Induction of Protective Immunity by Vaccination With Wild-Type Apo Superoxide Dismutase 1 in Mutant SOD1 Transgenic Mice

Article (PDF Available)inJournal of Neuropathology and Experimental Neurology 69(10):1044-56 · October 2010with30 Reads
DOI: 10.1097/NEN.0b013e3181f4a90a · Source: PubMed
Abstract
Vaccinations targeting extracellular superoxide dismutase 1 (SOD1) mutants are beneficial in mouse models of amyotrophic lateral sclerosis (ALS). Because of its misfolded nature, wild-type nonmetallated SOD1 protein (WT-apo) may have therapeutic application for vaccination of various SOD1 mutants. We compared the effects of WT-apo to those of a G93A SOD1 vaccine in low-copy G93A SOD1 transgenic mice. Both SOD1 vaccines induced antibody against G93A SOD1 and significantly delayed disease onset compared with saline/adjuvant controls. WT-apo SOD1 significantly extended the life span of vaccinated mice. The vaccines potentiated TH2 deviation in the spinal cord as determined by the ratio of interleukin-4 to interferon-γ (IFNγ) or tumor necrosis factor and induced C1q deposition around motor neurons. Transgenic mice had abundant microglial expression of signal transducers and activators of transcription 4, an activator of transcription of IFNγ, in the spinal cord implicating IFNγ in the pathogenesis. On the other hand, the sera from G93A SOD1-vaccinated mice showed higher IFNγ or tumor necrosis factor and yielded a lower IgG1/IgG2c ratio than the sera from WT-apo-vaccinated mice. These results indicate that the TH1/TH2 milieu is affected by specific vaccinations and that antigenicity might counteract beneficial effects by enhancing TH1 immunity. Thus, because of its lower TH1 induction, WT-apo may be a therapeutic option and have broader application in ALS associated with diverse SOD1 mutations.

Figures

Figure
Figure
Figure

Full-text (PDF)

Available from: Makoto Urushitani, Nov 20, 2014
Copyright @ 2010 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.
ORIGINAL ARTICLE
Induction of Protective Immunity by Vaccination With Wild-Type
Apo Superoxide Dismutase 1 in Mutant SOD1 Transgenic Mice
Shigeko Takeuchi, MS, Noriko Fujiwara, PhD, Akemi Ido, PhD, Miki Oono, MD, Yuki Takeuchi, MD,
Minako Tateno, PhD, Keiichiro Suzuki, MD, PhD, Ryosuke Takahashi, MD, PhD,
Ikuo Tooyama, MD, PhD, Naoyuki Taniguchi, MD, PhD, Jean-Pierre Julien, PhD,
and Makoto Urushitani, MD, PhD
Abstract
Vaccinations targeting extracellular superoxide dismutase 1
(SOD1) mutants are beneficial in mouse models of amyotrophic
lateral sclerosis (ALS). Because of its misfolded nature, wild-type
nonmetallated SOD1 protein (WT-apo) may have therapeutic appli-
cation for vaccination of various SOD1 mutants. We compared the
effects of WT-apo to those of a G93A SOD1 vaccine in low-copy
G93A SOD1 transgenic mice. Both SOD1 vaccines induced antibody
against G93A SOD1 and significantly delayed disease onset compared
with saline/adjuvant controls. WT-apo SOD1 significantly extended
the life span of vaccinated mice. The vaccines potentiated T
H
2devi-
ation in the spinal cord as determined by the ratio of interleukin-4 to
interferon-F(IFNF) or tumor necrosis factor and induced C1q depo-
sition around motor neurons. Transgenic mice had abundant microglial
expression of signal transducers and activators of transcription 4,
an activator of transcription of IFNF, in the spinal cord implicating
IFNFin the pathogenesis. On the other hand, the sera from G93A
SOD1-vaccinated mice showed higher IFNFor tumor necrosis factor
and yielded a lower IgG1/IgG2c ratio than the sera from WT-apoY
vaccinated mice. These results indicate that the T
H
1/T
H
2milieuis
affected by specific vaccinations and that antigenicity might counteract
beneficial effects by enhancing T
H
1 immunity. Thus, because of its
lower T
H
1 induction, WT-apo may be a therapeutic option and have
broader application in ALS associated with diverse SOD1 mutations.
Key Words: Acquired immunity, Amyotrophic lateral sclerosis,
Superoxide dismutase 1, Transgenic mice, Vaccination.
INTRODUCTION
Amyotrophic lateral sclerosis (ALS) is a lethal neuro-
degenerative disease characterized by progressive muscle
weakness and wasting. Although the precise pathogenetic
mechanisms of ALS remain elusive, diverse genetic mutations
have been identified in familial ALS cases and are risk factors
for both sporadic and familial ALS (1). Mutations in super-
oxide dismutase 1 (SOD1) account for 20% of familial ALS
and have been determined to be the cause of motor neuron
degeneration in many instances (2). Moreover, transgenic (Tg)
mice carrying the human SOD1 mutation represent an excellent
animal model of ALS (3). Importantly, the concept of nonYcell-
autonomous motor neuron death was derived from intensive
analyses of mutant SOD1 Tg mice, which has had a major
impact on understanding and treating not only mutant SOD1-
linked ALS but also other neurodegenerative diseases such
as Parkinson, Alzheimer, and Huntington diseases (4Y6). There
is also growing interest in the hypothesis of prion-like spread-
ing of disease-causing proteins in such diseases (7, 8).
On the basis of our findings that chromogranin A/B may
act as chaperone-like proteins to promote secretion of mutant
SOD1 (9), we targeted extracellular SOD1 of G37R SOD1 Tg
mice using a G93A mutant SOD1 vaccine, which resulted in
a significant delay in disease onset and extension of the life
span of the vaccinated mice (10). However, vaccination against
high-copy G93A SOD1 (G93AGur) mice was not effective.
We ascribed this failure to the extremely high expression level
of the transgene. It is also possible, however, that a combina-
tion of specific antigen and the host immune response can
counteract the beneficial effects of vaccination. For example, in
a report testing amyloid-A(AA) vaccines of wild-type (WT) or
various mutations linked to familial Alzheimer disease, dif-
ferent inflammatory responses and IgG subclasses were elic-
ited depending on the type of AAvaccine used; antibody titer
against AAin each vaccine was high overall (11). Another
approach by Kutzler et al (12) using DNA vaccines encoding
WT or Flemish/Dutch mutant AAshowed greater antigenicity
J Neuropathol Exp Neurol Volume 69, Number 10, October 20101044
J Neuropathol Exp Neurol
Copyright Ó2010 by the American Association of Neuropathologists, Inc.
Vol. 69, No. 10
October 2010
pp. 1044Y1056
From the Molecular Neuroscience Research Center (ST, AI, MO, IT, MU),
Shiga University of Medical Science, Shiga; Department of Biochem-
istry (NF, YT, KS), Hyogo College of Medicine, Hyogo; Department of
Peripheral Nervous System Research (MT), National Institute of Neuro-
science, National Center of Neurology and Psychiatry, Tokyo, Japan;
Department of Psychiatry and Neuroscience (J-PJ), Research Centre of
CHUQ, Laval University, Quebec, Canada; Department of Disease Gly-
comics (Seikagaku Corporation) (NT), The Institute of Scientific and
Industrial Research, Osaka University, Osaka; Systems Glycobiology
Group (NT), Disease Glycomics Team, Advanced Science Institute,
RIKEN, Saitama; and Department of Neurology (MO, RT), Kyoto Uni-
versity, Kyoto, Japan.
Send correspondence and reprint requests to: Makoto Urushitani, MD, PhD,
Unit for Neurobiology and Therapeutics, Molecular Neuroscience
Research Center, Shiga University of Medical Science; Shiga, Japan;
E-mail: uru@belle.shiga-med.ac.jp
This study was funded by the Japan Society for the Promotion of Science,
Japan Health and Labour Science Research Grants, Japan ALS Associa-
tion, and Takeda Science Foundation.
Supplemental digital content is available for this article. Direct URL citations
appear in the printed text and are provided in the HTML and PDF versions
of this article on the journal’s Web site (www.jneuropath.com).
Copyright @ 2010 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.
and T
H
1 immune response to the mutant than the WT AA
peptide. The T
H
1 cytokines tumor necrosis factor (TNF) and
interferon-F(IFNF) are implicated in motor neuron degener-
ation in ALS (13, 14), whereas the T
H
2 cytokine interleukin-4
(IL-4) provides protective immunity to prevent motor neuron
death (15). Moreover, T cells are present in the spinal cord of
ALS patients (16), and circulating CD4
+
T cells affect the dis-
ease course of ALS model mice (17Y19). In particular, CD4
+
T
cells stimulate astrocytes or microglia to express neuropro-
tective molecules including glutamate transporter or insulin-
like growth factor 1 (IGF-1) (17). The effects of different types
of acquired immunity induced by different vaccines on the
therapeutic outcome, however, have not been systematically
studied.
More than 120 mutations in SOD1 covering overall
domains have been reported. Therefore, it would be desirable
to develop a vaccine that is effective and not dependent on
thespecific SOD1 mutation associated with the disorder. One
approach is to target the core domain of the molecule, i.e.
the dimer interface (20). Another approach is to use apo WT
SOD1, the molecular behavior of which is similar to that of
the mutant molecules (21, 22). We previously reported that
wild-type (WT) SOD1 with posttranslational modifications,
including oxidative modification, gains properties similar to
those of mutant SOD1 (10). Other reports show possible
involvement of WT SOD1 in the pathogenesis of sporadic
ALS. For example WT SOD1 has been detected in cytoplas-
mic aggregates and abnormal dimer formations have been
reported in the spinal cord of sporadic ALS patients (23, 24).
Therefore, the development of SOD1 vaccine based on the
WT sequence deserves investigation because of its potential
broad application not only in mutant SOD1-associated but
also in sporadic ALS.
Here, we compared the effects of G93A-apo SOD1 and
WT-apo SOD1 vaccines on the survival and the life span of
low-copy G93A SOD1 Tg mice (G93AGur
dl
) and analyzed
the relationships to the cellular and humoral immune re-
sponses elicited by the vaccinations.
MATERIALS AND METHODS
Materials
All chemicals were obtained from Nacalai Tesque, Inc
(Kyoto, Japan), unless specified otherwise, and were of the
highest grade available.
Purification of Recombinant SOD1 From
Escherichia coli
Recombinant human WT and gly93ala (G93A) mutant
SOD1 were produced in Escherichia coli according to a
previous report (22). The eluates were dialyzed against
endotoxin-free saline and subsequently stored at j80-Cuntil
use. Metallation of the recombinant SOD1 was performed as
previously described (22). The metallated and nonmetallated
SOD1s were designated as holo-SOD1s and apo-SOD1s,
respectively. Control monomeric SOD1 was prepared by
treating recombinant human SOD1 chemically modified with
2-mercaptoethanol at Cys111 (2-ME-SOD1 (25). The purity
of the recombinant protein was assessed by sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)
and gel staining with Coomassie brilliant blue. Details
are provided in Methods, Supplemental Digital Content 1,
http://links.lww.com/NEN/A178.
Molecular-Size Filtration Chromatography
A total of 100 KL of SOD1 proteins (2Y3 mg/mL) in
phosphate-buffered saline (PBS) (j) were applied to a
molecular-size filtration high-performance liquid chromatog-
raphy (AKTA Explorer 10S) at a flow rate of 0.5 mL/min on
Superdex 75 10/30 (GE Healthcare, Piscataway, NJ) equili-
brated with 50 mmol/L sodium phosphate buffer containing
0.15 mol/L NaCl, pH 7.4. The calibration of the column for
the estimation of molecular weight was performed using
bovine serum albumin (Intergen, Milford, MA), ovalbumin
(GE Healthcare), and E. coli thioredoxin (Promega, Madison,
WI), as protein standards.
Animal Experiments
Tg mice harboring human G93A SOD1 (B6SJL-
TgN[SOD1-G93A]1Gur, hSOD1G93A; Jackson Laboratory,
Bar Harbor, ME) were backcrossed with C57BL/6 strain for
more than 20 generations (G93AGur
dl
). G93AGur
dl
mice
were vaccinated with apo recombinant human SOD1 (WT
or G93A) or human erythrocyte-derived SOD1 (Sigma) with
Monophosphoryl Lipid A- Trehalose Dicorynomycolate
(MPL-TDM; Ribi) adjuvant, as previously reported (10).
From day 100, motor performance was evaluated for onset
of decreased rotarod retention time (Muromachi, Tokyo,
Japan) and body weight (BW) change to determine the
time of onset (Methods, Supplemental Digital Content 1,
http://links.lww.com/NEN/A178). The data for the survival
and onset were analyzed by Kaplan-Meyer curve and log-rank
test using Prism software (GraphPad, La Jolla, CA).
FIGURE 1. Characterization of recombinant human G93A or
wild-type (WT)-apo superoxide dismutase 1 (SOD1) proteins.
The chromatogram profile of SOD1 proteins separated with
a gel filtration column. The native WT-SOD1, apo-WT, apo-
G93A, holo-WT, holo-G93A, and monomeric SOD1, were eluted
on Superdex 75 column. The elution profiles were monitored
by the absorbance change at 280 nm. apo- indicates non-
metallated; holo-, metallated.
J Neuropathol Exp Neurol Volume 69, Number 10, October 2010 Vaccination in ALS Model Mice
Ó2010 American Association of Neuropathologists, Inc. 1045
Copyright @ 2010 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.
FIGURE 2. Effects of vaccination with recombinant human G93A-apo or wild-type (WT)-apo superoxide dismutase 1 (SOD1)
proteins on clinical disease and antibody (Ab) responses in low-copy G93A SOD1 Tg mice (G93AGur
dl
). (A, B) Both the G93A-apo
and the WT-apo vaccines significantly delayed the disease onset of G93AGur
dl
mice. Time of the onset was determined based on
body weight (BW) loss (A) and impaired rotarod performance (B).(C) The WT-apo SOD1 vaccine significantly extended the life
span of G93AGur
dl
mice; the G93A-apo SOD1 vaccine showed a nonsignificant trend. p G0.05 by Kaplan-Meyer curve and log-
rank test; n = 13 for saline/adjuvant, n = 12 for the WT-apo, n = 9 for G93A vaccinations. (D) Ab titers against G93A SOD1 protein.
Sera from vaccinated mice (day 120 and end point) were analyzed by ELISA. The WT-apo SOD1 vaccine induced Ab against
G93A SOD1 as well as the G93A-apo SOD1 did even at the end point. *pG0.01 versus saline-injected mice at day 120;
#
pG0.01
versus saline-injected mice at end point by 1-way analysis of variance with Bonferroni post hoc test. (E, F) Correlation analysis
between Ab titer and clinical score including rotarod performance (E) and life span (F). Antibodies in both the G93A-apo and
the WT-apo vaccination were positively correlated with the onset timing of paralysis (E). The WT-apo SOD1 vaccination was
positively correlated with longevity (p = 0.00441, Spearman r= 0.6768); the G93A-apo SOD1 vaccine displayed a lower value for
Spearman r(0.0887).
Takeuchi et al J Neuropathol Exp Neurol Volume 69, Number 10, October 2010
Ó2010 American Association of Neuropathologists, Inc.
1046
Copyright @ 2010 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.
FIGURE 3. Induction of T
H
2 deviation in the spinal cord by superoxide dismutase 1 (SOD1) vaccination. (A, a) Real-time poly-
merase chain reaction analysis of mRNA of interferon-F(IFNF), tumor necrosis factor (TNF), interleukin-4 (IL-4), and transforming
growth factor A(TGFb) in spinal cord tissues from nontransgenic (Tg) wild-type (WT) and nontreated Tg mice at the early pre-
symptomatic stage (day 120). The mean mRNA of each cytokine standardized by actin mRNA is shown. (A, b) The ratio of IL-4 to
IFNFor to TNF in each mouse was averaged in each group to estimate the T
H
2/T
H
1 milieu. The mean IL-4/IFNFor IL-4/TNF ratio
shows a decrease in presymptomatic Tg mice (mean TSEM, n = 3 for non-Tg mice, and n = 4 for Tg mice). (B, a) Real-time
polymerase chain reaction analysis of the same cytokines in the spinal cord of the vaccinated mice with WT or G93A SOD1 proteins
or saline plus Ribi adjuvant and nontreated Tg controls. (B, b) The ratio of IL-4/IFNFor IL-4/TNF showing that the SOD1 efficiently
induced T
H
2 protective immunity by inhibiting IFNFand provoking IL-4. *pG0.05 versus nontreated Tg controls by 1-way analysis
of variance with Bonferroni post hoc test (mean TSEM, n = 3Y4 mice).
TABLE. Effect of Vaccinations on Days of Disease Onset and Survival of G93A Mice
Vaccine Saline WT-Apo G93A-Apo Saline WT-Apo G93A-Apo
Median survival 274 288 287.5 Mean survival (d) 272 T3.9 291 T7.9 283 T4.9
Median onset of rotarod impairment 225 243.5 249 Mean onset of Rotarod impairment 228 T5.4 239 T7.5 249 T6.2
Median onset BW loss 212 234 228 Mean onset BW loss 212 T6.2 247 T10.6 233 T5.9
Effect on survival V14 13.5 Effect on survival V19 11
Effect on rotarod impairment onset V18.5 24 Effect on rotarod impairment onset V11 21
Effect on onset BW loss V22 16 Effect for onset BW V35 21
The effect of vaccination with recombinant G93A-apo or the WT-apo SOD1 protein on the delay of the disease onset or extension of the life span of low-copyG93A
SOD1 (G93AGur
dl
) mice. The left columns are median survival or median onset determined by Kaplan-Myer curves. Data in the right columns show the mean days from each
vaccination TSEM (n = 13 for saline, n = 12 for WT, and n = 9 for G93A).
Effect on survival or onset of impaired rotarod performance and body weight (BW) loss were obtained by subtracting the date for saline injection from those for the WT-apo or
the G93A-apo SOD1 vaccination. Time of disease onset was determined by BW loss and rotarod performance. All data are days.
J Neuropathol Exp Neurol Volume 69, Number 10, October 2010 Vaccination in ALS Model Mice
Ó2010 American Association of Neuropathologists, Inc. 1047
Copyright @ 2010 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.
Takeuchi et al J Neuropathol Exp Neurol Volume 69, Number 10, October 2010
Ó2010 American Association of Neuropathologists, Inc.
1048
Copyright @ 2010 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.
Enzyme-Linked Immunosorbent Assay
At the ages of 120 days, 210 days, or at the end point,
sera were obtained for antibody titration for total IgG, IgG1,
IgG2b, and IgG2c against G93A SOD1 protein by enzyme-
linked immunosorbent assay (ELISA) as previously described
(10), with minor modifications (Methods, Supplemental Dig-
ital Content 1, http://links.lww.com/NEN/A178).
Immunofluorescent Analysis
The spinal cords at the level of L4-L5 were resected
from 275-day-old mice after perfusion with 4% parafor-
maldehyde. The sections were incubated in the same fix-
ative for 4 hours and subsequently in PBS containing 20%
sucrose and 0.1% sodium azide at 4-C. Cryosections 24 Km
thick were incubated with a mixture of primary antibodies
and subsequently with a mixture of corresponding second-
ary antibodies labeled with Alexa 488 or 594 (Invitrogen,
Carlsbad, CA). For counterstaining, 4,6-diamidino-2-
phenylindole dihydrochloride (Nacalai Tesque) was used.
The number of remaining motor neurons in the spinal cord
was obtained from averaged count of large NeuN+ cells
(G20 Km of the soma) in the anterior horn from 3 samples
per mouse. Fluorescent images were obtained using a
fluorescent microscope (Keyence, Mississauga, Canada) or
a confocal laser microscope equipped with the software
EZ-C1 (Nikon, Tokyo, Japan). Antibody information
is provided in Methods, Supplemental Digital Content 1,
http://links.lww.com/NEN/A178.
Serum Cytokine Quantification
Serum concentrations of TNF, IFNF, and IL-4 were de-
termined using suspension array system (Bio-Plex Pro Mouse
Cytokine; BioRad, Hercules, CA) according to the manu-
facturer’s protocol (Methods, Supplemental Digital Content 1,
http://links.lww.com/NEN/A178).
Cytokine Profiling in the Spleen and the Spinal
Cord by Real-Time Polymerase Chain Reaction
Cytokines and transcription factors including IFNF,
TNF, transforming growth factor A1 (TGFA), IL-4, forkhead
box P3, and RAR-related orphan receptor Fwere investigated
for their messenger RNA (mRNA) expression level by real-
time polymerase chain reaction (PCR) system, as described
with minor modifications (26). On days 120 and 210, total
RNA was purified using TRIzol and RNA purification kit
(Invitrogen), and an equal amount of total RNA was reacted
with reverse transcriptase (Superscript III; Invitrogen) to
generate complementary DNA (cDNA). mRNA of cytokines
or marker molecules for CD4
+
T lymphocytes was analyzed
by real-time PCR using the SYBR green system (Roche,
Basel, Switzerland). The cDNA level of the target molecules
was normalized to that of actin. Detailed procedures including
primer sequences are in Methods, Supplemental Digital
Content 1, http://links.lww.com/NEN/A178.
Statistics
The survival and clinical onset data were analyzed by
Kaplan-Meyer curve and log-rank tests. The effect of a
single factor on the difference among 3 or more groups was
determined by 1-way analysis of variance (ANOVA) with
Bonferroni post hoc test. Comparisons between 2 groups
were analyzed by unpaired ttest. p G0.05 was judged to be
significant.
RESULTS
Characterization of Recombinant SOD1s
The purities of recombinant SOD1s were more than
95%, as determined by SDS-PAGE (data not shown). To
investigate the precise molecular weight of SOD1 proteins of
as-isolated and nonmetallated state (i.e. WT-apo or G93A-
apo) versus the holo-state, we performed molecular-size fil-
tration chromatography because it is thought that apo-SOD1s
are constantly monomers (27). However, both apo- and holo-
SOD1s are dimeric, although the WT-apo sample appeared to
contain some monomer form (Fig. 1). Moreover, the elution
times of WT-apo and G93-apoA were slightly faster than
those of the holo-types, which could reflect looser structures
of apo-SOD1s.
Vaccination With Recombinant WT-Apo SOD1
Extends the Life Expectancy and Delays
Disease Onset
G93AGur
dl
mice were immunized with recombinant
G93A-apo SOD1 or WT-apo SOD1 using Ribi adjuvant.
Recombinant WT-apo SOD1 was used because of its broader
applications regardless of mutation types in SOD1 gene
and because, like mutant SOD1, it is misfolded (22). The
G93A-apo vaccine significantly delayed the onset of paralysis
assessed by BW change and rotarod performance tests
(Figs. 2A, B). The WT-apo SOD1 vaccination also sig-
nificantly suppressed BW loss and showed a trend to pre-
serving the rotarod performance, although this was not
significant. The difference of onset time was 18 or 22 days in
the WT-apo- and 24 or 16 days in the G93A-apo vaccinated
mice, as assessed rotarod tests or BW change, respectively
(Table). In addition, both vaccines prolonged the life spans
compared with saline/adjuvantYinjected controls by 14 and
13.5 days for WT-apo and G93A-apo vaccine, respectively.
The log-rank test determined statistical significance only for
the WT-apo SOD1 vaccinations (p = 0.0183 for the WT-apo;
p = 0.0720 for the G93A-apo). Notably, the mean life span
FIGURE 4. Signal transducers and activators of transcription 4 (STAT4), a transcription factor for interferon-F(IFNF), is abundantly
expressed in active microglia of G93A transgenic (Tg) mice. (A) Double immunostaining for STAT4 (red) and NeuN (green) in the
spinal cord of vaccinated 275-day-old mice. The left 2 columns (a, b, e, f, i, j, m, n) and the right 2 columns (c, d, g, h, k, l, o, p)
are images obtained from the fluorescent microscope and confocal laser microscope, respectively. Expression of STAT4 does not
merge with NeuN. Scale bar = 50 Km. (B, C) Confocal micrographs of double immunofluorescent analysis for STAT and Mac2 or
phosphorylated STAT4 and Mac2 in the spinal cord of G93AGur
dl
mice. (aYc) Saline/adjuvant-injected G93AGur
dl
mice (30 weeks).
(d) Non-Tg littermate. Scale bar = 30 Km.
J Neuropathol Exp Neurol Volume 69, Number 10, October 2010 Vaccination in ALS Model Mice
Ó2010 American Association of Neuropathologists, Inc. 1049
Copyright @ 2010 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.
Takeuchi et al J Neuropathol Exp Neurol Volume 69, Number 10, October 2010
Ó2010 American Association of Neuropathologists, Inc.
1050
Copyright @ 2010 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.
indicated that there were shorter durations of suffering in
G93A-apoYvaccinated mice (Fig. 2C; Table).
We next investigated whether the antibody titer corre-
lated with the therapeutic benefit. The WT-apo and the G93A
vaccination induced high titers of anti-G93A SOD1 antibody
compared with the native SOD1 vaccination or the saline
control (Fig. 2D). The WT-apo vaccine showed a significant
positive correlation between the life span and antibody titer
(r= 0.5758, p = 0.0041 by Spearman r) and a trend toward
a positive correlation of rotarod testing (Fig. 2E; r= 0.539,
p = 0.0569 by Spearman r). The G93A-apo vaccine showed
no significant correlation with either the onset timing (rotarod
and BW loss) or survival (Figs. 2E, F). The disease duration
was not associated with antibody titer in either the WT-apo
or G93A-apo vaccine groups (not shown).
SOD1 Vaccination Potentiates Protective
Immunity in the Spinal Cord
mRNA levels of IFNF, TNF, and IL-4 were determined
in the spinal cords of presymptomatic nontreated G93A
SOD1 Tg mice (120 days old; n = 3) and non-Tg WT litter-
mates (n = 5). There were slightly higher levels of IFNFand
TNF and lower levels of IL-4 in G93A Tg mice versus non-Tg
age-matched mice (Fig. 3Aa). The ratios of IL-4 to IFNF
(IL-4/IFNF) or TNF (IL-4/TNF), (termed BT
H
2/T
H
1 ratios[)
were much lower in the Tg mice (Fig. 3Ab). Comparison
between SOD1- or saline plus adjuvantYvaccinated groups
and nontreated Tg mice showed a significant reduction of
IFNFin both SOD1 vaccinations versus nontreated G93A Tg
controls. Interleukin-4 mRNA was markedly elevated by the
SOD1 vaccinations on both days 120 and 210, but only the
G93A vaccination on day 120 was statistically significant (by
1-way ANOVA, Bonferroni post hoc test). The saline/adjuvant
injection provided an effect comparable to that of the SOD1
vaccination in reducing IFNFand upregulating IL-4 (Fig. 3Ba).
The T
H
1/T
H
2 milieu is critically influenced by the adjuvant
as well as by antigens, and the MPL/TDM adjuvant used has
been reported to favor a T
H
2 rather than T
H
1 milieu (28).
Signal Transducers and Activators of
Transcription 4 Induction in Active Microglia in
the Spinal Cord
The role of IFNFin ALS pathogenesis is controversial
(14, 29Y31). Therefore, we examined the expression pattern
of IFNFin spinal cords of mutant SOD1 Tg mice. Immuno-
histochemical analysis was performed with an antibody against
signal transducers and activators of transcription 4 (STAT4),
an activator for transcription of IFNFunder the stimulation
of IL-12. There was STAT4 immunoreactivity in the spinal
cord of mutant SOD1 Tg mice with or without vaccination
(Fig. 4A). Notably, the lamina I in the dorsal horn, which
receives nociceptive input, was markedly STAT4+ in both Tg
and non-Tg mice (Fig. 4A, a, b, e, f, i, j, m, n). These cells
were also NeuN+ (not shown), which might imply a role for
IFNFin pain generation to nociceptive stimulation (32). How-
ever, the STAT4+ cells in other areas, including the anterior
horn are not costained with anti-NeuN antibody (Fig. 4A, c, d,
g, h, k, l, o, p). Therefore, we performed double immuno-
fluorescent staining using antibodies against STAT4 and
Mac2. STAT4 was exclusively expressed in Mac2+ active
microglia (Fig. 4B). Because phosphorylation is required to
activate STAT4 for IFNFtranscription, we examined the
presence of phosphorylated STAT4 (pSTAT4) by double
immunofluorescent study and found that certain populations
of active microglia were pSTAT4+ (Fig. 4C). There was no
difference in STAT4 or pSTAT4 levels between vaccinated
or nonvaccinated mice (not shown), but the presence of STAT4
in microglia provides evidence that spinal motor neurons are
affected by IFNFfrom activated microglia. Notably, the fluo-
rescent antimouse IgG antibodies showed meningeal enhance-
ment only in SOD1 vaccinated mice (Fig. 4A, arrowheads),
suggesting that vaccine-generated antibodies can reach the
spinal cord.
Vaccination With G93A-Apo or WT-Apo SOD1
Attenuates Motor Neuron Loss and Induces
Complement C1q Deposition
We next investigated whether the beneficial effect of
the vaccination with the G93A-apo or the WT-apo SOD1 cor-
related with motor neuron protection by counting of NeuN+
remaining anterior horn cells in 275-day-old Tg mice. Both
vaccinations with the G93A-apo and the WT-apo significantly
attenuated motor neuron loss (Fig. 5A), whereas there was no
significant difference in the number of active microglia
cells stained by antiYMac-2 between saline and SOD1 vacci-
nations was observed (Figure, Supplemental Digital Content 2,
http://links.lww.com/NEN/A179).
We then studied the activation of the classic pathway
of complement system after vaccination in mice of the same
age. C1q, which binds to antigen-antibody complexes at the
immunoglobulin Fc domain, is upregulated around motor
neurons in SOD1 mutant models before disease onset, sug-
gesting that Bdanger molecules[are leaked from degenerating
motor neurons (33). Double immunofluorescent staining with
rat monoclonal anti-C1q and mouse monoclonal anti-NeuN
detected C1q around motor neurons in both SOD1-vaccinated
and saline-control G93AGur
dl
mice, but not in age-matched
non-Tg littermates. More prominent C1q deposition was
detected in SOD1 vaccinated mice than in saline/adjuvant
controls (Fig. 5B). These data indicate that the apo-SOD1
vaccination activates the classic pathway in the spinal cord.
Because chromogranin interacts with mutant SOD1, promotes
FIGURE 5. Superoxide dismutase 1 (SOD1) vaccination attenuates motor neuron (MN) loss and induces complement deposition
around MNs. (A) MN count in the spinal cord of vaccinated 275-day-old mice. NeuN+ large anterior horn neurons (920 Kmin
size) were counted in 3 slices per mouse; mean numbers per slice are shown. *pG0.05 by 1-way analysis of variance of Bonferroni
test (n = 9 for transgenic [Tg], n = 4 for non-Tg; WT = wild-type. (B) Induction of complement C1q deposition around MNs by
SOD1 vaccination. Confocal micrographs of the spinal cord slices of the vaccinated mice (30 weeks) stained with antibodies
against NeuN (red, mouse monoclonal) and C1q (green, rat monoclonal). C1q staining surrounding MNs are indicated with
arrowheads; d, h, l, p show negative controls in which primary antibodies were eliminated. Scale bar = 30 Km.
J Neuropathol Exp Neurol Volume 69, Number 10, October 2010 Vaccination in ALS Model Mice
Ó2010 American Association of Neuropathologists, Inc. 1051
Copyright @ 2010 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.
its secretion, and activates microglia (9), the complex for-
mation comprising mutant SOD1Yanti-SOD1 antibody-C1q
may be facilitated by the SOD1 vaccination.
WT-Apo SOD1 Vaccination Induces a Higher
T
H
2/T
H
1 Milieu Than G93A Vaccination
There were differences in the slowing of the progression
in Tg mice between the WT-apo and the G93A SOD1 vac-
cinations. Because the antibody titer in the G93A-apo vacci-
nation unexpectedly showed no correlation with disease onset
or life span as opposed to the WT-apo (Figs. 2E, F), we
compared the IgG subclasses induced by the WT-apo and
G93A-apo vaccines. Sera of SOD1-vaccinated mice were
analyzed on days 120 and 210 by ELISA for IgG1, 2b, and 2c
titers. Both the WT-apo and the G93A-apo vaccines induced
considerable amounts of each subclass and the G93A vacci-
nation induced more IgG than the WT vaccination. Overall,
the difference between the 2 vaccines was statistically sig-
nificant only on day 120 for IgG1 and day 210 for IgG2b and
IgG2c (Fig. 6A, p G0.05 by unpaired ttest). The mean ratios
of IgG1 to IgG2c (IgG1/IgG2c) (an indicator of the T
H
2/T
H
1
milieu) were not significantly different between WT and
G93A vaccinations on day 120, but on day 210, this ratio was
significantly higher in the WT-apo vaccination than in the
G93A (Fig. 6Ab, p G0.05 by 1-way ANOVA).
We further measured the quantity of serum cytokines
induced by these SOD1 vaccinations using a suspension array
system. The antisera from immunized mice with WT or G93A
SOD1 vaccine (n = 3/group) on day 120 was analyzed for
TNF, IFNF, and IL-4. The G93A-apo vaccination induced
more of the T
H
1 cytokines TNF and IFNFcompared with
the WT-apo vaccination (Fig. 6B). Interleukin-4 levels were
below the detection limit (0.59 pg/mL) in either non-Tg or Tg
mice regardless of the type of the vaccination. The analysis
of the spleen tissues of the same mice for mRNA quantity
of TNF, IFNF, IL-4, and TGFAby real-time PCR revealed a
clear trend in which IL-4 mRNA in the WT or the G93A vac-
cination was higher than that in the saline control on day 120.
The G93A vaccination tended to induce higher IL-4 than the WT
did on day 120. On day 210, IL-4 mRNA remained high in the
WT but not in the G93A vaccination sera (Figure, part A, Sup-
plemental Digital Content 2, http://links.lww.com/NEN/A179).
No significant differences in expression levels of forkhead
box P3 or RAR-related orphan receptor Ft, a marker for reg-
ulatory T-cell or a T
H
1 7 cells, respectively, between control
and vaccinated groups were observed (data not shown).
We then analyzed the relationship between IgG sub-
class and the therapeutic effect of the vaccinations. In the
WT-apo vaccination there was a positive correlation between
IgG2b titer and the date of the onset (Figure, part B, Supple-
mental Digital Content 3, http://links.lww.com/NEN/A180, b;
FIGURE 6. Analysis of sera for acquired immunity by super-
oxide dismutase 1 (SOD1) vaccination. (A, a) ELISA for IgG
subclasses induced by the SOD1 vaccination. The titer for
IgG1, 2b, and 2c against the G93A-apo SOD1 was obtained
from sera of wild-type (WT) and G93A SOD1-vaccinated mice
at day 120 and 210 and expressed as optical density (OD)
405 nm (mean TSEM; each group, n = 4 or 5). *pG0.05, NS
indicates not significant by unpaired ttest. (A, b) The ratio
of IgG1 to IgG2c (IgG1/IgG2c) was obtained in each mouse
and the data averaged (mean TSEM for each group, n = 4
or 5). *pG0.05 by 1-way analysis of variance with Bonferroni
post hoc test. (B) Serum cytokine concentrations. Sera from
WT or G93A SOD1 vaccination (n = 3) were analyzed for the
quantification of interferon gamma (IFNF) and tumor necrosis
factor (TNF) using a suspension array system. Each bar shows
the mean concentration from 3 mice TSEM. *pG0.05 by
unpaired ttest. ND indicates not detected (under the lowest
value of standard curve (0.17 pg/mL).
Takeuchi et al J Neuropathol Exp Neurol Volume 69, Number 10, October 2010
Ó2010 American Association of Neuropathologists, Inc.
1052
Copyright @ 2010 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.
r= 0.7333, p = 0.0156, by Spearman r), whereas IgG1 and
IgG2c showed no significant correlation (Figure, part B, a, c,
Supplemental Digital Content 3, http://links.lww.com/NEN/
A180). By contrast, there was no positive correlation of the
G93A vaccination with any IgG subtype. Rather, the IgG1 titer
was negatively correlated to the survival effect (Figure, part B,
d, Supplemental Digital Content 3, http://links.lww.com/NEN/
A180; r=j0.5879, p = 0.0403, by Spearman r).
Antibody Induction in Non-Tg Mice by
Vaccination With Murine WT SOD1
Because the human WT SOD1 is a foreign antigen to
mice, antibody is generally easily induced. For human trials,
induction of the antibody against human WT-SOD1 in human
body would be required. Therefore, we investigated whether
the murine WT-apo SOD1 induced antibody by breaking
immunologic tolerance to endogenous SOD1 in mice. Vac-
cination of non-Tg C57BL/6 mice with recombinant mouse
WT-apo SOD1 induced antibody against mouse SOD1 at
days 120 and 210 (Fig. 7A). Simultaneously, the adverse
reactions of the vaccination were monitored for BW loss
(Fig. 7B), movement, hair length, spontaneous activity, and
provoked biting as well as observation of tissue hemorrhage
or anemia at autopsy. These analyses displayed no abnor-
malities that could be attributed to vaccination. The analysis
of IgG subclass profile showed that all the subclasses in-
cluding IgG1, IgG2b, and IgG 2c were comparably elevated
(Fig. 7C), indicating that both T
H
1 and T
H
2 immunity is in-
duced in non-Tg mice.
DISCUSSION
In this report, we demonstrate a beneficial effect of
vaccination with the WT-apo SOD1 protein to prolong the life
FIGURE 7. Vaccination of nontransgenic (non-Tg) mice using recombinant wild-type (WT) superoxide dismutase 1 (SOD1) protein
effectively induces antibody with no detectable detrimental effect. (A) ELISA of sera of non-Tg mice vaccinated with mouse SOD1
and saline with adjuvant at days 120 and 210 (n = 5 in each group). Data show mean TSEM (n = 5). *pG0.01 by 1-way analysis
of variance of Bonferroni test. (B) Comparison of body weights after vaccination between mouse SOD1-vaccinated and saline/
adjuvantYinjected controls. Each point shows mean body weight (g) of 5 mice per group. (C) ELISA for IgG subclasses induced
by mouse SOD1 vaccination. The titer for IgG1, 2b, and 2c against recombinant mouse SOD1 was determined in sera from
SOD1-vaccinated and control saline/adjuvantYinjected mice at day 210 and expressed as OD 405 nm (mean TSE forSE for each
group, n = 5). *pG0.01 by 1-way analysis of variance of Bonferroni test.
J Neuropathol Exp Neurol Volume 69, Number 10, October 2010 Vaccination in ALS Model Mice
Ó2010 American Association of Neuropathologists, Inc. 1053
Copyright @ 2010 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.
span of G93A SOD1 Tg mice. The chief effect was delaying
the disease onset rather than slowing its progression. The
G93A SOD1 vaccine also delayed the onset significantly and
showed a trend to prolonging the life span. This result is
in agreement with our previous finding that the vaccination
with the G93A-apo SOD1 protein against G37R SOD1 mice
(5 times higher) led to a 3-week delay of onset and a 4-week
prolongation of life span, whereas the same vaccination of
high-copy G93A SOD1 (G93AGur) mice (17 times higher)
was not effective (10). Therefore, the expression level of the
mutant protein in the Tg mice may determine the vaccine
effect, particularly in the late stage. Because there generally
are 1 or 2 copies of the mutant SOD1 gene in ALS patients,
our results indicate that further development of vaccination
aiming not only for prevention but also for disease inhibition
after onset is warranted.
Unexpectedly, the apo-form of the SOD1 was a dimer,
although the precise molecular features of the WT-apo SOD1
are unclear. The apo-SOD1 of both WT and G93A are slightly
larger molecules than the holo-SOD1s, indicating that the
apo-form of SOD1s form nonnative dimers. Intriguingly, the
previous report showed that a nonnative SOD1 dimer is a
common finding in mutant and sporadic ALS (24). Therefore,
the misfolded dimeric conformation might mediate the
vaccine effect of the WT-apo SOD1. Because of its utility
regardless of the mutation type, the protective effect of the
WT-apo SOD1 in G93AGur
dl
mice may represent a widely
applicable paradigm. We observed no detrimental effect of
WT-apo SOD1 vaccination in G37R Tg mice, but there
seemed to be some beneficial effect (personal observation).
We and others have reported potential misfolding property
of the WT SOD1 when posttranslationally modified such as
monomerization (21) or oxidation (34), which may account
for the effect of the WT-apo SOD1 vaccination. Therefore,
further modification by narrowing the critical domains for
misfolding or by specifying more aberrant structures such
as oligomer or aggregates is a potential strategy for develop-
ing vaccines against mutant SOD1-linked ALS.
The precise mechanisms of the beneficial effects of
vaccination with the misfolded SOD1 protein are unclear. We
found that C1q increased around the remaining motor neurons
and that IgG immunoreactivity increased in the spinal cord
meninges. Moreover, the blood-brain barrier of mutant SOD1
Tg mice is fragile (35), and recent studies show CD4
+
T-cell
proliferation in the spinal cord of mutant SOD1 Tg mice
(17, 18). Therefore, it is conceivable that immune-related
molecules or cells may easily penetrate into the CNS in
mutant SOD1-linked ALS patients and that vaccine-induced
antibodies could directly target extracellular SOD1 in the
spinal cord. Although immunohistochemistry showed no
obvious change in the SOD1 aggregates by the vaccination
(not shown), this indicates that only small fractions of solu-
ble misfolded species might participate in the extracellular
toxicity.
Our results indicate that there is induction of protective
immunity by vaccination with the Ribi adjuvant. The effect of
both SOD1 vaccinations on T
H
2 deviation in the spinal cord
is striking. Moreover, STAT4 analysis suggests that activated
microglia in the spinal cord of mutant SOD1 Tg mice are the
source of IFNF. Indeed, a recent report also showed that
STAT4 is chiefly expressed in the active microglia in the
cerebral white matter in multiple sclerosis (36). Active mi-
croglia expressing Mac-2 in the spinal cords are detected in
presymptomatic mutant SOD1 Tg mice but not in non-Tg or
WT SOD1 Tg mice (personal observation). Thus, microglia-
derived IFNFmight be a pathogenic proinflammatory factor
in mutant SOD1-linked ALS. Indeed, accumulating evidence
also indicates the detrimental effect of IFNFin the CNS,
including oxidative stress (14), mitochondrial damage (30),
and MHC class II+ cell proliferation (18). On the other hand,
it should be noted that the toxicity of IFNFis context-
dependent and that it is regulated by a complex crosstalk
between microglia and T-cell subtypes (37). It was reported
that the stimulation of microglial cells with IFNFrescues cells
with various toxic challenges through the induction of IGF-1
(38) or IGF-2 (39). A recent study also shows that a low dose
of IFNFprovides neuroprotection by enhancing neurogenesis
in the amyloid precursor protein Tg mice (40). Therefore, the
effects of IFNFin the CNS are complex, and synergistic ef-
fects of other proinflammatory cytokines (e.g. TNF) in spe-
cific conditions likely are involved. Interestingly, the saline/
adjuvant injection also promoted T
H
2 deviation, possibly as
a result of using the Ribi adjuvant (29). This result indicates
that the adjuvant is crucial to acquire desirable effect of
the vaccination, especially in ALS mutant mice. However,
because of the lack of a therapeutic effect of saline/adjuvant
injection, T
H
2 deviation is not the primary mediator of the
beneficial effects of vaccination.
We previously reported that late-stage G37R Tg mice
with SOD1 vaccination had greater Mac2+ microglia than
control mice (10). However, in the present study, analysis of
the mice at the same point showed no significant difference
in the number of Mac2+ microglia. Therefore, it may be that
longer-lived mice with SOD1 vaccination showed more
abundant microgliosis. Further analyses including in situ hy-
bridization are needed to clarify this issue because of the
diverse and complicated functions of microglia.
We show here that the both SOD1 vaccinations in-
creased IL-4 mRNA levels in the spleen at an early time point
and that there was a significant difference in IgG2c titer
and IgG1/IgG2c ratio between the WT-apo and the G93A-apo
vaccinations at the late stage. These results suggest that
peripheral IL-4 may drive a T
H
2 milieu from the early pre-
symptomatic stage in the spinal cord, and circulating IFNF
may have some detrimental effect in the later period. Thus,
the amount of IFNFor TNF in the serum at the preclinical
stage might be a useful marker for selecting an appropriate
vaccine and adjuvant. Furthermore, the titer of each IgG
subclass was not positively correlated in G93A-apo vacci-
nation; rather, IgG1 titer was negatively correlated to the
life span (Figure, Supplemental Digital Content, part B, d,
http://links.lww.com/NEN/A180). These results suggest that
T
H
1/T
H
2 regulation is unbalanced by the G93A vaccination
in the Tg mice with the same mutation. Taken together, T
H
1
deviation in the G93A vaccination in the later stage might
counteract the effect to slow the progression. A recent report
shows that IFNFis increased in the serum of ALS patients
(41). Moreover, systemically injected IFNFdistributes in the
Takeuchi et al J Neuropathol Exp Neurol Volume 69, Number 10, October 2010
Ó2010 American Association of Neuropathologists, Inc.
1054
Copyright @ 2010 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.
brain (42) and it is known to disturb the immunologic priv-
ilege by the induction of MHC class I antigen in CNS cells
(43). On the other hand, ablation of CD4
+
cells in mutant
SOD1 Tg mice accelerated the ALS phenotype, which is
mediated by microglial inhibition of IGF-1 (17). It was also
reported that intravenous injection of regulatory or effecter
T-cells significantly prolongs the longevity of G93A Tg mice
(19). In a spinal cord injury model, externally injected mac-
rophages potentiated the repair of damaged tissue by halting
microglial activation (44). In such cases, anti-inflammatory
cytokines including IL-4 (17) and IL-10 (45) mediate transfer
of protective signals to microglia. Because there is potentially
aberrant immunity in ALS, vaccinations should be cautiously
designed. Considering recent conflicting reports as for ther-
apeutic outcomes of a glatiramer acetate vaccination (46Y49),
in which an adjuvant is used only in the successful case (49)
but not in other failures (46Y48), it is possible that the repet-
itive challenges of a neuroprotective antigen without adjuvant
may not suffice to evoke protective immunity. Therefore, in
light of current understanding of the role of the inflamma-
tion in neurodegeneration, T
H
1/T
H
2 balance may determine a
detrimental/protective direction of neurons by regulating
microglial function (50). Thus, it is important to recognize
that a proper combination of antigen and adjuvant may syn-
ergistically augment the therapeutic effect of vaccination.
Notably, the positive correlation between IgG2b (a subclass
induced by TGFA) and the therapeutic effect, including delay
of disease onset and prolongation of the life span, prompts
us to consider the benefit of mucosal immunity such as nasal
or intestinal vaccination in which neuroprotective cytokines
such as TGFAand IL-4 are induced (51).
Vaccination of non-Tg mice with mouse WT SOD1
effectively induced antibody but did no overt harm. This is
very important for the future application of anti-SOD1 anti-
body in terms of potential adverse reactions and immunologic
tolerance. Abundant IgG2c as well as IgG1 was induced by
the vaccination indicating that T
H
1 immunity is provoked.
Therefore, further modifications for the improvement of
the vaccine effect include the development of misfolding
structureYtargeted vaccines, such as an oligomer or more
narrowed sequences involved in the pathogenesis, the choice
of adjuvant, and the route of vaccination.
In conclusion, our results indicate that WT-apo SOD1
may be a candidate as a vaccine for familial ALS patients
with SOD1 mutations. Moreover, they suggest that induction
of protective immunity and immunomodulation to suppress
T
H
1 immunity, particularly at the later stage, may enhance the
outcome. Despite the advantage of passive immunization,
which can abolish such immune response considerably, fur-
ther improvement of vaccines will prove beneficial to familial
ALS patients because of their long-lasting effects and of po-
tential noninvasive ways for overcoming adverse reactions.
ACKNOWLEDGMENTS
The authors thank H. Kita and S. Nakamura for the
technical assistance. The authors also thank Central Research
Laboratory of Shiga University of Medical Science for exper-
imental support. The authors are grateful to Ube Industries
Ltd for kindly providing the recombinant human SOD1,
chemically modified with 2-mercaptoethanol (2-ME-SOD1).
REFERENCES
1. Dion PA, Daoud H, Rouleau GA. Genetics of motor neuron disorders:
New insights into pathogenic mechanisms. Nat Rev Genet 2009;10:
769Y82
2. Rosen DR, Siddique T, Patterson D, et al. Mutations in Cu/Zn superoxide
dismutase gene are associated with familial amyotrophic lateral sclerosis.
Nature 1993;362:59Y62
3. Gurney ME, Pu H, Chiu AY, et al. Motor neuron degeneration in mice
that express a human Cu,Zn superoxide dismutase mutation. Science
1994;264:1772Y75
4. Clement AM, Nguyen MD, Roberts EA, et al. Wild-type nonneuronal
cells extend survival of SOD1 mutant motor neurons in ALS mice.
Science 2003;302:113Y17
5. Lobsiger CS, Cleveland DW. Glial cells as intrinsic components of
nonYcell-autonomous neurodegenerative disease. Nat Neurosci 2003;10:
1355Y60
6. Ilieva H, Polymenidou M, Cleveland DW. NonYcell autonomous toxicity
in neurodegenerative disorders: ALS and beyond. J Cell Biol 2009;187:
761Y72
7. Aguzzi A, Rajendran L. The transcellular spread of cytosolic amyloids,
prions, and prionoids. Neuron 2009;64:783Y90
8. Frost B, Diamond MI. Prion-like mechanisms in neurodegenerative dis-
eases. Nat Rev Neurosci 2010;11:155Y59
9. Urushitani M, Sik A, Sakurai T, et al. Chromogranin-mediated secretion
of mutant superoxide dismutase proteins linked to amyotrophic lateral
sclerosis. Nat Neurosci 2006;9:108Y18
10. Urushitani M, Ezzi SA, Julien JP. Therapeutic effects of immunization
with mutant superoxide dismutase in mice models of amyotrophic lateral
sclerosis. Proc Natl Acad Sci U S A 2007;104:2495Y500
11. Cao C, Lin X, Wahi MM, et al. Successful adjuvant-free vaccination
of BALB/c mice with mutated amyloid beta peptides. BMC Neurosci
2008;9:25
12. Kutzler MA, Cao C, Bai Y, et al. Mapping of immune responses fol-
lowing wild-type and mutant Abeta42 plasmid or peptide vaccination
in different mouse haplotypes and HLA class II Tg mice. Vaccine 2006;
24:4630Y39
13. He BP, Wen W, Strong MJ. Activated microglia (BV-2) facilitation of
TNF-alphaYmediated motor neuron death in vitro. J Neuroimmunol
2002;128:31Y38
14. Mir M, Asensio VJ, Tolosa L, et al. Tumor necrosis factor alpha and
interferon gamma cooperatively induce oxidative stress and motoneuron
death in rat spinal cord embryonic explants. Neuroscience 2009;162:
959Y71
15. Zhao W, Xie W, Xiao Q, et al. Protective effects of an anti-inflammatory
cytokine, interleukin-4, on motoneuron toxicity induced by activated
microglia. J Neurochem 2006;99:1176Y87
16. Kawamata T, Akiyama H, Yamada T, et al. Immunologic reactions in
amyotrophic lateral sclerosis brain and spinal cord tissue. Am J Pathol
1992;140:691Y707
17. Beers DR, Henkel JS, Zhao W, et al. CD4
+
T cells support glial neuro-
protection, slow disease progression, and modify glial morphology in
an animal model of inherited ALS. Proc Natl Acad Sci U S A 2008;105:
15558Y63
18. Chiu IM, Chen A, Zheng Y, et al. T lymphocytes potentiate endogenous
neuroprotective inflammation in a mouse model of ALS. Proc Natl Acad
Sci U S A 2008;105:17913Y18
19. Banerjee R, Mosley RL, Reynolds AD, et al. Adaptive immune neuro-
protection in G93A-SOD1 amyotrophic lateral sclerosis mice. PLoS One
2008;3:e2740
20. Rakhit R, Robertson J, Vande Velde C, et al. An immunological epitope
selective for pathological monomerYmisfolded SOD1 in ALS. Nat Med
2007;13:754Y59
21. Furukawa Y, O’Halloran TV. Amyotrophic lateral sclerosis mutations
have the greatest destabilizing effect on the apo- and reduced form
of SOD1, leading to unfolding and oxidative aggregation. J Biol Chem
2004;279:17266Y74
J Neuropathol Exp Neurol Volume 69, Number 10, October 2010 Vaccination in ALS Model Mice
Ó2010 American Association of Neuropathologists, Inc. 1055
Copyright @ 2010 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.
22. Urushitani M, Kurisu J, Tateno M, et al. CHIP promotes proteasomal
degradation of familial ALS-linked mutant SOD1 by ubiquitinating
Hsp/Hsc70. J Neurochem 2004;90:231Y44
23. Shibata N, Hirano A, Kobayashi M, et al. Presence of Cu/Zn superoxide
dismutase (SOD) immunoreactivity in neuronal hyaline inclusions in
spinal cords from mice carrying a transgene for Gly93Ala mutant human
Cu/Zn SOD. Acta Neuropathol 1998;95:136Y42
24. Gruzman A, Wood WL, Alpert E, et al. Common molecular signature in
SOD1 for both sporadic and familial amyotrophic lateral sclerosis. Proc
Natl Acad Sci U S A 2007;104:12524Y29
25. Fujiwara N, Nakano M, Kato S, et al. Oxidative modification to cysteine
sulfonic acid of Cys111 in human copper-zinc superoxide dismutase.
J Biol Chem 2007;282:35933Y44
26. Sato T, Takeuchi S, Saito A, et al. Axonal ligation induces transient
redistribution of TDP-43 in brainstem motor neurons. Neuroscience
2009;164:1565Y78
27. Tiwari A, Liba A, Sohn SH, et al. Metal deficiency increases aberrant
hydrophobicity of mutant superoxide dismutases that cause amyotrophic
lateral sclerosis. J Biol Chem 2009;284:27746Y58
28. Baldridge JR, Yorgensen Y, Ward JR, et al. Monophosphoryl lipid A
enhances mucosal and systemic immunity to vaccine antigens following
intranasal administration. Vaccine 2000;18:2416Y25
29. Ferri A, Nencini M, Cozzolino M, et al. Inflammatory cytokines increase
mitochondrial damage in motoneuronal cells expressing mutant SOD1.
Neurobiol Dis 2008;32:454Y60
30. Ford L, Rowe D. Interleukin-12 and interferon-gamma are not detectable
in the cerebrospinal fluid of patients with amyotrophic lateral sclerosis.
Amyotroph Lateral Scler Other Motor Neuron Disord 2004;5:118Y20
31. Rivest S. Regulation of innate immune responses in the brain. Nat Rev
Immunol 2009;9:429Y39
32. Racz I, Nadal X, Alferink J, et al. Interferon-gamma is a critical modu-
lator of CB(2) cannabinoid receptor signaling during neuropathic pain.
J Neurosci 2008;28:12136Y45
33. Lobsiger CS, Boillee S, Cleveland DW. Toxicity from different SOD1
mutants dysregulates the complement system and the neuronal regener-
ative response in ALS motor neurons. Proc Natl Acad Sci U S A 2007;
104:7319Y26
34. Ezzi SA, Urushitani M, Julien JP. Wild-type superoxide dismutase
acquires binding and toxic properties of ALS-linked mutant forms
through oxidation. J Neurochem 2007;102:170Y78
35. Zhong Z, Deane R, Ali Z, et al. ALS-causing SOD1 mutants generate
vascular changes prior to motor neuron degeneration. Nat Neurosci 2008;
11:420Y22
36. Zeis T, Graumann U, Reynolds R, et al. Normal-appearing white matter
in multiple sclerosis is in a subtle balance between inflammation and
neuroprotection. Brain 2008;131:288Y303
37. Kipnis J, Avidan H, Caspi RR, et al. Dual effect of CD4
+
CD25
+
regu-
latory T cells in neurodegeneration: A dialogue with microglia. Proc
Natl Acad Sci U S A 2004;101:14663Y69
38. Gao X, Gillig TA, Ye P, et al. Interferon-gamma protects against
cuprizone-induced demyelination. Mol Cell Neurosci 2000;16:338Y49
39. Nicholas RS, Stevens S, Wing MG, et al. Microglia-derived IGF-2
prevents TNF>induced death of mature oligodendrocytes in vitro.
J Neuroimmunol 2002;124:36Y44
40. Baron R, Nemirovsky A, Harpaz I, et al. IFN-gamma enhances neuro-
genesis in wild-type mice and in a mouse model of Alzheimer’s disease.
FASEB J 2008;22:2843Y52
41. Babu GN, Kumar A, Chandra R, et al. Elevated inflammatory markers in
a group of amyotrophic lateral sclerosis patients from northern India.
Neurochem Res 2008;33:1145Y49
42. Hozumi H, Asanuma M, Miyazaki I, et al. Protective effects of
interferon-gamma against methamphetamine-induced neurotoxicity.
Toxicol Lett 2008;177:123Y29
43. Takeuchi H, Wang J, Kawanokuchi J, et al. Interferon-gamma induces
microglial-activationYinduced cell death: A hypothetical mechanism of
relapse and remission in multiple sclerosis. Neurobiol Dis 2006;22:33Y39
44. Shechter R, London A, Varol C, et al. Infiltrating blood-derived macro-
phages are vital cells playing an anti-inflammatory role in recovery from
spinal cord injury in mice. PLoS Med 2009;6:e1000113
45. Weiner HL, Lemere CA, Maron R, et al. Nasal administration of
amyloid-beta peptide decreases cerebral amyloid burden in a mouse
model of Alzheimer’s disease. Ann Neurol 2000;48:567Y79
46. Haenggeli C, Julien JP, Mosley RL, et al. Therapeutic immunization with
a glatiramer acetate derivative does not alter survival in G93A and G37R
SOD1 mouse models of familial ALS. Neurobiol Dis 2007;26:146Y52
47. Habisch HJ, Schwalensto¨cker B, Danzeisen R, et al. Limited effects of
glatiramer acetate in the high-copy number hSOD1-G93A mouse model
of ALS. Exp Neurol 2007;206:288Y95
48. Meininger V, Drory VE, Leigh PN, et al. Glatiramer acetate has no
impact on disease progression in ALS at 40 mg/day: A double- blind,
randomized, multicentre, placebo-controlled trial. Amyotroph Lateral
Scler 2009;10:378Y83
49. Angelov DN, Waibel S, Guntinas-Lichius O, et al. Therapeutic vaccine
for acute and chronic motor neuron diseases: implications for amyo-
trophic lateral sclerosis. Proc Natl Acad Sci U S A 2003;100:4790Y95
50. Appel SH, Beers DR, Henkel JS. T cell-microglial dialogue in Parkin-
son’s disease and amyotrophic lateral sclerosis: Are we listening? Trends
Immunol 2010;31:7Y17
51. Bhowmick S, Mazumdar T, Ali N. Vaccination route that induces
transforming growth factor beta production fails to elicit protective
immunity against Leishmania donovani infection. Infect Immun 2009;77:
1514Y23
Takeuchi et al J Neuropathol Exp Neurol Volume 69, Number 10, October 2010
Ó2010 American Association of Neuropathologists, Inc.
1056
    • "Wild-type non-metallated SOD1 (WT-apo) and G93A SOD1 vaccines have been studied in the SOD1 G93A mouse model. Both vaccines significantly delayed disease onset and increased the lifespan of SOD1 G93A mice [191]. A vaccine approach is being taken by the company Amorfix together with Biogen Idec. "
    [Show abstract] [Hide abstract] ABSTRACT: Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease, associated with motor neuron degeneration, muscle weakness, paralysis and finally death. The proposed mechanisms of ALS include glutamate excitotoxicity, oxidative stress, inflammation, mitochondrial dysfunction, apoptosis and proteasomal dysfunction. Although numerous pathological mechanisms have been explained, ALS remains incurable disease because of failure of clinical trials and lack of any effective therapy. The rapid advancement in genetic discoveries in ALS emphasizes the point that ALS is a multi-subtype syndrome rather than a single disease. This can be argued as one of the single reason why many previous therapeutic drug trials have failed. Efforts to develop novel ALS treatments which target specific pathomechanisms are currently being pursued. Herein, we review the recent discovery and preclinical characterization of neuroprotective compounds and compare their effects on disease onset, duration and survival. Furthermore, the structure-activity relationships of these agents are analyzed with the overall goal of developing a screening strategy for future clinical applications.
    Full-text · Article · Jun 2016
    • "This strategy delays disease onset and increases the life span of mSOD1 mice longer than other therapeutic methods [95]. A SOD1 vaccine was developed for the wild type protein and used on mSOD1 mice, also achieving a delay of disease onset [96,97]. A very promising strategy for ALS patients is vaccination with effector T cells against neural constituents during the symptomatic disease phase; which combined with autologous neural stem cell treatment , achieves neurologic improvements probably by establishment of self-tolerance against neural constituents [98]. "
    [Show abstract] [Hide abstract] ABSTRACT: The progressive loss of neurons and inflammation characterizes neurodegenerative diseases. Although the etiology, progression and outcome of different neurodegenerative diseases are varied, they share chronic inflammation maintained largely by central nervous system (CNS)-derived antigens recognized by T cells. Inflammation can be beneficial by recruiting immune cells to kill pathogens or to clear cell debris resulting from the primary insult. However, chronic inflammation exacerbates and perpetuates tissue damage. An increasing number of therapies that attempt to modulate neuroinflammation have been developed. However, so far none has succeeded in decreasing the secondary damage associated with chronic inflammation. A potential strategy to modulate the immune system is related to the induction of tolerance to CNS antigens. In this line, it is our hypothesis that this could be accomplished by using anterior chamber associated immune deviation (ACAID) as a strategy. Thus, we review current knowledge regarding some neurodegenerative diseases and the associated immune response that causes inflammation. In addition, we discuss further our hypothesis of the possible usefulness of ACAID as a therapeutic strategy to ameliorate damage to the CNS.
    Full-text · Article · Jan 2016 · Frontiers in Cellular Neuroscience
    • "While the forest plots look at studies compared to each other, the meta-analysis using normalized data operates under a similar fixed effects model in which the value of each study, instead of the study itself, is weighted using sample size. Hensley et al., 2002 Hensley et al., , 2003 Hensley et al., , 2006 Yoshihara et al., 2002; Kang et al., 2003; Chen et al., 2004; Weydt et al., 2004; Xie et al., 2004; Perrin et al., 2005; Henkel et al., 2006; Kiaei et al., 2006; Kim et al., 2006; Liu and Martin, 2006; Ohta et al., 2006; Pardo et al., 2006; Guan et al., 2007; Petrik et al., 2007; Yin et al., 2007; Beers et al., 2008; Pitzer et al., 2008; Poesen et al., 2008; Shibata et al., 2008; Zhang et al., 2008; Boucherie et al., 2009; Cheroni et al., 2009; Kassa et al., 2009; Keller et al., 2009 Keller et al., , 2011 Neymotin et al., 2009; Ringer et al., 2009; Sekiya et al., 2009; Yang et al., 2009 Yang et al., , 2011 Fang et al., 2010; Guo et al., 2010 Guo et al., , 2011 Israelsson et al., 2010; Jokic et al., 2010; Moreno-Igoa et al., 2010; Shimazawa et al., 2010; Steinacker et al., 2010; Suzuki et al., 2010; Takeuchi et al., 2010; Tsai et al., 2010; Zhao et al., 2010; Berger et al., 2011; Finkelstein et al., 2011; Genestine et al., 2011; Pollari et al., 2011; Audet et al., 2012; Dibaj et al., 2012; Gifondorwa et al., 2012; Miquel et al., 2012; Valente et al., 2012 Type II total 10 25 Hensley et al., 2002 Hensley et al., , 2003 Weydt et al., 2004; Xie et al., 2004; Kim et al., 2006; Beers et al., 2008; Chiu et al., 2008; Finkelstein et al., 2011; Audet et al., 2012; Dibaj et al., 2012 Information gathered on each cytokine that was suitable for use in the statistical analysis. "
    [Show abstract] [Hide abstract] ABSTRACT: Amyotrophic Lateral Sclerosis (ALS) is a fatal motoneuron disease that is characterized by the degradation of neurons throughout the central nervous system. Inflammation have been cited a key contributor to ALS neurodegeneration, but the timeline of cytokine upregulation remains unresolved. The goal of this study was to temporally examine the correlation between the varying levels of pro-inflammatory type I cytokines (IL-1β, IL-1α, IL-12, TNF-α, and GFAP) and anti-inflammatory type II cytokines (IL-4, IL-6, IL-10) throughout the progression of ALS in the SOD1 G93A mouse model. Cytokine level data from high copy SOD1 G93A transgenic mice was collected from 66 peer-reviewed studies. For each corresponding experimental time point, the ratio of transgenic to wild type (TG/WT) cytokine was calculated. One-way ANOVA and t-tests with Bonferonni correction were used to analyze the data. Meta-analysis was performed for four discrete stages: early, pre-onset, post-onset, and end stage. A significant increase in TG cytokine levels was found when compared to WT cytokine levels across the entire SOD1 G93A lifespan for majority of the cytokines. The rates of change of the individual cytokines, and type I and type II were not significantly different; however, the mean fold change of type I was expressed at significantly higher levels than type II levels across all stages with the difference between the means becoming more pronounced at the end stage. An overexpression of cytokines occurred both before and after the onset of ALS symptoms. The trend between pro-inflammatory type I and type II cytokine mean levels indicate a progressive instability of the dynamic balance between pro- and anti-inflammatory cytokines as anti-inflammatory cytokines fail to mediate the pronounced increase in pro-inflammatory cytokines. Very early immunoregulatory treatment is necessary to successfully interrupt ALS-induced neuroinflammation.
    Full-text · Article · Dec 2015
Show more