Neurobiology of Disease 37 (2010) 370383

Contents lists available at ScienceDirect

Neurobiology of Disease
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y n b d i

A systematic study of brainstem motor nuclei in a mouse model of ALS,

the effects of lithium
Michela Ferrucci a,1, Alida Spalloni b,1, Alessia Bartalucci a, Emanuela Cantafora a, Federica Fulceri a,
Michele Nutini b, Patrizia Longone b, Antonio Paparelli a, Francesco Fornai a,c,
a
b
c

Department of Human Morphology and Applied Biology, University of Pisa, via Roma 55, 56126 Pisa, Italy
Molecular Neurobiology Unit, Santa Lucia Foundation, via del Fosso di Fiorano, 00143 Rome, Italy
Laboratory of Neurobiology of Movement Disorders, I.N.M. I.R.C.C.S. Neuromed, Localit Camerelle, 86077 Pozzilli (IS), Italy

a r t i c l e

i n f o

Article history:
Received 14 August 2009
Revised 1 October 2009
Accepted 14 October 2009
Available online 27 October 2009
Keywords:
Amyotrophic lateral sclerosis
Transgenic G93A mouse
Brainstem motor neurons
Synaptic boutons
Choline acetyl transferase
Lithium

a b s t r a c t
Transgenic mice expressing the human superoxide dismutase 1 (SOD-1) mutant at position 93 (G93A)
develop a phenotype resembling amyotrophic lateral sclerosis (ALS). In fact, G93A mice develop progressive
motor decits which nally lead to motor palsy and death. This is due to the progressive degeneration of
motor neurons in the ventral horn of the spinal cord. Although a similar loss is reported for specic cranial
motor nuclei, only a few studies so far investigated degeneration in a few brainstem nuclei. We recently
reported that chronic lithium administration delays onset and duration of the disease, while reducing
degeneration of spinal motor neuron. In the present study, we extended this investigation to all somatic
motor nuclei of the brain stem in the G93A mice and we evaluated whether analogous protective effects
induced by lithium in the spinal cord were present at the brain stem level. We found that all motor but the
oculomotor nuclei were markedly degenerated in G93A mice, and chronic treatment with lithium
signicantly attenuated neurodegeneration in the trigeminal, facial, ambiguus, and hypoglossal nuclei.
Moreover, in the hypoglossal nucleus, we found that recurrent collaterals were markedly lost in G93A mice
while they were rescued by chronic lithium administration.
2009 Elsevier Inc. All rights reserved.

Introduction
Amyotrophic lateral sclerosis (ALS) is a progressive, devastating
neurodegenerative disease that affects primarily motor neurons (MN)
of the spinal cord, brainstem and motor cortex (Wijesekera and Leigh,
2009; Cleveland and Rothstein, 2001). The clinical symptoms include
weakness, muscle atrophy and fasciculations, death occurs in 3
5 years from diagnosis (Rowland and Shneider, 2001).
When the motor nuclei of the brainstem are involved at the onset
of disease (roughly, 25% of cases), the clinical outcome is the most
severe leading to death in less than 1 year from diagnosis (Khnlein et
al., 2008). In fact, the involvement of the brainstem leads to
dysarthria, dysphagia and impairment of swallowing and breathing
all due to a damage of hypoglossal, trigeminal and ambiguus nuclei.
Also the facial nucleus is affected and contributes to these symptoms
(Kusaka et al., 1988; Hartmann et al., 1989). In contrast, oculomotor
nuclei are relatively spared (Gizzi et al., 1992; Okamoto et al., 1993).

Corresponding author. Department of Human Morphology and Applied Biology,

University of Pisa, 56126, Pisa, Italy. Fax: +39 050 2218606.
E-mail address: f.fornai@med.unipi.it (F. Fornai).
1
Equally contributed to the present work.
Available online on ScienceDirect (www.sciencedirect.com).
0969-9961/$ see front matter 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.nbd.2009.10.017

Based on the knowledge of inherited ALS in humans, a variety of

mouse strains have been generated. Among these, the most studied is
the G93A mouse, where a point mutation in the human gene coding the
enzyme superoxide dismutase type-1 (SOD1) leads to a 93 glycine/
alanine substitution. These mice develop a rapidly progressive MN
disease, which leads to hindlimb paralysis and death (Gurney et al.,
1994; Ripps et al., 1995; Gurney, 1997). This phenotype recapitulates
several clinical and histopathological features of both familial and
sporadic forms of the human disease (Gurney et al., 1994; Gurney, 1997;
Newbery and Abbott, 2001). Although lumbar spinal cord is the site of
onset and it is mostly affected, MN degeneration also occurs at higher
levels of the cord and brainstem. It is surprising that, despite brainstem
degeneration was described in this mouse model, and the brainstem
involvement leads to the worst prognosis of ALS in humans, only a few
studies investigated the brainstem motor nuclei in the G93A mouse.
Therefore, we decided to analyze the entire population of somatic
brainstem motor nuclei of the G93A mouse also including the
parasympathetic nucleus dorsalis of the vagus.
Within these nuclei, we counted the degeneration occurring
at the end of the natural course of the disease, and we evaluated
the effects of lithium. In fact, recent reports (Shin et al., 2007; Fornai
et al., 2008a,b; Feng et al., 2008; Pasquali et al., 2009) indicate that
chronic lithium administration protects G93A mice from MN
degeneration (Shin et al., 2007; for a review, see Pasquali et al.,

M. Ferrucci et al. / Neurobiology of Disease 37 (2010) 370383

2009 ; Young, 2009) and rescues the behavioural and pathological

decit occurring following a spinal cord transection (Dill et al., 2008).
These effects are concomitant with lithium-induced increased
neuron number and decreased gliosis as shown both in degenerating
(Fornai et al., 2008a,b) and lesioned (Su et al., 2007, 2009) spinal
cords as well as lithium-induced sprouting of corticospinal (Dill et al.,
2008) and rubrospinal (Yick et al., 2004) tracts.
Materials and methods
Animals
We used male mice from our own colony originated from the G93A
mouse strain originally obtained from the Jackson laboratories (Bar
Harbor, ME, USA). Selective breeding maintained the transgene in the
hemizygous state in an F1 hybrid C57BL6xSJL genetic background.
Colony maintenance and screening for the presence of the human
transgene were performed as described (Spalloni et al., 2006). In
addition, we replicated the experiments in G93A mice directly
purchased from Jackson laboratories (Bar Harbor, ME, USA) via
Charles River (Calco, LC, Italy) as provider from Jackson laboratories

371

for Italy. All experiments were conducted in compliance with the

European Council directive (86/609/EEC) for the use and care of
laboratory animals. The experiments were carried out in male mice,
treated from postnatal day 67 until the tetraplegic stage (end point).
Experimental groups
The G93A transgenic mice and their non-transgenic wild-type
(WT) littermates were divided into 4 experimental groups (n = 9):
WT mice receiving saline; WT mice receiving lithium; G93A mice
receiving saline; and G93A mice receiving lithium.
In the main experiment, for each experimental group, 9 mice were
used as follows: 3 mice per group were used for confocal microscopy
while 6 were processed for plain light microscopy. These very same
mice were used for behavioural analysis (n = 6). In this experiment,
mice received lithium carbonate.
Four additional experimental groups, each containing n = 3
animals, used both for plain light microscopy and behavioural
analysis, were used to evaluate a potential change using lithium
chloride, which was not the case. We reported the rough number
obtained using lithium carbonate.

Fig. 1. Representative cresyl violet and H&E-stained sections through the hypoglossal nuclei (12N) in WT saline and lithium (A, B) and in G93A saline and lithium (C, D) mice. The MN
count (J) showed a decrease in neuron amount in the G93A saline mice (C) compared to WT (A, B) and the partial recovery following lithium treatment (D). Values are mean SEM
(P 0.001 vs. WT sal.; #P 0.05 vs. G93A sal, WT sal, and WT li, F-value = 17.168). The circle indicates the hypoglossal area considered for the cell count, also showed with the arrow
in the image extracted from the mouse atlas of Paxinos and Franklin (2004, I). (EH) Representative high magnication of H&E-stained neurons in the hypoglossal nucleus of WT
saline- and lithium-treated (E, F) and G93A saline- and lithium-treated (G, H) mice. Note the decrease in intracellular and extracellular vacuolization and the neuronal recovery in the
lithium-treated compared to the saline-treated G93A mouse. The asterisks indicate the intracellular and extracellular vacuoles. Scale bars = 367.64 m (AD) and 33.33 m (EH).

372

M. Ferrucci et al. / Neurobiology of Disease 37 (2010) 370383

Fig. 2. Neurons and astrocytes in the hypoglossal nucleus area (12N) in WT (A) and G93A saline- and lithium-treated (B, C) mice. The representative double immunouorescence of
NeuN and GFAP revealed the presence of astrocytes in hypoglossal nucleus of G93A saline (B) and lithium (C) compared to the WT (A), and the preservation in shape and neuron
density of the NeuN+ cells in the G93A lithium-treated mouse (C) compared with the saline-treated (B). Scale bar = 250 m.

In detail, mice were administered with either saline solution (0.9%

sodium chloride) or lithium carbonate (Sigma, St. Louis, MO, USA)
1 mEq/kg, dissolved in saline solution) every other day, i.p., starting at
67 days of age. This seems a convenient time even considering three
recent works that looked at the brainstem degeneration. At rst the
papers by Lever et al. (2009a,b) where the authors demonstrate that
early at pre-motor stage (60 days), the mice show impairments in
licking and mastication rate. Moreover, Bucher et al. (2007), by using
magnetic resonance imaging and H&E, have shown in the SOD1G93A
mice that the facial hypoglossal and trigeminal nucleus degenerate
starting at 80 days.
An additional group of mice (n = 3) received lithium chloride
(Sigma).
The volume of the injection was adjusted to the mouse weight
to keep constant the dose of 1 mEq/kg. Saline or lithium were

Fig. 3. Motor neuron loss in the hypoglossal area (12N) in WT saline- and lithiumtreated (A, B) and G93A saline- and lithium-treated (C, D) mice. The immunouorescence against ChAT shows the partial MN loss in the G93A saline (C) compared to the
WT (A, B) and the recovery after lithium treatment (D). ChAT, choline acetyltransferase; CC, central canal. Scale bar = 250 m.

administered in the morning, between 0900 and 1200 h; after

injections, mice were tested for behavior (see later).
Behavior
Motor tests started at 65 days of age, 2 days before treatment
started. Tests were performed weekly for all animal groups (n = 6 per
group). The behavioural observations were made by blind observers.
Survival time, motor impairment, motor strength and coordination
were evaluated with the following tests.
Stride length test
The stride length test was performed with slight modications
accordingly to the method reported by Fernagut et al. (2002). The

Fig. 4. ChAT+ cholinergic boutons amount on motor neuron body in the hypoglossal
area in WT saline and lithium (A, B) and G93A saline and lithium (C, D) mice. The single
immunouorescence shows the cholinergic boutons loss in G93A saline (C) compared
with WT (A, B) and the recovery after lithium treatment (D). Arrowheads indicate the
distinct bouton. Represented in the table are the ChAT+ boutons showing the decrease
in the G93A saline-treated mice and the statistically signicant recovery under lithium
treatment. Scale bar = 11.36 m.

M. Ferrucci et al. / Neurobiology of Disease 37 (2010) 370383

Table 1
Lithium attenuates the loss of ChAT+ cells and counteracts the loss of ChAT+ boutons
per cell in the hypoglossal nucleus.

No. mice
No. cells counted in
the nucleus
No. boutons counted
per cell

WT saline

WT lithium

G93A saline

G93A lithium

3
75.8 2.4

3
81.4 3.2

3
45.3 2.0

3
58.5 3#

2.556 0.10

2.511 0.15

0.827 0.21

2.036 0.17#

The quantication of ChAT+ cholinergic cells was carried out using progressive slices
(15 m) observed at confocal microscopy. ChAT+ cholinergic boutons were counted in
an average of 100 positive cells per group. Values represent the mean SEM.
P b 0.01 compared with WT saline and #P b 0.01 compared with G93A saline by oneway ANOVA testing with NewmanKeuls post hoc test.

apparatus was composed of an open eld (80 80 30 cm), illuminated by a light, in which a runway wide (75 5 cm) was arranged to
lead out into a dark box (20 15 10 cm). The hind paws of the mouse
were painted with blue ink and the mouse was allowed to run on a
strip of paper down the brightly runway towards the dark box. Stride
lengths were measured manually as the distance between two
pawprints. The three longest stride lengths, corresponding to
maximal velocity, were measured from each run. Data were obtained
as the mean of the three longest stride lengths. Mouse unable to walk
was graded as zero.
Paw grip endurance test
The motor strength was assessed with the paw grip endurance
(PaGE) test (Weydt et al., 2003). Animals were placed individually on
a meshed wire lid shaken to induced the mice to grip the grid. Then
the lid was gently turned upside-down and the latency until a mouse
remained hold on was recorded with a cutoff time of 90 s. Each mouse
was given three consecutive trials and the longest latency was
recorded. Animals unable to grip the grid received a score of zero.

373

Rotarod
Motor impairment was evaluated with a rotarod. Each mouse was
place on a rod rotating at 15 rpm and the time the mouse stayed on
the rod until it fell off (during a 10-min interval) was recorded. Three
consecutive trials were performed and the best result of the three
trials was recorded. As an additional parameter, body weight was
monitored.
Histological analysis
We carried out the same procedures as described in Fornai et al.
(2008a). Briey, mice were anaesthetized by chloral hydrate and then
perfused transcardially with saline solution, followed by a xative
solution consisting of 4% paraformaldehyde in 0.1 M phosphate buffer,
pH 7.3. The brain and the spinal cord, at a level corresponding to the
lumbar tract (identied by the presence of an increased diameter),
were dissected out and postxed in the same xative solution for 24 h
and then transferred in 70% ethylic alcohol overnight at 4 C. Samples
were dehydrated in increasing alcohol solutions, immersed in xylene
for several hours, and nally embedded in parafn. Brains were
sectioned coronally using a microtome in order to obtain 7- to 8-mthick slices.
Spinal cord
Sections of lumbar spinal cord were analyzed in order to conrm
the MN loss in the G93A mice and the effects of lithium treatment on
neuron number and morphology as previously described (Fornai
et al., 2008a).
Brainstem
Sections corresponding to the motor nuclei of the brainstem were
collected in strict anatomical order and mounted on polylysinated
slides. In order to perform a stereological analysis of the neurons

Fig. 5. The synaptic cholinergic bouton clusters on motor neuron membrane in the hypoglossal nucleus area. The immunouorescence depicts as a single channel the presynaptic
marker synaptotagmin 1 (Syn1, A) and the cholinergic boutons (ChAT, B) in WT. (C) Merge of the two images. Hoechst (blue) indicates the nuclei. Arrowheads (in the enlarged
image) indicate ChAT+ clusters that colocalize with synaptotagmin1. The square dotted shows the cell magnied in the bottom panels. Scale bars = 45.44 m (low magnications)
and 15.15 m (high magnications).

374

M. Ferrucci et al. / Neurobiology of Disease 37 (2010) 370383

Fig. 6. Representative cresyl violet- and H&E-stained sections through the dorsal motor nucleus of the vagus (10N) in WT saline- and lithium- (A, B) and in G93A saline- and lithiumtreated (C, D) mice. The MN count (F) shows a decrease in neuron number in the G93A saline-treated mice compared with WT and the lack of recovery following lithium treatment.
Values are mean SEM (P 0.001 vs. WT). The high H&E magnication shows the absence of recovery in neuron structure following lithium exposure in the G93A. The circles
indicate the area considered for the cell count, also showed with arrows in the image extracted from the mouse atlas of Paxinos and Franklin (2004, E). Scale bars = 35.71 m (H&E,
high magnication) and 367.64 m (cresyl violet, low magnication).

belonging to the brainstem motor nuclei, serial sections (collected as

described below in the dedicated paragraph) were stained by cresyl
violet. Briey, sections were dried at 37 C, immersed in xylene to
remove the parafn, re-hydrated by immersion in decreasing alcohol
solutions, and then stained by cresyl violet. After dehydration, slides
were immersed in xylene, coverslipped with DPX plastic mounting
media (Sigma), and observed at light microscope Nikon Eclipse 80i.
Proper identication of the brainstem nuclei was conrmed by using
comparable tables of the mouse atlas by Paxinos and Franklin (2004)
based on stereotaxic coordinates.
Sections which were not used for the stereological analysis were
stained by H&E and used to analyze at light microscopy the
morphological properties of the brainstem neurons.
Stereological analysis
The stereological analysis was carried out in 6 animals per group.
The neuronal loss in the ventral horn of the lumbar spinal cord and
within the brainstem motor nuclei was estimated by counting the
number of neurons at 20 magnication by two different observers,
blind to treatments.

Spinal cord
Spinal MN within the Lamina IX were identied by combining
morphological (multipolar cells with not condensed nucleus and
well-evident nucleolus) and size exclusion criteria, consisting of
counting only those neurons exhibiting a diameter of at least 30 m.
This latter criterion allowed us to consider with high specicity the
population of alpha-MN (see Fornai et al., 2008a). For each mouse,
counts of the Lamina IX MN were carried out on a total of 125
consecutive, not serial, cresyl violet-stained slices, spaced 80 m each
other.
Brainstem
For the exact identication of each brainstem motor nucleus, we
used the Paxinos and Franklin atlas (2004). As cranial motor nuclei are
readily dened, the neurons of the entire nuclei were counted. For
each nucleus, we analyzed the sections within the following AP
stereotaxic coordinates (expressed in mm posterior to the bregma):
a) 3.804.24 for the oculomotor nucleus;
b) 4.364.48 for the trochlear nucleus;
c) 4.965.34 for the trigeminal motor nucleus;

M. Ferrucci et al. / Neurobiology of Disease 37 (2010) 370383

d) 5.525.80 for the abducens nucleus;

e) 5.686.48 for the facial nucleus;
f) 6.708.00 for the ambiguus nucleus;
g) 6.507.90 for the dorsal motor nucleus of the vagus nerve;
h) 7.008.12 for the hypoglossal nucleus.
The size exclusion criteria used in Fornai et al. (2008a) to count
alpha-MN in the spinal cord was not feasible for the brainstem, where
a general size exclusion does not allow to identify selectively alphaMN in each nucleus. This is due both to the general smaller size of
brainstem MN compared with those in the spinal cord and to the high
variability in MN size between different brainstem nuclei.
In fact, the size of MN is related to the dimension of the motor unit
and the metameric organization of the efferent projections from the
nucleus. Some brainstem motor nuclei, such as trigeminal, facial, and
hypoglossal nucleus, exhibit a very heterogeneous population of MN,
characterized by a different size of the cell body. In particular, within
the trigeminal motor nucleus, MN diameter ranges from about 15 m

375

(small MN) to 40 m (large MN); in the facial motor nucleus, the

range of the MN diameter is even wider, from 5 to 55 m; nally, in
the hypoglossal nucleus, MN size is about 1750 m.
For this reason, the analysis was not limited to alpha-MN (N30 m)
and we chose to count all neurons sized more than 15 m. In this way,
we used an extended size exclusion criterion and we counted within
each brainstem motor nucleus all neurons longer than 15 m. When
the MN diameter was very similar to other small neurons, we carefully
combined dimensional and morphological exclusion criteria.
In detail, the neuronal counts were carried out in every fth
section spaced about 50 m, accordingly to several authors (Llad
et al., 2006; Haenggeli and Kato, 2002), in which we counted all
neurons longer than 15 m.
Measurement of the neuronal diameter was determined by an image
analysis software (Molecular Machine & Industries AG, Glattburgh,
Switzerland) which allowed to draw the diameter of the cell under
observation and automatically read the related numerical value. The
analysis of each section was carried out at 20 magnication with a

Fig. 7. Representative cresyl violet- and H&E-stained sections through the facial nucleus (7N) in WT saline- and lithium-treated (A, B) and in G93A saline- and lithium-treated (C, D)
mice. Note the normal histological appearance of the nucleus in WT (A, B) compared with the extensive degeneration observed in the G93A saline (C) mice. The MN count (F)
showed a decrease in neuron amount in the G93A saline-treated mice compared with WT and the partial recovery following lithium treatment. Values are mean SEM (P 0.001
vs. WT sal.; #P 0.05 vs. G93A sal, WT sal, and WT li, F-value = 13.353). The high H&E magnication shows the recovery in neuron structure following lithium exposure. The circles
indicate the nucleus area considered for the cell count, also showed with arrows in the image extracted from the mouse atlas of Paxinos and Franklin (2004, E). Scale bars = 50 m
(H&E, high magnication) and 588 m (cresyl violet, low magnication).

376

M. Ferrucci et al. / Neurobiology of Disease 37 (2010) 370383

Fig. 8. Neurons and astrocytes in the facial nucleus (7N) in G93A saline-treated (B) and lithium-treated (C) mice compared to WT (A). Double immunouorescence of NeuN and
GFAP reveals the neuronal loss in the saline-treated mouse (B) and the recovery after lithium (C), while the reactive gliosis is not controlled by the treatment with lithium. The dotted
square indicates the area magnied in panels D (saline) and E (lithium). Scale bars = 190 m.

motorized system that allowed to scan the slice throughout the XY plan.
The movement of the stage was controlled by an automatic scanning
motorized system 120 100 mm (Marzhauser, Molecular Machine &
Industries Smart-CutR, Glattbrugg, Switzerland). This procedure
allowed us to analyze in a very careful manner the tissue region
corresponding to the nucleus under observation and to count within
such a well-dened area all neurons exhibiting sizes which matched
with the above mentioned range.
In summary, this procedure provided a useful tool to virtually rebuild the neuronal 3D size, avoiding to count the same neuron twice
(sections spaced 50 m) and taking into account only motor neurons
(mostly larger than 15 m and exhibiting a typical feature).
Brainstem section preparation for immunouorescence and confocal
microscopy
Saline and lithium-treated mice were deeply anesthetized with
500 mg/kg i.p. chloral hydrate and perfused intracardially with 100 ml
of saline followed by 100 ml of 4% paraformaldehyde in a phosphate
buffer (PB; 0.1 M; pH 7.4). Each brain was immediately removed,
postxed in the same xative for 12 h, and after three washes in PB
transferred to 30% sucrose in PB solution at 4 C until it sank. Brainstem
transversal cryostat cut sections ( 10 m) were thaw-mounted onto
polylysine-coated slides and stored at 70 C until use.
Before incubation with primary antibodies, sections were incubated in a blocking solution (5% normal donkey serum, 0.3% Triton X100, PB) for 1 h at room temperature. All solutions containing the
primary antibodies were prepared in PB and 0.3% Triton X-100 and
incubated overnight at 4 C. Sections were incubated with the
following antibodies: mouse anti-NeuN (1:100; Chemicon), rabbit
anti-GFAP (1:500; Chemicon), goat anti- choline acetyltransferase
(ChAT) (1:100; Chemicon), and rabbit anti-synaptotagmin 1 cytoplasmatic domain (1:100; Synaptic System). Subsequently, sections
were incubated for 2 h at room temperature with the secondary

antibodies Cy3-conjugated donkey anti-mouse IgG, Cy2-conjugated

donkey anti-rabbit IgG, and Cy3-conjugated donkey anti-goat IgG
(1:100; Jackson Immunoresearch Laboratories). For nuclei staining,
after secondary antibodies incubation, the sections were exposed to
Hoechst 33342 (Molecular Probes). The sections were examined using
a confocal laser scanning microscope (Leica SP5, Leica Microsystems,
Wetzlar, Germany).
The ChAT+ boutons were quantied in four to ve sections (40
enlargement) for each animal at comparable levels along the
rostrocaudal axis of the hypoglossal nucleus (Bregma 7.008.00;
Paxinos and Franklin, 2004). The ChAT+ terminals were counted
directly on the confocal images using the Leica SP5 proprietary
program. Only boutons clearly located around the pericaryon of the
ChAT+ neurons within the hypoglossal nuclei were counted. The
arrowhead in Fig. 5 indicates the ChAT+ boutons. The images are a
63 enlargement to clearly show the boutons.
Statistics
Measurements of neuronal number from each mouse carried out
by each observer were used to obtain the mean value referred to each
group. Comparisons among groups were made by using a one-way
analysis of variance ANOVA combined with Fisher and Scheff post
hoc tests. F values for these comparisons were calculated. Null
hypothesis was rejected for P b 0.05.
Counts of the ChAT+ boutons was carried out by an observer blind
to the treatment and presented as mean SEM of the ratio between
the cholinergic boutons number and the amount of the ChAT+ cells.
The values were compared by using the one-way analysis of variance
ANOVA testing with NewmanKeuls post hoc test. Null hypothesis
was rejected for P b 0.01.
Measurement of behavioural performance is expressed as mean
SEM and values were compared by using Student's t-test. Null
hypothesis was rejected for P b 0.05.

M. Ferrucci et al. / Neurobiology of Disease 37 (2010) 370383

Results
Chronic lithium administration extended the survival time and
signicantly improved the motor function of the G93A mouse used
in this study, conrming our previous publication (Fornai et al.,
2008a,b). In Supplementary Fig. 1, presented are the results of the
motor function tests (Rotarod, PaGe and stride length) and the
survival time. In this experimental context, the increased neuron
number and the decreased vacuolization in the spinal cord were
conrmed here as shown in the representative Supplementary Figs.
2 and 3.
The brainstem nuclei have been roughly divided in oromotor
nuclei and oculomotor nuclei considering the muscles that they
innervate and the extent of their degeneration in ALS.
Oromotor nuclei
These nuclei innervate and control the muscles of mastication as
well as facial and lingual muscles during feeding, grooming, oral
reexes, swallowing, and respiration. We have included in this section
also the dorsal motor nucleus of the vagus and the nucleus ambiguus.
All these nuclei degenerate in ALS.

377

Hypoglossal nucleus (XII) and dorsal motor nucleus of the vagus (X)
The hypoglossal nucleus innervates the tongue, and it is a key
motor nucleus that is affected in human ALS. Few years ago in a
different SOD1 ALS mouse model carrying the G86R mutation,
Nimchinsky and coworkers (2000) found that the hypoglossal
nucleus is spared until end stage. In the present report, we showed
a massive cell loss (Figs. 1C and J) in the G93A mice hypoglossal
nucleus compared to the WT (Figs. 1A, B, and J), and we found that
this loss was attenuated following lithium treatment (Figs. 1D and J).
The effects of lithium were conrmed with NeuN immunostaining
that showed an increased number of large, well-shaped NeuN+ cells
(Fig. 2C) compared to the saline-treated mouse (Fig. 2B). Lithium also
decreased the vacuolization of the cell body (Figs. 1EH).
Most of the cells within the hypoglossal nucleus are MN, although
the nucleus contains also interneurons (Paxinos, 1997). Immunocytochemical localization of the synthesizing enzyme ChAT indicates a
decrease in the ChAT+ neurons in the G93A mouse (Fig. 3C), which
was attenuated following lithium administration (Fig. 3D) in
agreement with the NeuN immunostaining (Fig. 2). We also measured
the number of ChAT+ boutons within the hypoglossal nucleus as
shown in Fig. 4 and Table 1. We found a profound decrease of ChAT+
boutons in the G93A mice (Fig. 4C), which was reduced by lithium

Fig. 9. Representative cresyl violet- and H&E-stained sections through the trigeminal nuclei in WT saline- and lithium-treated (A, B) and in G93A saline- and lithium-treated (C, D)
mice. The MN count (F) showed a decrease in neuron amount in the G93A saline (C) compared to WT (A, B) and the partial recovery following lithium treatment (D). Values are
mean SEM (P 0.001 vs. WT sal.; #P 0.05 vs. G93A sal, WT sal, and WT li, F-value = 15.912). The high H&E magnication shows the recovery in neuron structure following
lithium treatment. The circles indicate the trigeminal area considered for the cell count, also showed with arrows in the image extracted from the mouse atlas of Paxinos and Franklin
(2004, E). Scale bars = 41.66 m (H&E, high magnication) and 447.76 m (cresyl violet, low magnication).

378

M. Ferrucci et al. / Neurobiology of Disease 37 (2010) 370383

Fig. 10. Neurons and astrocytes in the trigeminal nucleus area (5N) in G93A saline-treated (A) and lithium-treated (B) mice. The representative double immunouorescence of NeuN
and GFAP reveals the presence of astrocytes in trigeminal nucleus of the G93A mice and the neuronal loss (NeuN) in the G93A saline-treated mouse (A) and recovery after lithium
treatment (B). The square dotted shows the trigeminal area magnied in panels D (saline) and E (lithium) compared to WT (C). Scale bar = 250.31 m (A, B) and 93.75 m (CE).

(Fig. 4D) compared with saline. Most of ChAT+ boutons colocalize

with the presynaptic marker synaptotagmine I (Fig. 5). This result
indicates that ChAT+ boutons we counted are localized in the nerve
endings.
In contrast, in the dorsal motor nucleus of the vagus (X), lithium
treatment was unsuccessful. In fact, the nucleus showed a dramatic
decrease in neuron number in the G93A mouse compared with WT
(Figs. 6A, C, and F) while lithium did not change these values (Figs. 6C,
D, and F).
Facial nucleus (VII)
The facial nucleus degenerates in ALS. Most of the neurons in this
nucleus are MN that control and project to the facial musculature. In
agreement with previous reports (Nimchinsky et al., 2000; Zang et al.,
2004) compared with WT littermates (Figs. 7A, B, F) in the G93A
mouse, we observed an intense tissue vacuolization and a massive
neuronal loss, as indicated by cresyl violet and H&E staining (Fig. 7C),
and a statistically signicant protection following lithium treatment
(Figs. 7C, D, and F). While preserving the number of neurons, lithium
was not able to decrease the extent of gliosis as suggested by the nonquantitative NeuN/GFAP immunostaining in Figs. 8B and D (saline)
compared with Figs. 8C and E (lithium).
Trigeminal motor nucleus (V)
It is a voluminous nucleus which supplies the mastication muscles.
This nucleus is known to degenerate in ALS. Again we found a
signicant decrease in neuron number in the G93A mouse compared
with the WT (Figs. 9A, C, F) which was occluded by lithium (Figs. 9C,
D, and F). The non-quantitative NeuN/GFAP staining was in line with
total counts showing a higher NeuN+ in the area and within the
nucleus, while the GFAP positivity was comparable in saline- (Figs.
10A and D) and lithium-treated mice (Figs. 10B and E).

Nucleus ambiguus (IX-X)

This nucleus by the Merriam-Webster's Medical Dictionary
denition (Merriam-Webster Inc., 2002) is an elongated nucleus in
the medulla oblongata that is a continuation of a group of cells in the
ventral horn of the spinal cord and gives rise to the motor bers of the
glossopharyngeal (IX), and vagus (X), supplying striated muscle of the
larynx and pharynx. Nucleus ambiguus degenerates in human ALS. In
the G86R ALS mouse model, Nimchinsky et al. (2000) did not nd any
visible neuronal loss, while Zang et al. (2004), by using magnetic
resonance imaging coupled with histological analyses, observed a
signicant neuronal loss and intense vacuolization in the nucleus
ambiguus of 120 days old G93A male mice.
In line with this, in the G93A mouse, we found a signicant decrease
in the nucleus neuron number compared to WT (Figs. 11A, C, and J) and
a signicant protection after lithium (Figs. 11C, D, and J). This effect
was accompanied by a decrease in intracellular vacuolization (see
Figs. 11EH). The NeuN/GFAP immunostaining following lithium
showed a decrease in reactive gliosis surrounding this nucleus, as
evident in Figs. 12B and D (saline) compared with 12C, 12E (lithium).
Oculomotor nuclei
The abducens (VI), trochlear (IV), and oculomotor (III) are all
nuclei that control the eye movements, and it seems that they are not
affected in human ALS. The oculomotor nucleus has been analyzed in
the G86R model by Nimchinsky et al. (2000). These authors found a
substantial preservation of the nucleus, which was partly affected
only in one case and associated with a very severe disease stage.
Abducens (VI) and trochlear (IV) nuclei
The abducens nerve innervates the lateral rectus muscle of the eye.
Neuronal counts following cresyl violet staining did not reveal any

M. Ferrucci et al. / Neurobiology of Disease 37 (2010) 370383

decrease in G93A mouse compared with WT (Supplementary Figs. 4A,

4C, 4F) and any effect related to lithium treatment (Supplementary
Figs. 4C, 4D, and 4F). Confocal images of ChAT+/NeuN+ neurons
conrmed the substantial neuronal preservation in this nucleus
(Supplementary Figs. 5A, 5B, 5D, and 5E) bluntly revealed by the
ChAT single channel image showing the conservation of the ChAT+
cells and projections (Supplementary Figs. 5C and 5F). The trochlear
nucleus innervates the superior oblique muscle of the eye. Similarly to
the abducens nucleus, the neuronal count did not show any decrease
in the G93A mice compared with the WT (Supplementary Figs. 6A, 6C,
and 6F) and any effect following lithium treatment (Supplementary
Figs. 6C, 6D, and 6F). The NeuN/GFAP double immunostaining
revealed a generalized preservation of the NeuN+ cells in the nucleus
and surrounding area and a similar GFAP positivity in the lithium
treatment (Supplementary Figs. 7B and 7D) as compared with the
saline (Supplementary Figs. 7A and 7C).
Oculomotor nucleus (III)
The somatic motor component of the oculomotor nucleus plays a
major role in controlling the muscles responsible for the precise
movement of the eyes for visual tracking or xation on an object.

379

The nucleus contains densely packed neurons easily identiable with

cresyl violet staining. This nucleus displayed no obvious histopathological alterations and neuronal loss in saline-treated G93A mice
(Figs. 13A, C, and F) and any effect following lithium administration
as shown by the cresyl violet and H&E staining (Figs. 13C, D, and F).
Data are supported by the NeuN/GFAP representative immunostaining of a saline (Figs. 14A and C) and a lithium-treated mouse (Figs.
14B and D).
Discussion
In the present paper, we analyzed all brainstem motor nuclei in the
G93A mouse showing the reliability of this experimental model to
reproduce the bulbar involvement of ALS. Moreover, by proting of
this experimental setting, we characterized the effects of lithium in
counteracting the loss of brainstem motor neurons thus extending
previous ndings obtained in the spinal cord.
In fact, ALS is characterized by the degeneration of motor neurons
in the spinal cord, brainstem, and motor cortex (Rowland and
Shneider, 2001). Degeneration of MN in the brainstem produces
bulbar symptoms, such as dysphagia and dysarthria, which can be

Fig. 11. Representative cresyl violet- and H&E-stained sections through the ambiguus nucleus in WT saline- and lithium-treated (A, B) and in G93A saline- and lithium-treated (C, D)
mice. The MN count (J) showed a decrease in neuron amount in the G93A saline-treated mice (C) compared to WT (A, B) and the partial recovery following lithium (D). Values are
mean SEM (P 0.001 vs. WT sal.; #P 0.05 vs. G93A sal, WT sal, and WT li, F-value = 15.953). The circle indicates the area of the nucleus considered for the cell count also showed
with the arrow in the image extracted from the mouse atlas of Paxinos and Franklin (2004, I). (EH) Representative high magnication images of H&E-stained neurons of the nucleus
ambiguus in the WT saline- and lithium-treated (E, F) and G93A saline- and lithium-treated mice (G, H). Note the recovery in the neuronal structure in the lithium-treated compared
with saline-treated G93A mouse. Scale bars = 400 m (AD) and 22.72 m (EH).

380

M. Ferrucci et al. / Neurobiology of Disease 37 (2010) 370383

Fig. 12. Neurons and astrocytes in the ambiguus nucleus (Amb) in WT (A) and in G93A saline-treated (B) and lithium-treated (C) mice. The double immunouorescence of NeuN and
GFAP reveals the presence of reactive astrocytes in the nucleus of the saline-treated mouse (B) and their decrease following lithium (C) compared to WT (A). The NeuN indicates the
neuronal loss in the G93A saline (B) and the recovery after lithium (C) compared to WT (A). The G93A ambiguus nucleus region is magnied in panels D (saline) and E (lithium).
Scale bars = 175 m.

observed at disease onset in up to 30% of the patients. On the other

hand, almost all patients demonstrate some kind of bulbar involvement at later stage of the disease (Oliver, 1996). The involvement of
the brainstem is, with the progression of the disease, more and more
disturbing and results in a drop of quality of life and life expectancy
(Bourke et al., 2004). For these reasons, it is mandatory to study the
brainstem involvement at experimental level. The G93A mouse model
was used extensively to study the degeneration of the spinal cord and
sometimes it was questioned since it leads to a quick disease course
which can erase the subtle changes inherent to benecial or
detrimental effects of drugs under testing. Moreover, the apparent
lack of cortical degeneration makes this model incomplete to
understand ALS; nonetheless, previous evidence already suggested
the occurrence of brainstem degeneration reminiscent of ALS in the
G93A model.
Despite the severity of the brainstem involvement in human ALS,
only a few studies have evaluated the neuropathology of the
brainstem, taking into consideration, in general, the nuclei trigeminus
(V), facial (VII), and hypoglossal (XII) as degenerating nuclei and
oculomotor as a non-degenerating nucleus (Nimchinsky et al., 2000;
Angenstein et al., 2004; Zang et al., 2004; Niessen et al., 2006; Bucher
et al., 2007). In the present study, following a cell count which
encompassed the entire population of motor neurons of the
brainstem, we demonstrated that in the G93A mouse all branchial
motor nuclei (motor trigeminal, facial, ambiguus) and the hypoglossal
nucleus were signicantly degenerated. On the other hand, oculomotor nuclei (III, IV, and VI) were spared. These ndings are consistent
with what reported in humans thus providing a solid experimental
setting to analyze the so-called bulbar involvement of ALS. This
formal description adds on previous important ndings which
converge to validate the G93A mouse model to study the brainstem
degeneration in ALS. For instance, in recent studies, Lever et al.
(2009a,b) have demonstrated a correlation between neurodegeneration of the trigeminal and hypoglossal nuclei and the development of

dysphagia and oral dysfunction in the G93A mouse. They found that
oral dysfunction (i.e. lick and mastication rates) could be detected as
early as 60 days of age, suggesting that oral dysphagia could be one of
the earliest clinical symptoms in the G93A mouse. This extends the
usefulness of the G93A mouse as an animal model for dysphagia in
ALS. In human ALS patients difculties in swallowing occur early in
the course of the disease and nearly all patients ultimately develop
some kind of impairments in swallowing food, liquid or saliva (Hillel
and Miller, 1989; Kawai et al., 2003).
Besides the description of these nuclei at the end of the natural
course of the disease, we assessed the protective effects of lithium in
modifying tissue vacuolization, neuronal cell loss, and reducing the
loss of recurrent axon boutons. Chronic lithium treatment reduced the
massive motor neuron loss in all but one, the dorsal motor nucleus of
the vagus (X), motor nuclei. On a general basis, this specic lack of
effects of lithium might rely on the peculiarity of visceral compared
with somatic motor neurons. This may be the consequence of either
increased susceptibility to the neurotoxic effects of the G93A
mutation or a refractoriness to the protective effects of lithium. In
line with the rst hypothesis, neurons of the dorsal motor nucleus of
vagus are more vulnerable to axonal injuries compared with somatic
motor nuclei (Lewis et al., 1972; Aldskogius et al., 1980; Yu, 2002). In
line with this, axotomy induces a stronger expression of markers of
neuronal in the dorsal motor nucleus of vagus compared with other
somatic motor nuclei (Chang et al., 2001; Yu, 2002; Wei et al., 2008),
and axotomy of the vagus nerve induces cell death within the visceral
(dorsal motor nucleus) but not somatic (nucleus ambiguus) nuclear
component (Yu, 1994).
Lending substance to the hypothesis which relates neuronal
vulnerability to axonal length, we have to consider that the G93A
mutation leads to a more marked loss of those neurons possessing
longer axons (lumbar compared with cervical spinal cord). This is in
line with the unusual length of axons from vagal visceral neurons
compared with axons from brainstem somatic motor neurons. In

M. Ferrucci et al. / Neurobiology of Disease 37 (2010) 370383

381

Fig. 13. Representative cresyl violet- and H&E-stained sections of the oculomotor nuclei in WT saline- and lithium-treated (A, B) and in G93A saline- and lithium-treated (C, D) mice.
Representative cresyl violet- and H&E-stained images do not show any alteration in neuronal density either neuronal structure in the nucleus of the transgenic mice compared with
WT. The MN count (F) showed no difference in neuron amount in the G93A saline- and lithium-treated mice compared to the WT. The circles indicate the oculomotor area
considered for the cell count, also showed with arrows in the image extracted from mouse atlas of Paxinos and Franklin (2004, E). Scale bars = 38.46 m (H&E, high magnication)
and 551.18 m (cresyl violet, low magnication).

addition, the poor regeneration of the visceral compared with somatic

motor system following neuronal damage needs to be considered (Yu,
1997). In fact, previous studies demonstrated that the dorsal motor
nucleus of vagus possesses reduced synaptic reorganization and
neuronal plasticity compared with nucleus ambiguus (Lan et al., 2000,
2004). Similarly, neuroprotective strategies following neuronal injury
produce a reduced effect on this visceral nucleus compared with
somatic motor nuclei (Wei et al., 2008).
Similarly to what we have observed in the spinal cord, also in the
brainstem, lithium treatment recovers MN-like cells in all nuclei and
ameliorates both the extracellular and intracellular tissue vacuolization (see Figs. 1EH and Figs. 11EH) a hallmark of ALS (Jaarsma et al.,
2000; Bucher et al., 2007). Qualitatively we observed an increase in
NeuN-positive and ChAT-positive neurons in agreement with the
Nissl counts as well as an improvement in their morphology and
shape (compare Fig. 2B with Fig. 2C). We also examined the extent of
gliosis in the saline- and lithium-treated mice. While in spinal cord we
observed a lithium-mediated decrease in reactive gliosis, in the
brainstem the qualitative analyses shows a clear decrease in GFAP
immunostaining at the level of the nucleus ambiguus in the lithiumtreated mice (Figs. 12D, E) whereas in the other nuclei the decrease in
GFAP staining is less clear or not present at all. We do not possess a
clear explanation to address such a discrepancy. In fact, while a

decrease in GFAP staining conrms the data obtained in the spinal

cord, a lack of effect represents a novelty. It is likely that nucleusspecic glianeuron interactions may underlie such a difference.
Further studies aimed at addressing the specic role of glia in each
brainstem nucleus are needed to solve this point.
In a previous study (Fornai et al., 2008a,b), we have described
lithium efcacy in delaying the disease course in terms of survival rate
and motor performances while improving neuronal survival, stimulating mitochondriogenesis, limiting reactive gliosis, and activating
autophagy. Such a symptomatic effect was conrmed here as reported
in the supporting Supplementary Figs. 1, 2, and 3. Lithium's ability,
alone or in combination, to delay disease progression in the G93A ALS
mouse model has been also described by others. For instance, Shin
et al. (2007) using lithium in combination with the antioxidant drug
Neu2000 linked its neuroprotective ability to the inhibition of the Fasmediated apoptosis. While Feng et al. (2008), using lithium in
combination with valproic acid, linked lithium efcacy to the
potentiation of the inhibition of the glycogen synthase kinase-3
(GSK-3) activity. These ndings were recently translated in two pilot
studies where lithium was found as a potential candidate to slow
progression of ALS in human patients (Fornai et al., 2008a; Payne,
2009; for a review, see Eisen, 2009). It is likely that the main
mechanism of action which underlies the effects of lithium on motor

382

M. Ferrucci et al. / Neurobiology of Disease 37 (2010) 370383

with a deletion of the hypoxia-response element in the VEGF

promoter develop a severe adult-onset muscle weakness due to
lower MN degeneration (Oosthuyse et al., 2001). The benecial effects
of lithium in counteracting motor impairment caused by a disease of
the spinal cord was demonstrated to be associated with (i) increased
motor neurons (Shin et al., 2007; Fornai et al., 2008a,b; Feng et al.,
2008); (ii) sprouting of the lateral motor tracts such as the
corticospinal (Dill et al., 2008) and rubrospinal (Yick et al., 2004)
which are both involved also in ALS; and (iii) neuronogenesis in the
spinal cord (Su et al., 2007; 2009; Fornai et al., 2008a,b).
We are presently investigating the occurrence of neuronogenesis
in the brainstem of lithium-treated G93A and we possess at present
fascinating data which cover the facial nucleus where a widespread
Bromodeoxyuridine+/calbindind D28K+ immunostaining is observed only in lithium-treated G93A mice (data not shown).
Despite these open issues, the present data extend the neuroprotective effects of lithium to brainstem MN while offering a solid
evidence on the brainstem involvement in the G93A mouse model.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.nbd.2009.10.017.
References

Fig. 14. Neurons and astrocytes in the oculomotor nucleus (3N) in G93A saline-treated
(A, C) and lithium-treated (B, D) mice. Representative double immunouorescence
revealed a comparable NeuN and GFAP staining in the oculomotor nucleus of the salinetreated (A, C) and the lithium-treated (B, D) mice. The dotted square represents the
oculomotor region magnied in panels C (saline) and D (lithium). Aq, aqueduct. Scale
bars = 375 m (A, B) and 125 m (C, D).

neurons is based on the activation of the autophagy pathway which is

decient in ALS independently of the specic models (Fornai et al.,
2008a,b; Laird et al., 2008; Pasquali et al., 2009). In fact, in a recent
study, Pizzasegola et al. (2009), using experimental conditions in
which the pharmacological effects of lithium on autophagy were not
achieved, failed to document any effect on motor neurons.
Moreover, the positive effects of lithium toward MN goes behind
ALS and it has been shown also in relation to axon sprouting in the
caudal spinal cord and in promoting locomotor functional recovery
after thoracic spinal cord transaction or contusion injuries (Dill et al.,
2008). Accordingly the therapeutic potentials of lithium either as
autophagy inducer (Sarkar et al., 2005) or GSK-3 regulator (Aghdam
and Barger, 2007) are increasingly recognized as a possible therapeutical strategy for neurological disorders including Alzheimer's
disease, Parkinson's disease, and ischemic brain injury (Meijer et al.,
2004; Engel et al., 2008).
Autophagy is a key cellular pathway for the clearance of aggregates
prone proteins and malfunctioning organelles (Rubinsztein et al.,
2005; Ventruti and Cuervo 2007), and it is becoming a central
pathway in the most common neurodegenerative diseases including
ALS (Nedelsky et al., 2008; Winslow and Rubinsztein, 2008).
Finally, in addition to stimulating autophagy and/or inactivating
GSK-3, lithium may exert its protective action in ALS via additional
pathways, such as the upregulation of the endothelial or astrocytic
vascular endothelial growth factor (VEGF; Silva et al., 2007; Guo et al.,
2009). VEGF lentiviral delivery prolonged survival and protected
neurons in an experimental animal model of ALS (Azzouz et al., 2004),
while a bloodspinal cord barrier compromise has been recently
describe as one of the early symptoms of ALS (Garbuzova-Davis et al.,
2007; Zhong et al., 2008). These data indicate that a decreased
expression or an abnormal VEGF regulation may predispose to
pathological disorders including MN degeneration. Indeed, mice

Aghdam, S.Y., Barger, S.W., 2007. Glycogen synthase kinase-3 in neurodegeneration and
neuroprotection: lessons from lithium. Curr. Alzheimer Res. 4, 2131.
Aldskogius, H., Barron, K.D., Regal, R., 1980. Axon reaction in dorsal motor vagal and
hypoglossal neurons of the adult rat. Light microscopy and RNA-cytochemistry.
J. Comp. Neurol. 193, 165177.
Angenstein, F., Niessen, H.G., Goldschmidt, J., Vielhaber, S., Ludolph, A.C., Scheich, H.,
2004. Age-dependent changes in MRI of motor brain stem nuclei in a mouse model
of ALS. Neuroreport 15, 22712274.
Azzouz, M., Ralph, G.S., Storkebaum, E., Walmsley, L.E., Mitrophanous, K.A., Kingsman,
S.M., Carmeliet, P., Mazarakis, N.D., 2004. VEGF delivery with retrogradely
transported lentivector prolongs survival in a mouse ALS model. Nature 429,
413417.
Bourke, S.C., McColl, E., Shaw, P.J., Gibson, G.J., 2004. Validation of quality of life
instruments in ALS. Amyotroph. Lateral. Scler. Other Mot. Neuron Disord. 5, 5560.
Bucher, S., Braunstein, K.E., Niessen, H.G., Kaulisch, T., Neumaier, M., Boeckers, T.M.,
Stiller, D., Ludolph, A.C., 2007. Vacuolization correlates with spin-spin relaxation
time in motor brainstem nuclei and behavioural tests in the transgenic G93A-SOD1
mouse model of ALS. Eur. J. Neurosci. 26, 18951901.
Chang, H.M., Lue, J.H., Wen, C.Y., Shieh, J.Y., 2001. Axotomy along with hypoxia
enhances the neuronal NADPH-d/NOS expression in lower brain stem motor
neurons of adult rats. Exp. Neurol. 171, 116126.
Cleveland, D.W., Rothstein, J.D., 2001. From Charcot to Lou Gehrig: deciphering
selective motor neuron death in ALS. Nat. Rev. Neurosci. 2, 806819.
Dill, J., Wang, H., Zhou, F., Li, S., 2008. Inactivation of glycogen synthase kinase 3
promotes axonal growth and recovery in the CNS. J. Neurosci. 28, 89148928.
Eisen, A., 2009. Amyotrophic lateral sclerosis: a 40-year personal perspective. J. Clin.
Neurosci. 16, 505512.
Engel, T., Goni-Oliver, P., Gomez de Barreda, E., Lucas, J.J., Hernandez, F., Avila, J., 2008.
Lithium, a potential protective drug in Alzheimer's disease. Neurodegener. Dis. 5,
247249.
Feng, H.L., Leng, Y., Ma, C.H., Zhang, J., Ren, M., Chuang, D.M., 2008. Combined lithium
and valproate treatment delays disease onset, reduces neurological decits and
prolongs survival in an amyotrophic lateral sclerosis mouse model. Neuroscience
155, 567572.
Fernagut, P.O., Diguet, E., Labattu, B., Tison, F., 2002. A simple method to measure stride
length as an index of nigrostriatal dysfunction in mice. J. Neurosci. Methods 113,
123130.
Fornai, F., Longone, P., Cafaro, L., Kastsiuchenka, O., Ferrucci, M., Manca, M.L., Lazzeri, G.,
Spalloni, A., Bellio, N., Lenzi, P., Modugno, N., Siciliano, G., Isidoro, C., Murri, L.,
Ruggieri, S., Paparelli, A., 2008a. Lithium delays progression of amyotrophic lateral
sclerosis. Proc. Natl. Acad. Sci. U. S .A. 105, 20522057.
Fornai, F., Longone, P., Ferrucci, M., Lenzi, P., Isidoro, C., Ruggieri, S., Paparelli, A., 2008b.
Autophagy and amyotrophic lateral sclerosis: the multiple roles of lithium.
Autophagy 4, 527530.
Garbuzova-Davis, S., Haller, E., Saporta, S., Kolomey, I., Nicosia, S.V., Sanberg, P.R., 2007.
Ultrastructure of bloodbrain barrier and bloodspinal cord barrier in SOD1 mice
modeling ALS. Brain Res. 1157, 126137.
Gizzi, M., DiRocco, A., Sivak, M., Cohen, B., 1992. Ocular motor function in motor neuron
disease. Neurology 42, 10371046.
Guo, S., Arai, K., Stins, M.F., Chuang, D.M., Lo, E.H., 2009. Lithium upregulates vascular
endothelial growth factor in brain endothelial cells and astrocytes. Stroke 40,
652655.

M. Ferrucci et al. / Neurobiology of Disease 37 (2010) 370383

Gurney, M.E., Pu, H., Chiu, A.Y., Dal Canto, M.C., Polchow, C.Y., Alexander, D.D., Caliendo,
J., Hentati, A., Kwon, Y.W., Deng, H.X., 1994. Motor neuron degeneration in mice
that express a human Cu,Zn superoxide dismutase mutation. Science 264,
17721775.
Gurney, M.E., 1997. Transgenic animal models of familial amyotrophic lateral sclerosis.
J. Neurol. 244, 1520.
Haenggeli, C., Kato, A.C., 2002. Differential vulnerability of cranial motoneurons in
mouse models with motor neuron degeneration. Neurosci. Lett. 335, 3943.
Hartmann, H.A., McMahon, S., Sun, D.Y., Abbs, J.H., Uemura, E., 1989. Neuronal RNA in
nucleus ambiguus and nucleus hypoglossus of patients with amyotrophic lateral
sclerosis. J. Neuropathol. Exp. Neurol. 48, 669673.
Hillel, A.D., Miller, R., 1989. Bulbar amyotrophic lateral sclerosis: patterns of
progression and clinical management. Head Neck 11, 5159.
Jaarsma, D., Haasdijk, E.D., Grashorn, J.A., Hawkins, R., van Duijn, W., Verspaget, H.W.,
London, J., Holstege, J.C., 2000. Human Cu/Zn superoxide dismutase (SOD1)
overexpression in mice causes mitochondrial vacuolization, axonal degeneration,
and premature motoneuron death and accelerates motoneuron disease in mice
expressing a familial amyotrophic lateral sclerosis mutant SOD1. Neurobiol. Dis. 7,
623643.
Kawai, S., Tsukuda, M., Mochimatsu, I., Enomoto, H., Kagesato, Y., Hirose, H., Kuroiwa, Y.,
Suzuki, Y., 2003. A study of the early stage of dysphagia in amyotrophic lateral
sclerosis. Dysphagia 18, 18.
Khnlein, P., Gdynia, H.J., Sperfeld, A.D., Lindner-Peghar, B., Ludolph, A.C., Prosiegel,
M., Riecker, A., 2008. Diagnosis and treatment of bulbar symptoms in amyotrophic
lateral sclerosis. Nat. Clin. Pract. Neurol. 4, 366374.
Kusaka, H., Imai, T., Hashimoto, S., Yamamoto, T., Maya, K., Yamasaki, M., 1988.
Ultrastructural study of chromatolytic neurons in an adult-onset sporadic case of
amyotrophic lateral sclerosis. Acta Neuropathol. 75, 523528.
Laird, F.M., Farah, M.H., Ackerley, S., Hoke, A., Maragakis, N., Rothstein, J.D., Grifn, J.,
Price, D.L., Martin, L.J., Wong, P.C., 2008. Motor neuron disease occurring in a
mutant dynactin mouse model is characterized by defects in vesicular trafcking.
J. Neurosci. 28, 19972005.
Lan, C.T., Liu, J.C., Hsu, J.C., Ling, E.A., 2000. Different astroglial reaction between the
vagal dorsal motor nucleus and nucleus ambiguus following vagal-hypoglossal
nerve anastomosis in cats. Brain Res. 881, 222226.
Lan, C.T., Hsu, J.C., Tseng, C.Y., Chang, C.N., Ling, E.A., 2004. Synaptic remodeling in the
nucleus ambiguus following vagal-hypoglossal nerve anastomosis in the cat. Brain
Res. 1013, 6073.
Lever, T.E., Gorsek, A., Cox, K.T., O'Brien, K.F., Capra, N.F., Hough, M.S., Murashov, A.K.,
2009a. An animal model of oral dysphagia in amyotrophic lateral sclerosis.
Dysphagia 24, 180195.
Lever, T.E., Simon, E., Cox, K.T, Capra, N.F., O'Brien, K.F., Hough, M.S., Murashov, A.K.,
2009b. A mouse model of pharyngeal dysphagia in amyotrophic lateral sclerosis.
Dysphagia. doi:10.1007/s00455-009-9232-1.
Lewis, P.R., Jones, P.B., Breathnach, S.M., Navaratnam, V., 1972. Regenerative capacity of
visceral preganglionic neurones. Nat. New Biol. 236, 181182.
Llad, J., Haenggeli, C., Pardo, A., Wong, V., Benson, L., Coccia, C., Rothstein, J.D., Shefner,
J.M., Maragakis, N.J., 2006. Degeneration of respiratory motor neurons in the SOD1
G93A transgenic rat model of ALS. Neurobiol. Dis. 21, 110118.
Meijer, L., Flajolet, M., Greengard, P., 2004. Pharmacological inhibitors of glycogen
synthase kinase 3. Trends Pharmacol. Sci. 25, 471480.
Merriam-Webster Inc., 2002. Merriam-Webster Collegiate Dictionary, Eleventh Edition.
Encyclopedia Britannica Company.
Nedelsky, N.B., Todd, P.K., Taylor, J.P., 2008. Autophagy and the ubiquitin-proteasome
system: collaborators in neuroprotection. Biochim. Biophys. Acta 1782, 691699.
Newbery, H.J., Abbott, C.M., 2001. Of mice, men and motor neurons. Trends Genet. 17,
S2S6.
Niessen, H.G., Angenstein, F., Sander, K., Kunz, W.S., Teuchert, M., Ludolph, A.C., Heinze,
H.J., Scheich, H., Vielhaber, S., 2006. In vivo quantication of spinal and bulbar
motor neuron degeneration in the G93A-SOD1 transgenic mouse model of ALS by
T2 relaxation time and apparent diffusion coefcient. Exp. Neurol. 201, 293300.
Nimchinsky, E.A., Young, W.G., Yeung, G., Shah, R.A., Gordon, J.W., Bloom, F.E., Morrison,
J.H., Hof, P.R., 2000. Differential vulnerability of oculomotor, facial, and hypoglossal
nuclei in G86R superoxide dismutase transgenic mice. J. Comp. Neurol. 416,
112125.
Okamoto, K., Hirai, S., Amari, M., Iizuka, T., Watanabe, M., Murakami, N., Takatama, M.,
1993. Oculomotor nuclear pathology in amyotrophic lateral sclerosis. Acta
Neuropathol. 85, 458462.
Oliver, D., 1996. The quality of care and symptom controlthe effects on the terminal
phase of ALS/MND. J. Neurol. Sci. 139, S134S136.
Oosthuyse, B., Moons, L., Storkebaum, E., Beck, H., Nuyens, D., Brusselmans, K., Van
Dorpe, J., Hellings, P., Gorselink, M., Heymans, S., Theilmeier, G., Dewerchin, M.,
Laudenbach, V., Vermylen, P., Raat, H., Acker, T., Vleminckx, V., Van Den Bosch, L.,
Cashman, N., Fujisawa, H., Drost, M.R., Sciot, R., Bruyninckx, F., Hicklin, D.J., Ince, C.,
Gressens, P., Lupu, F., Plate, K.H., Robberecht, W., Herbert, J.M., Collen, D., Carmeliet,
P., 2001. Deletion of the hypoxia-response element in the vascular endothelial

383

growth factor promoter causes motor neuron degeneration. Nat. Genet. 28,
131138.
Pasquali, L., Longone, P., Isidoro, C., Ruggieri, S., Paparelli, A., Fornai, F., 2009. Autophagy,
lithium, and amyotrophic lateral sclerosis. Muscle Nerve 40, 173194.
Paxinos, G., 1997. The rat nervous system, 2nd Edition. Academic Press, San Diego
(USA).
Paxinos, G., Franklin, K.B.J., 2004. The Mouse Brain in Stereotaxic Coordinates, 2nd
Edition. Academic Press, San Diego (USA).
Payne, A.G., 2009. Experimental regimen targeting the ependyma slows disease
progression in four patients with amyotrophic lateral sclerosis. Med. Hypotheses
72, 548550.
Pizzasegola, C., Caron, I., Baleno, C., Ronchi, A., Minoia, C., Carr, M.T., Bendotti, C., 2009.
Treatment with lithium carbonate does not improve disease progression in two
different strains of SOD1 mutant mice. Amyotroph. Lateral. Scler. 10, 221228.
Ripps, M.E., Huntley, G.W., Hof, P.R., Morrison, J.H., Gordon, J.W., 1995. Transgenic mice
expressing an altered murine superoxide dismutase gene provide an animal model
of amyotrophic lateral sclerosis. Proc. Natl. Acad. Sci. U. S. A. 92, 689693.
Rowland, L.P., Shneider, N.A., 2001. Amyotrophic lateral sclerosis. N. Engl. J. Med. 344,
16881700.
Rubinsztein, D.C., DiFiglia, M., Heintz, N., Nixon, R.A., Qin, Z.H., Ravikumar, B., Stefanis,
L., Tolkovsky, A., 2005. Autophagy and its possible roles in nervous system diseases,
damage and repair. Autophagy 1, 1122.
Sarkar, S., Floto, R.A., Berger, Z., Imarisio, S., Cordenier, A., Pasco, M., Cook, L.J.,
Rubinsztein, D.C., 2005. Lithium induces autophagy by inhibiting inositol monophosphatase. J. Cell Biol. 170, 11011111.
Shin, J.H., Cho, S.I., Lim, H.R., Lee, J.K., Lee, Y.A., Noh, J.S., Joo, I.S., Kim, K.W., Gwag, B.J.,
2007. Concurrent administration of Neu2000 and lithium produces marked
improvement of motor neuron survival, motor function, and mortality in a
mouse model of amyotrophic lateral sclerosis. Mol. Pharmacol. 71, 965975.
Silva, R., Martins, L., Longatto-Filho, A., Almeida, O.F., Sousa, N., 2007. Lithium prevents
stress-induced reduction of vascular endothelium growth factor levels. Neurosci.
Lett. 429, 3338.
Spalloni, A., Geracitano, R., Berretta, N., Sgobio, C., Bernardi, G., Mercuri, N.B., Longone,
P., Ammassari-Teule, M., 2006. Molecular and synaptic changes in the hippocampus
underlying superior spatial abilities in pre-symptomatic G93A+/+ mice overexpressing the human Cu/Zn superoxide dismutase (Gly93ALA) mutation. Exp.
Neurol. 197, 505514.
Su, H., Chu, T.H., Wu, W., 2007. Lithium enhances proliferation and neuronal
differentiation of neural progenitor cells in vitro and after transplantation into
the adult rat spinal cord. Exp. Neurol. 206, 296307.
Su, H., Zhang, W., Guo, J., Guo, A., Yuan, Q., Wu, W.., 2009. Lithium enhances the
neuronal differentiation of neural progenitor cells in vitro and after transplantation
into the avulsed ventral horn of adult rats through the secretion of brain-derived
neurotrophic factor. J. Neurochem. 108, 13851398.
Ventruti, A., Cuervo, A.M., 2007. Autophagy and neurodegeneration. Curr. Neurol.
Neurosci. Rep. 7, 443451.
Yick, L.W., So, K.F., Cheung, P.T., Wu, W.T., 2004. Lithium chloride reinforces the
regeneration-promoting effect of chondroitinase ABC on rubrospinal neurons after
spinal cord injury. J. Neurotrauma 21, 932943.
Young, W., 2009. Review of Lithium Effects on Brain and Blood. Cell Transplant.
doi:10.3727/096368909X471251.
Yu, W.H., 1994. Nitric oxide synthase in motor neurons after axotomy. J. Histochem.
Cytochem. 42, 451457.
Yu, W.H., 1997. Regulation of nitric oxide synthase expression in motoneurons
following nerve injury. Dev. Neurosci. 19, 247254.
Yu, W.H., 2002. Spatial and temporal correlation of nitric oxide synthase expression
with CuZn-superoxide dismutase reduction in motor neurons following axotomy.
Ann. N. Y. Acad. Sci. 962, 111121.
Wei, I.H., Huang, C.C., Tseng, C.Y., Chang, H.M., Tu, H.C., Tsai, M.H., Wen, C.Y., Shieh, J.Y.,
2008. Mild hypoxic preconditioning attenuates injury-induced NADPH-d/nNOS
expression in brainstem motor neurons of adult rats. J. Chem. Neuroanat. 35,
123132.
Weydt, P., Hong, S.Y., Kliot, M., Moller, T., 2003. Assessing disease onset and progression
in the SOD1 mouse model of ALS. Neuroreport 14, 10511054.
Wijesekera, L.C., Leigh, P.N., 2009. Amyotrophic lateral sclerosis. Orphanet J. Rare Dis.
4, 3.
Winslow, A.R., Rubinsztein, D.C., 2008. Autophagy in neurodegeneration and
development. Biochim. Biophys. Acta 1782, 723729.
Zang, D.W., Yang, Q., Wang, H.X., Egan, G., Lopes, E.C., Cheema, S.S., 2004. Magnetic
resonance imaging reveals neuronal degeneration in the brainstem of the
superoxide dismutase 1 transgenic mouse model of amyotrophic lateral sclerosis.
Eur. J. Neurosci. 20, 17451751.
Zhong, Z., Deane, R., Ali, Z., Parisi, M., Shapovalov, Y., O'Banion, M.K., Stojanovic, K.,
Sagare, A., Boillee, S., Cleveland, D.W., Zlokovic, B.V., 2008. ALS-causing SOD1
mutants generate vascular changes prior to motor neuron degeneration. Nat.
Neurosci. 11, 420422.