вход по аккаунту


Benefit of vitamin E riluzole and gababapentin in a transgenic model of familial amyotrophic lateral sclerosis.

код для вставкиСкачать
Benefit of Vitamin E, N m l e , and
babapenun in a Transgenic Model of
~ d AmyotropfUc
Lateral sclerosis
Mark E. Gurney, PhD,*t Frank B. Cutting, MS,* Ping Zhai, MD, PhD,* Adam Doble, PhD,$
Charles P. Taylor, PhD,$ Paula K. Andrus, MS,t and Edward D. Hall, PhDt
Familial amyotrophic lateral sclerosis (FALS) has been linked in some families to dominant mutations of the SODl
gene encoding Cu,Zn superoxide dismutase (Cu,Zn SOD). We have used a transgenic model of FALS based on expression
of mutant human Cu,Zn SOD to explore the etiology and therapy of the genetic disease. Expression of mutant, but
not wild-type, human Cu,Zn SOD in mice places the brain and spinal cord under oxidative stress. This causes depletion
of vitamin E, rather than the typical age-dependent increase in vitamin E content as occurs in nontransgenic mice and
in mice expressing wild-type human Cu,Zn SOD. Dietary supplementation with vitamin E delays onset of clinical disease
and slows progression in the transgenic model but does not prolong survival. In contrast, two putative inhibitors of
the glutamatergic system, riluzole and gabapentin, prolong survival. However, riluzole did not delay disease onset. Thus,
there was clear separation of effects on onset, progression, and survival by the three therapeutics tested. This suggests
the hypothesis that oxidative damage produced by the expression of mutant Cu,Zn SOD causes slow or weak excitotoxicity that can be inhibited in part by altering glutamate release or biosynthesis presynaptically.
ME, Cutting FB, Zhai P, Doble A, Taylor CP, Andrus PK, Hall ED. Benefit of
vitamin E, riluzole, and gabapentin in a transgenic model of familial
amyotrophic lateral sclerosis. Ann Neurol 1996;39:147- 157
matergic neurotransmission, show that survival of ALS
patients is extended by 12% [12].
The familial form of ALS (FALS) has been linked
in some patients to dominantly inherited mutations in
the SOD1 gene encoding Cu,Zn superoxide dismutase
(Cu,Zn SOD) [13] and in others to variants in the
gene encoding the heavy subunit of the neurofilament
triplet protein [14]. So far, more than 20 different missense mutations of Cu,Zn SOD that substitute one
amino acid for another have been identified in FALS
families [15]. Studies of FALS patients with mutations
of SODl indicated that Cu,Zn SOD activity is decreased 20 to 50% [16, 171. This suggested initially
that disease was due to free-radica! damage resulting
from a structurally defective enzyme with reduced activity [16]. However, no deletions of the SODl gene
have been found in FALS families, which implies that
expression of the mutant protein is required for pathogenesis.
Rather than causing a loss of function, however,
Amyotrophic lateral sclerosis (ALS) is caused by the
degeneration of motor neurons in cortex, brainstem,
and spinal cord [I]. The disease is progressive and fatal,
with survival a median of 3 years after diagnosis [2].
The disease occurs in both sporadic and familial forms
whose clinical courses are highly similar. The etiology
of the sporadic disease has been linked to alterations
of glutamatergic neurotransmission [ 3 ] ,environmental
toxins [4],
autoimmunity [5], and metals [6]. Glutamate, the primary excitatory neurotransmitter in the
central nervous system, has been linked to neuronal
damage in acute illnesses such as stroke and head
trauma [7].Supporting the involvement of glutamate
in the pathogenesis of ALS ate elevated measurements
of glutamate in serum, cerebrospinal fluid, and brain
[S, 91, decreased high-affinity glutamate uptake by synaptosomes prepared from spinal cord and motor cortex
[lo], and decreased expression of the primarily glial
GLT-1 glutamate transporter [ 1 I]. In addition, clinical
trials with riluzole, a drug that incerferes with gluta-
From the *Department of Cell and Molecular Biology, Northwestern University Medical School, Chicago, IL; $Rh6ne-Poulenc
Rorer, Vitry-sur-Seine, France; SDepartment of Neurological and
Neurodegenerative Diseases, Parke-Davis Pharmaceutical Research
Division, Warner-Lambert Company, Ann Arbor, MI; and t Central Nervous System Diseases Research Unit, Pharmacia & Upjohn,
Inc, Kalamazoo, MI.
Received Aug 28, 1995, and in revised form Oct 20 and Nov 2.
Accepted for publication Nov 6, 1995.
Address correspondence to Dr Gurney, Central Nervous System
Diseases Research Unit, 725 1-209-505, Pharmacia & Upjohn, Inc,
301 Henrietta Street, Kalamazoo, MI 49001.
0 1996 by
the American Neurological Association
studies i n transgenic mice suggest the mutations of
C u , Z n SOD i n FALS patients cause a gain of function
that results in neuronal degeneration [ 181. So far, three
of four SODl mutations (glf’ -+ ala, glyR5+ arg, a n d
gly” -+ arg) found i n FALS kindreds has caused motor
neuron disease when expressed i n transgenic mice [ 18201. Mice develop a progressive, paralytic illness whose
general features resemble ALS in humans. Unlike the
human disease, the mouse model has prominent vacuolization of mitochondria a n d endoplasmic reticulum
[20, 211. T h e glf’ + ala and gly3’
arg mutations
have little effect on C u , Z n SOD enzymatic activity, so
expression of the mutant h u m a n SODl gene causes an
increase in brain superoxide dismutase activity [ 18,
201. Because transgenic mice expressing wild-type hum a n C u , Z n SOD with comparable elevation of brain
superoxide dismutase activity do n o t develop motor
neuron disease [IS, 20, 221 and, i n fact, s h o w enhanced resistance to different types of oxidative insults
[23, 241, disease is due to expression of the m u t a n t
protein and not t o elevation of superoxide dismutase
activity in brain [25].
W e undertook a n initial screen of potential therapeutics i n this model [26].We examined dietary supplementation with an antioxidant, vitamin E [27],a n d
treatment with two putative modulators of the glutamatergic system, riluzole and gabapentin [ 12, 281.
Should riluzole have therapeutic benefit in the transgenic model, the result would suggest that the sporadic
a n d familial forms of ALS share c o m m o n elements of
Materials and Methods
Transgenic Mice
The transgene used for expression of human Cu,Zn S O D
in mice consisted of an 1 1.7-kb genomic fragment that contained the entire human SOD1 gene [29], together with promoter and enhancer sequences required for transcription in
mice [30], as described previously [18]. The G l H / + line
has an expansion in transgene copy number of 40% over the
G 1 line (now designated GIL/+) described by Gurney and
colleagues (181. The line ha5 been deposited with the
Induced Mutant Resource operated by The Jackson Laboratory (Bar Harbor, ME) under the strain designation
BGSJL-TgN(SODl-G93A)1Gur. N29/+ mice [also deposited with The Jackson Laboratory under the strain designation TgN(SOD1)2Gur] contain a wild-type human SODl
transgene and express levels of human Cu,Zn SOD protein
comparable with the level of mutant Cu,Zn S O D protein expressed in GlH/+ mice. The transgenic mouse lines
were maintained as hemizygotes by mating transgenic males
with BGSJL F, hybrid females (The Jackson Laboratory).
Transgenic progeny were identified by an enzyme immunoassay specific for human Cu,Zn SOD [18]. Nontransgenic
(nonTg) littermates were used as controls for vitamin E determinations.
Transgenic mice were housed in microisolator cages
within a modified harrier facility and were seronegative for
148 Annals of Neurology
Vol 39
No 2
February 1996
mouse hepatitis virus, Sendai virus, mouse parvovirus, and
other common viral and bacterial pathogens. During the
course of these experiments, however, infection by Syphacia
obvelata was discovered in several GIHIS mice that were
not enrolled in treatment protocols. Consequently, all of the
mice in the room were treated for Syphacia obvelata with
four weekly spray treatments of aqueous ivermectin (0.9- 1.8
ing per cage) followed by treatment for 3 weeks with piperazine hydrochloride in their water (4 mg/ml, changed twice
weekly). Use of both agents to treat murine pinworms was
chosen to increase the possibility of eliminating the infection
[31]. Seven mice enrolled in treatment protocols died after
spray treatment with ivcrmectin and two more died during
the period of piperazine treatment (see Table 1).
Vitamin E Measurements in Brain, Spinal Cord,
and Liver
Samples for vitamin E determination consisted of pooled
brainstem and spinal cord, the remainder of the brain (midbrain, cerebellum, and cerebrum), and liver. Pathology in
GIH/+ transgenic mice is restricted to the brainstem and
spinal cord. The cerebellum and cerebrum are histologically
normal as is the liver [18, 21, 32, 331. After anesthetization
with Metofane (Pitman-Moore, Mundelein, IN), mice were
transcardially perfused with 0.13 M sodium phosphate buffer (pH 7.2) to wash blood from the brain and spinal cord.
The tissue samples were harvested, flash frozen in liquid nitrogen, and then stored at -70°C. Vitamin E was assayed
as described in detail elsewhere with only slight modifications
Treatment Protocols
Animals were fed a basal AIN-93G diet (Dyets, Inc, Bethlehem, PA) containing 75 IU vitamin E and 0.15-mg selenium
per kilogram of diet [35].The vitamin E-supplemented diet
consisted of AIN-93G containing 200 IU vitamin E and 8mg selenium per kilogram of diet. Riluzole was delivered in
the water at a concentration of 100 pglml with the solution
changed three times weekly. Gabapentin was formulated at
3% by weight in pelleted AIN-93G chow. Animals were
given ad libitum access to food and water. The number of
mice entered in each protocol and their age at entry is given
in Table 1. Mice were randomized across litters when entered
into each protocol. Mice fed the diet supplemented with
vitamin E were entered into the study at 30 days of age.
Mice were entered into the other protocols at 50 days of
age. At 50 to 60 days of age, GlHI+ males weigh 22.1 ?
1.4 gm (mean -C SD) and females weigh 17.4 5 1.2 gm
[32]. Treatment was continued until the mice reached
end-stage disease. When paresis became marked, mice
were fed on the bottom of their cagc with moistened food
Clinical End Points
The clinical condition of mice was monitored three times
per week after entry into each protocol. Onset of clinical
disease was scored by examining the mouse for shaking of
its limbs when suspended in the air by its tail. Clonus, hyperreflexia, and crossed spread of spinal reflexes were also detectable in most mice at this stage of disease, which suggests
that the shaking of the limbs may be due to clinical involvement of upper motor centers [32].At end-stage disease, mice
laid on their sides in their cage. They were alert but did not
attempt to move when gently prodded [32]. Mice were killed
if they could not right themselves within 30 seconds when
placed on their sides on a flat surface or if they were unable
to groom their faces and severe infections developed in one
or both eyes.
Trentment Groups
Mice were fed vitamin E, riluzole, or gabapentin
through their water or chow at the dosages shown in
Table 1. The treatment protocols were continued until
the mice reached end-stage disease. Three different
GIHI+ control groups were run concurrently with
treatment groups (see Table 1). Dietary supplementation with vitamin E at 200 IU per kilogram of mouse
chow produced an 18-fold increase in liver content of
vitamin E (see below). The dosages of riluzole and
gabapentin used in the study were chosen based on
their anticonvulsant effects in rodents (unpublished
Two different sublines of transgenic mice carrying a
mutant human SOD1 gene (designated GIHI+ and
GILI+) were used for the study. The two lines have
the same integration site and show Mendelian segregation of the transgene locus but differ in transgene copy
number and in severity of disease. GIHI+ mice have
higher levels of expression of mutant Cu,Zn SOD, an
earlier onset of clinical disease, and a shorter life span
than do GILIS mice. N29/+ mice carry a wild-type
human SODI gene and show about the same elevation
of total Cu,Zn SOD activity in brains as do GZHIS
mice (data not shown). N29/+ mice do not develop
clinical disease (at least up to 600 days of age).
Wbeel Running Activity
Mice were housed individually in small microisolator cages
that contained an 11.43-cm (4.5 in.) diameter running wheel
suspended from the wire cage lid within 1 cm of the bedding
surface. Magnets glued to opposite sides of the wheel activated a magnetic switch whose closure was detected twice
per revolution by a Mini Mitter DataCol 3 system (Sunriver,
OR). The total number of switch closures per 30-minute
bin (“wheel activity”) was logged continuously for the duration of the experiment. For each mouse, average daily wheel
activity was calculated for the 12-hour period 8 I’M to 8 AM.
Mice were introduced into the cages at 40 to 50 days of age.
Prior to the onset of clinical disease at 90 days of age, daily
wheel activity was relatively constant for each GIH/+
mouse, while wheel activity varied between mice from 470
to 1,620 switch closures per 30 minutes. To allow statistical
comparison of wheel activity between mice, daily wheel activity for each mouse was normalized to average wheel activity
over the period from 65 to 90 days of age. The period of
the data record chosen for normalization was prior to the
expected onset of clinical disease in untreated mice. Total
normalized wheel activity was summed from 90 to 135 days
of age for each mouse to allow statistical comparison between
study groups using a two-tailed Student‘s t test.
Oxidative Stress in Brain and Spinal Cord
We find a gradual, age-dependent increase of vitamin
E in brain and spinal cord of nonTg adult mice and
Table 1. Treatment Groups, Dosages, Routes of Deliuey, and Censorship of Data
Treatment Group
Transgenic Dosage and Route
Mouse Line of Delivery
Vitamin E
Untreated group A GIH/+
Vitamin E
(running wheels)
Untreated group B GIH/+
(running wheels)
Untreated group C
200 IU vitamin E per kg in
AIN-93G defined diet
with 8 ppm selenium
AIN-93G defined diet
200 IU vitamin E per kg in
AIN-93G defined diet
with 8 ppm selenium
AIN-93G defined diet
Age at Entry Number
Enrolled Censored Data
1 death due to unknown causes
45 (wheels)
1 death during piperazine treatment
100 pg/ml in water
3% by weight in AIN93G defined diet
AIN-93G defined diet
1 death after ivermectin treatment,
1 killed by cage mates, 1 death
during piperazine treatment
1 dearh after ivermectin treatment
1 death after ivermectin treatment,
1 death due to unknown causes
3 deaths after ivermectin treatment
3% by weight in A N
93G defined diet
AIN-93G defined diet
30 (chow)
45 (wheels)
Gurney et al: Transgenic Model of FALS
in N29/+ mice expressing wild-type human Cu,Zn
SOD that do not develop motor neuron disease. Vitamin E accumulates at a rate of about 14 ng/gm of
tissudday at essentially identical rates in brain and spinal cord (Figs 1A and 2A). This age-dependent increase
is absent in G I f f / + and GILI+ mouse brain and spinal
cord (Figs 1B and 2B). Vitamin E protects membranes
against oxidative injury by donating a hydrogen atom
to lipid peroxyl radicals [27].The depletion of vitamin E from the brain and spinal cord of GIH/+ and
GILI mice, therefore, implies that cellular membranes
have been exposed to lipid peroxidative attack.
We sought to restore brain vitamin E levels in
GIN/+ and GILI+ mice by dietary supplementation
with vitamin E. After absorption from the gut, vitamin
E is stored primarily in the liver where it is rapidly
turned over [36]. In nonTg and GIH/+ mice, there
was no significant difference in liver vitamin E content,
nor was there a change with age (Fig 2C). Liver vitamin E content in nonTg mice was 7.2 +- 0.38 pg/gm
(mean +- SEM), compared with 7.0 2 0.55 pgigm in
GIHI+ mice. Dietary supplementation with vitamin E
beginning at 30 days of age significantly increased liver
vitamin E content in GIH/+ mice a maximum of 18fold at 70 days of age (see Fig 2C, two-tailed Student's
t test, p < 0.005). Vitamin E content subsequently declined at 100 and 135 days of age; however, the decrease may be due to decreased food intake with advancing paresis. Dietary supplementation of GIH/+
and GILl+ mice with vitamin E partially restored the
increase in spinal cord vitamin E (14 2 7 ng/gm/day;
Fig 1C; linear regression, p = 0.0806).
Vitamin E Delays Onset of CLinical Disease
Only vitamin E significantly delayed the onset of clinical disease (see Materials and Methods) in the transgenic
model. GIH/+ mice normally show onset of clinical
signs at a mean of 92 to 96 days of age (Table 2). In
two studies conducted several months apart, vitamin E
delayed onset of clinical symptoms by 12 to 15 days of
age, an improvement of 14% (Fig3; seeTable2). Riluzole
had no effect on diseaseonset (see Fig 5), while gabapentin
seemed to unmask early clinical signs, although the effect
was not statistically significant (see Table 2). Gabapentin
sometimes causes myoclonus in humans, which could explain the earlier onset of clinical signs in gabapentintreated G I f f I + mice.
Etamin E Delays Disease Progression
Locomotor activity was assessed by monitoring a natural behavior, running in a wheel. When provided with
a wheel in their cage, mice quickly learned to run in
the wheel and maintained high levels of the behavior
throughout the night. In our study, many mice averaged more than 500 revolutions of the running wheel
per 30 minutes over the 12-hour data collection pe-
150 Annals of Neurology Vol 39 No 2 February 1996
Age (days)
Age (days)
2 SO
Fig I . (A) Age-dependent increase in vitamin E content of
mouse spinal cord in nontransgenic (non TR, (*) and in
N 2 9 f + mice (A) that express wild-type human Cu,ZrI superoxide dismutuse und that do not develop diseaxe. For nonTg
mice, the rate of increase (solid line) w a ~I I C 5 ng (mem
2 SEM) vitamin E per gram of spinul cord per day (slope o f
regiession, p = 0,0338).For N29/+ mice the rate of
increase (dotted line) was 17 2 6 ng/gm/day (p = 0.0269).
(B) No increase in vitamin E content with time zuas seen in
C l H / + (0)or GlL/+ spinal cord (0).
A line of reyression
drawn through the combined data fiom G l H / + und GlL/
+ mice shows a downward trend; however, the slope of the
regression (-5 i 5 ng/gm/day) was not stutistically different
from zero (p = 0.329) (C) Dietary supplementation with
vitamin E and selenium slight4 increases spinal cord vitamin
E content in GlH/+ ( 0 ) and GlL/+ mice (m). A line of
regression through the combined datu fiom GlH/+ and
G 1L/+ mice shows a positive trend (14 -+ 7 ng/gm/day),
although the slope of the regression is only weakly signijcant
(p = 0.0806).
70 d
Fig 2. (A) Age-dependent increase in vitamin E content of brain in nontransgenic (nonTg) (+) and N29/+ mice (A) that
express wild-type human Cu,Zn superoxide dismutase. The vitamin E content of nonTg mouse brain increases at a rate of 12 2
5 nglgmlday (p = 0.021), while in N291i- mouse brain the rate of increase is 16 2 6 ng/gmlday (p = 0.010). (B) The vitamouse brain shows a downward trend with age, although the slope of the regresmin E content o f GlH/+ (0)or GlL/+ (0)
sion is not statistically signifcant. (C) Liver vitamin E content does not change with age in either nonTg (0)
or GlH/+ mice
In G1H/+ mice (0)fed a diet supplemented with vitamin E beginning at 30 days of age, the vitamin E content of liver is
signifcantly elevated at 70, 100, and 135 days of age (two-tailed Student: t test, p = 0.005).
Table 2. Effects of Vitamin E, Riluzole, and Gabapentin on the Onset of Clinical Diseme (See Materials and Methods)
in GI H/+ Mice
Vitamin E
Vitamin E’
(running wheels)
Contemporaneous Control
Groups (A, B, and C)a
108 & 10 (13)
107 2 10 (10)
95 +- 13 (23) A
92 2 13 (19) B
98 +- 11 (10)
83 ? 18 (10)
95 t 12 (9)
95 2 12 (9)
- 1.66
Values tabulated are mean 2 SD (sample size). Statistical comparisons were with a two-tailed Student’s t test (NS
not significant).
‘No significant statistical differences were observed between the three control groups (A, B, and C).
riod. Given that the diameter of the running wheel was
11.43 cm, we calculate that the mice were running an
average of4.3 km (2.7 mi) per night. Untreated GIH/+
mice maintained relatively consistent levels of wheel
activity until the onset of clinical disease, after which
they showed a monotonic decline in the behavior (Fig
4). The decline in wheel running was due to weakness
and to increasing paresis. As mice approached end-stage
disease, they continued to pull themselves into the wheel
but were unable to rotate it. We often found them
sitting in the wheel in preference to the cage bedding.
Thus, the assay provided a continuous, quantitative measure of disease progression.
Mice given access to wheels and who therefore were
more active, showed no difference in onset of clinical
disease or in survival in comparison with more sedentary mice housed in standard mouse cages (untreated
group B vs groups A and C, see Tables 2 and 3).
Therefore, oxidative stress caused by physical exertion
[37] was not a risk factor in the transgenic model.
Gurney et al: Transgenic Model of FALS
mice fed vitamin E, an improvement of 60% (twotailed Student’s t test, p = 0.0021). Put another way,
mice fed vitamin E ran 45 km (28 mi) farther than
mice fed the basal diet over that period of time. Thus,
dietary supplementation with vitamin E significantly
delayed progression of disease in GIH/+ mice.
Riluzole and Gabapentin Extend Survival
Despite the positive effects of vitamin E on disease onset, no effect on survival was noted (see Fig 3 and
Table 3). Therefore, measures of progression and sur-
0 untrcated
Age (days)
5 .4-
0 vitaminE
0 untreated
.2 -
, ,
, , ,
Age (days)
Fig 3. (Top) Disease onset (see Materials and Methods) was
sign$cantly delayed in G 1H/+ mice f . d a diet supplemented
with vitamin E. The graph shows cumulative probability of
disease onset f o r C; 1H/+ mice f . d vitamin E supplemented
(u) or control diets (0).
(Bottom) Dietary supplementation
with vitamin E had no effect on survival o f G1H/S mice.
The graph plots cumulative probabiliy of survival f o v mice
assigned to each ti’eatment group. For statistical unalJLtis of
data see 7;zbles 2 and 3.
Dietary supplementation with vitamin E significantly delayed the progression of disease in G1HI-t
mice. Normal levels of wheel activity were prolonged
by 12 to 15 days (see Fig 4 ) , which was roughly the
amount that clinical onset of disease was delayed, as
shown in the previous experiment (see Fig 3). Furthermore, mice fed vitamin E maintained higher average
levels of wheel activity at all ages than did age-matched
controls (see Fig 4). For statistical comparison, we calculated the total wheel activity for each mouse from
the equation.
Total wheel activity
activity (12 hr)
Mice fed vitamin E recorded greater total wheel activity
from 90 to 135 days of age, the period after the onset
of clinical disease in untreated controls (see Fig 4).
GIH/+ mice fed a basal diet recorded a total wheel
activity of 16.7 ? 5.0 compared with 27.2 ? 7.4 for
152 Annals of Neurology
Vol 39 No 2
February 1996
vival are separable in the transgenic model and may be
affected somewhat independently by therapeutic
agents. This inference is strengthened further by studies
of the neuroprotective and anticonvulsant drugs riluzole and gabapentin.
Although riluzole did not delay onset of clinical disease, it significantly prolonged survival of GIN/+ mice
(Fig 5 ) . Survival improved 13 to 15 days (see Table
3), an increase of about 11 %, which roughly equals
the improvement in survival seen in patients with the
sporadic form of ALS [12]. With gabapentin treatment
of GIH/+ mice there was a positive trend with a slight
prolongation of survival of about 8 days (Fig 6; see
Table 3). G IH / + mice have a fairly severe form of
motor neuron disease characterized by short survival.
We, therefore, decided to test gabapentin in a second
line of transgenic mice designated G1Lli- . These have
the same mutant SOD1 transgene as GIH/+ mice but
carry fewer transgene copies and, consequentIy, have a
less severe form of motor disease that is characterized
by longer survival (see Fig 6). Gabapentin had better
clinical efficacy in GIL/+ mice and extended survival
by 9 days (two-tailed Student’s t test, p = 0.012). Survival of untreared GIL/+ mice was 168 i 10 d (mean
? SD, n = 25) as compared to 177 ? 11 d ( n =
13) for mice fed 30/0 gabapentin in chow.
Vitamin E levels in brain normally increase with age
in rodents [34, 361. In GZH/+ and GILI+ transgenic
mice, however, the normal age-dependent increase in
vitamin E fails to occur. Vitamin E is the major cellular
antioxidant defense against lipid peroxidation [27],
which suggests that expression of mutant human
Cu,Zn S O D in transgenic mice a t levels sufficient to
cause motor neuron disease causes slow peroxidative
attack on membrane lipids. Peroxidative damage is not
due simply to elevation of Cu,Zn S O D activity in
brain and spinal cord [25], since no depletion of vitamin E is seen in N29/+ mice that have elevated SOD
activity but that do not develop disease. In other
words, simply increasing the dismutation of superoxide
to hydrogen peroxide by increasing enzyme levels is
not the basis for disease in the GIH/+ mice, because
this also would occur in NZY+ mice.
45 -
I .25
40 v,
35 30 25 -
Fig 4. (Lefi) Normalized daily wheel activig for G 1H/+ animals f i d a control diet (0,
n = 18) or a diet supplemented with
vitamin E (a, n = 10). Animals fed the vitamin E-supplemented diet maintained greater wheel activity at all ages a$er the
mean onset of clinical disease in the control group at 92 t 13 days of age. (Right) Comparison of total wheel activig between
GIH/+ animalr fed a control diet (0)
and a vitamin E-supplemented diet (a). Total wheel activity was calculated as described
in Results. GI H/+ animals fed the vitamin E-supplemented diet recorded sign&antly higher total wheel activity during the
symptomatic phase of the disease fiom 90 to 135 days of age (two-tailed Student; t test, A = 10.5, df = 15, r = 3.72, p =
Table 3. Effects of Vitamin E, Riluzole, and Gabapentin on Survival (See Materials and Methods) o f G I W + Mice
Conrrol Groupa
Vitamin E
Vitamin E
(running wheels)
134 5 9 (13)
131 t 11 (10)
136 ? 15 (23) A
130 t 11 (18) B
148 2 14 (8)
141 -t- 12 (9)
134 t 8 (8) C
134 t 8 (8) C
Values tabulated are mean C SD (sample size). Statistical comparisons were with a two-tailed Student's t test (NS = not significant).
'No significant statistical differences were observed herween the three control groups (A, B, and C ) .
In support of the biochemical measurements, dietary
supplementation of GIN/+ mice with vitamin E delayed onset of clinical disease and disease progression
but failed to extend survival. In contrast, riluzole and
gabapentin, two putative presynaptic modulators of the
glutamatergic system [ 12, 281, both extended survival,
while neither drug significantly affected the onset of
clinical disease. Riluzole had about the same clinical
benefit in the transgenic model as was seen in a recent
clinical trial in the sporadic form of ALS [12]. There
was clear separation of drug effects on onset, progression, and survival for the three therapeutics tested. Pathology during the presymptomatic phase of disease
in G I H I t mice is characterized by the slow accumulation of damage to intracellular, membranous organelles
such as the mitochondria and endoplasmic reticulum,
whereas the symptomatic phase of disease is associated
with motor neuron loss [21, 32, 331. Thus, the therapeutic benefit of vitamin E on disease onset implicates
lipid peroxidation in the presymptomatic phase of the
disease, while glutamate-mediated excitotoxic mechanisms may be associated with motor neuron loss during
the symptomatic phase of disease.
How mutation of human Cu,Zn SOD might cause
damage to motor neurons and initiate pathogenesis
during the presymptomatic phase of disease is unclear.
Gurney et al: Transgenic Model of FALS
4 I
Age (days)
Age (days)
A s (days)
Age (days)
Fig 5. (Top) Treatment o f G1H/+ mice with riluzole (0)
had no effect on disease onset in comparison with controls
(Bottom) Riluzole significantly extended survival of
G1H/f mice (p = 0.039, see Table 3). the outlier
riluzole-treated GI H/+ mouse with the longest survival
(181 days) is removed fiom the data set, the prolongation of
urvival remains statistically signzjkant (df = 13, t = 2.39,
p = 0.034I).
Hypotheses include generation of hydroxyl radical
(HO.) from hydrogen peroxide [38], formation of
a nitronium-like intermediate with peroxynitrite
(NO2+OCu-SOD), which subsequently catalyzes nitration of proteins on tyrosyl residues [39], and the
release of free copper, which could promote free radical-mediated damage to membranes [40]. Reaction
pathways that increase free-radical formation will promote oxidation of polyunsaturated fatty acids (lipid
peroxidation) and cause the formation of lipid peroxyl
radicals [4O]. Because lipid peroxidation is self-propagating, the attack of one reactive free radical on a polyunsaturated fatty acid side chain can be amplified into
the formation of multiple lipid peroxides.
Lipid peroxidation is opposed by vitamin E and is
implicated in the presymptomatic phase of pathogenesis by our results. Vitamin E (a-tocopherol) partitions
to membranes where it protects against lipid peroxidemediated damage by donating a hydrogen atom to a
lipid peroxide radical (LOO.) to give a lipid hydroper-
154 Annals of Neurology Vol 39 No 2
February 1996
Fig 6 Effect of gabapentin treutment on the cumulative probability of survival of GI H/+ (Top) or GlL/+ mice (Bottom). G1L/+ mice have a lower transgene copy number and
slower progression of disease than do GIH/+ mice. For
G 1H/+ mice, there was a trend for improved survival (see
Table 3) in mice j d gabapentin (0)
compared with controls
while for GlL/+ mice, gabapentin signijcantly
improved survival in comparison with controls (two-tailed Student: t test, A = 9.3, d€ = 36 t = 2.63, p = 0.0I2).
oxide (LOOH) and a vitamin E radical (T.). In the
absence of iron, lipid hydroperoxides are eliminated by
soluble or membrane-bound forms of glutathione peroxidase. Some reduced vitamin E (TH) can be regenerated from the oxidized form (T-) by donation of an
electron from reduced ascorbate, but the remainder is
further oxidized to tocopherol quinone and cannot be
reconverted back to vitamin E [27]. In addition, vitamin E transport into the brain is limited, so vitamin
E levels can only be replenished slowly after an oxidative insult [41]. Thus, vitamin E depletion is an indication of free-radical flux in tissues and of the degree to
which membranes are placed under oxidative stress.
This has been shown previously in models of acute
spinal cord injury [42]and focal cerebral ischemia [43]
in which there occurs a rapid, postinsult depletion of
tissue vitamin E indicative of lipid peroxidation. During the symptomatic phase of disease, which is characterized in the transgenic model by motor neuron loss,
oxidative and excitotoxic mechanisms of damage may
become mutually reinforcing and self-sustaining once
a threshold for damage has been passed. For example, xanthinelxanthine oxidase-derived oxygen radicals
have been shown to increase glutamate release from
brain slices in vitro [44], while glutamate, in turn, is
known to increase brain levels of hydroxyl radicals
impairing the integrity of intracellular, membranous
organelles. Membrane calcium permeability [48],calcium-ATPase-associated extrusion [43], and mitochondrial sequestration of calcium [50, 511 are all exquisitely sensitive to lipid peroxidation. Mitochondria
may be one of the early targets of damage in the
transgenic model due to the high flux of free radicals
generated by electron transport [20, 331. A defect in
mitochondrial energy metabolism is also suggested in
Although vitamin E depletion occurs in both brain
FALS patients by an increase in mitochondrial complex
and spinal cord, in the G1HI-k and GlLI+ transgenic
I enzymatic activity in motor cortex [17]. Since mitolines only motor neurons and motor related areas of
chondria sequester intracellular calcium, a mitochonthe brainstem develop pathology [18, 21, 32, 331. Padrial defect could cause loss of calcium homeostasis eithology has not been observed elsewhere in the brain,
ther directly through a failure to sequester calcium or
nor is there damage to somatic tissues [32]. Apparently,
indirectly due to an energy deficit [15, 171. Once damdamage can spread in brain beyond the motor system,
age reaches a threshold, one or more of these alterations
however, if expression of mutant human Cu,Zn SOD
may make the motor neuron sensitive to excitotoxic
is sufficiently high, as reported by Wong and associates
damage by ambient levels of glutamate during the
[2O]. The changes in vitamin E levels that we detect
symptomatic phase of the disease. Motor neurons may
are small and may reflect the chronic nature of damage
be particularly vulnerable to loss due to the number
in this model compared with the rapid and acute
and activity of their glutamatergic inputs, selective exchanges in brain vitamin E elicited by trauma or focal
cerebral ischemia [42, 431. Because the changes were
pression of subtypes of glutamate receptors with enhanced calcium permeability [52],or low expression of
small, we used linear regression to analyze the vitamin
calcium-binding proteins [53].
E data due to the small sample size and the age distriThe therapeutic benefit of riluzole and gabapentin
bution of the data.* Further support for peroxidative
on survival of mutant SOD1 transgenic mice implicates
damage to membrane lipids in the transgenic model
should come from measurement of lipid hydroperoxglutamate in the mechanism of neuron death during
ides (which are formed transiently) and more stable
the symptomatic phase of the disease. Riluzole inhibits
lipid peroxidation breakdown products such as maloglutamate release from nerve terminals [54], perhaps by
stabilizing the inactivated form of voltage-gated sodium
nyldialdehyde. Such experiments are in progress.
Confirming that a similar peroxidative attack on
channels [55, 561, and also acts as a noncompetitive
membrane lipids occurs in FALS patients may be difan tagon ist at N-me thyl- n-asparta te (NMDA) recepficult. Global depletion of vitamin E in the brain and
tors [57]. The drug is neuroprotective in a number of
spinal cord of transgenic mice might be due to our
models of cerebral ischemia involving excitotoxic
achieving such high levels of mutant protein expression
injury [58, 591 but also protects against N-methyl-4[le]. In patients who inherit only a single copy of the
phenyl- 1,2,3,6-tetrahydropyridine(MPTP) neurotoxicmutant SOD1 gene, lower levels of mutant protein exity, which is thought to involve a free-radical mechapression may limit lipid peroxidative attack on memnism [60].The mechanism of gabapentin action is not
branes to spinal cord motor neurons or other neuronal
well understood, but it may act as a presynaptic inhibipopulations that preferentially express high levels of
tor of glutamate biosynthesis [GI]. Gabapentin inhibits
Cu,Zn SOD [46,471.
the enzyme branched-chain amino acid transferase
Formation of lipid peroxyl radicals due to expression
(BCAA-T) [62]. BCAA-T in brain contributes to the
of mutant human Cu,Zn SOD might precipitate
biosynthesis of glutamate from a-ketoglutarate by catapathogenesis by oxidizing membrane proteins or by
lyzing the transfer of an amino group from a branchedchain amino acid donor (leucine, isoleucine, or valine).
In brain, this pathway may account for a small but
*We estimate a sample size of at least 13 animals per group would
significant fraction of glutamate biosynthesis in neube necessary to detect statistical significance using Student’s t test
rons and glia. In addition, the drug may increase glutaas follows. At 75 days of age, we found that spinal cord vitamin E
mate metabolism into y-aminobutyric acid (GABA)
levels averaged 5.7 t 1.04 pgigm (mean ‘iSD) in a sample of 12
nonTg mice. We also estimate that vitamin E content increases at
[63]. Both compounds protect rat motor neurons from
a rate of about 14 ng/gm/day and thus should reach a value of 6.5
glutamate-mediated toxicity in a spinal cord explant
pglgm in nonTg mice by 135 days of age. This increase does not
model of neuroprotectioii [64).
occur in GlH/+ mice, so we expect a difference of 0.84pglgm or
14% between the nvo groups. From an estimate of the 95% confiThese studies link the oxidative and excitotoxic thedence interval given by to,,[.\i(2o2/n)]or n = 2(l.04’)/(0.84/2.0)z ories of the etiology of ALS. They show the utility
for n, + n, - 2 degrees of freedom, we estimate a sample size of
of transgenic models of neurodegenerative diseases and
13 or more mice per group is need for statistical significance at the
0.05 level.
suggest that the sporadic and familial forms of ALS
Gurney et al: Transgenic Model of FALS
share common underlying mechanisms of pathogenesis. They place excitotoxic mechanisms of damage
downstream in the pathway of cellular damage leading
to neurodegeneration. The precipitating events that
initiate pathogenesis in the sporadic form of ALS are
unknown, while in the familial form of the disease
linked to mutation of Cu,Zn SOD, the precipitating
event is likely to involve peroxidative attack on membrane lipids. Glutamate-mediated excitotoxic mechanisms appear to dominate in the symptomatic phase
of the disease, which is characterized by motor neuron
These studies were conducted by the FALS Therapeutics Working
Group with funding provided by the Muscular Dystrophy Association.
We thank Margarita Dubocovich and Susan Benloucif for help with
the initial running-wheel experiments.
1. Mulder DW. Clinical limits of amyotrophic lateral sclerosis.
In: Rowland LP, ed. Human motor neuron diseases. New
York: Raven Press, 1982:15-22
2. Emery AEH, Holloway S. Familial motor neuron diseases. In:
Rowland LP, ed. Human motor neuron diseases. New York:
Raven Press, 1982:139-147
3. Rothstein JD, Jin L, Dykes-Hoberg M, Kuncl RW. Chronic
inhibition of glutamate uptake produces a model of slow neurotoxicity. Proc Natl Acad Sci USA 1993;90:6591-6595
4. Spencer PS, Allen C N , Kisby GE, et al. Lathyrism and western
Pacific amyotrophic lateral sclerosis: etiology of short and long
latency motor system disorders. Adv Neurol 199 1;56:287-299
5. Appel SH, Smith RG, Engelhardt JI, Stefani E. Evidence for
autoimmunity in amyotrophic lateral sclerosis. J Neurol Sci
6. Yasui M, Yase Y, Ota K, Garruto RM. Aluminum deposition
in the central nervous system of patients with amyotrophic
lateral sclerosis from the Kii Peninsula of Japan. Neurotoxicology 1991;12:6 15-620
7. Choi DW. Glutamate neurotoxiciry and diseases of the nervous
system. Neuron 1988;1:623-634
8. Canu W, Billiard M, Baldy-Mouliner M. Fasting plasma and
CSF amino acid levels in ALS. Acta Neurol Scand 1993;88:
9. Plaitakis A, Constantakakis E. Altered metabolism of excitatory
amino acids, N-acetyl-aspartate and N-acetyl-aspartyl-glutamate in ALS. Brain Res Bull 1993;30:381-386
10. Rothstein J , Martin IJ, Kuncl RW. Decreased glutamate transport by the brain and spinal cord in ALS. N Engl J Med 1992;
326: 1464-1468
11. Rothstein JD, Van Kammen M , Levey AI, et al. Selective loss
of glial gluraniate transporter GL1‘-1 in amyotrophic lateral
sclerosis. Ann Neurol 1995;38:73-84
12. Rensitnon G, Lacomblez L, Meininger V, and the ALS/Kiluzole Study Group. A controlled trial of riluzole in amyotrophic
lateral sclerosis. N Engl J Med 1994;330:585-591
13. Rosen DR, Siddique T, Patterson D, et al. Mutations in Cu,Zn
superoxide dismurase gene are associated with familial amyotrophic lateral sclerosis. Nature 1993;362:59-62
14. Figlewicz DA, Krizus A, Martinoli MG, et 31. Variants of the
heavy neurofilament subunit are associated with the develop-
156 Annals of Neurology Vol 39 No 2 February 1996
ment of amyotrophic lateral sclerosis. H u m Mol Genet 1994;
3:1757- 1761
15. Brown RH Jr. Amyotrophic lateral sclerosis: recent insights
from genetics and transgenic mice. Cell 1995;80:687-692
16. Deng HX,Hentati A, Tainer JA, et al. Amyotrophic lateral
sclerosis and structural defects in Cu,Zn superoxide dismutase.
Science 1993;261:1047-1051
17. Bowling AC,Schulz JB, Brown R H Jr, Beal MF. Superoxide
dismurase activity, oxidative damage, and mitochondrial energy
metabolism in familial and sporadic amyotrophic lateral sclerosis. J Neurochem 1993;61:2322-2325
18. Gurney ME, Pu H. Chiu AY, et al. Motor neuron degeneration in mice expressing a human Cu,Zn superoxide dismutase
mutation. Science 1934;264:1772- 1775
19. Ripps ME, Huntlep GW, Hof PR, et al. Transgenic mice expressing an altered murine superoxide dismutase gene provide
an animal model of amyotrophic lateral sclerosis. Proc Natl
Acad Sci USA 1995;92:689-693
20. Wong PC, Pardo CA, Borchelr DR, et al. An adverse property
of a familial ALS-linked SOD1 mutation causes motor neuron
disease characterized by vacuolar degeneration of mitochondria.
Neuron 1995;14:1105-1116
21. Dal Canto MC, Gurney ME. The development of CNS pathology in a murine transgenic model of human ALS. Am J
Pathol 1994;145:1271-1280
22. Avraham KB, Sugarman H, Rotshenker S, Groner Y. Down’s
syndrome: morphological remodeling and increased complexity
in the neuromuscular junction of transgenic CuZn-superoxide
dismutase mice. J Neurocytol 1991;20:208-215
23. Przedborski S, Kostic V, Jackson-Lewis V, et al. Transgenic mice with increased Cu/Zn-superoxide dismutase activity
are resistant to N-methyl-4-phenyl-l,2,3,6-tetrahydropyridineinduced neurotoxicity. J Neurosci 1992;12:1658-1667
24. Yang G, Chan PH, Chen J, et al. Human copper-zinc superoxide dismutase transgenic mice are highly resistant to reperfusion
injuty after focal cerebral ischemia. Stroke 1994;25: 165-1 70
25. McCord JM. Mutant mice, Cu,Zn superoxide dismutase, and
motor neuron degeneration. Science 1995;266:1586-1587
26. Gurney ME. Transgenic-mouse model of amyotrophic lateral
sclerosis. N Engl J Med 1994;331:1721-1722 (Letter)
27. McCay 1’B. Vitamin E: interactions with free radicals and
ascorbate. Annu Rev Nutr 1985;5:323-340
28. Taylor CP. Emerging perspectives on the mechanism of action
of gabapentin. Neurology 1994;44(suppl 5):SIO-SlG
29. Hallewell RA,Puma JP, Mullenbach GT, Najarian RC. Superoxide and superoxide dismutase. In: Rotilo G, ed. Chemistry,
biology and medicine. New York: Elsevier, 1986249-256
30. Epstein CJ, Avraham KB, Lovett M, er al. Transgenic mice
with increased Cu/Zn-superoxide dismutase activity: animal
model of dosage effects in Down syndrome. Proc Natl Acad
Sci USA 1987;84:8044-8048
31. Lipman NS, Dalton SD, Stuart AR, Arruda K. Eradication of
pinworms (Sypharia obvelara) from a large mouse breeding colony by combination oral antihelmintic therapy. Lab Anim Sci
32. Chiu AY, Zhai P, Dal Canto MC, et al. Age-dependent penetrance of disease in a transgenic model of familial amyotrophic
lateral sclerosis. Mol Cell Neurosci 1995;6:349-362
33. Dal Canto MC, Gurney ME. Neuropathological changes in
two lines of mice carrying a transgene for mutant human
Cu,%n SOD, and in mice overexpressing wild type human
SOD: a model of familial amyotrophic lateral sclerosis. Brain
Res 1995;676:25 -40
34. Zhang J-R, Andrus PK, Hall ED. Age-related changes in hydroxyl radical stress and antioxidants in gerbil brain. J Neurochem 1993;61:1640-1647
35. Reeves ED, Nielsen FH, Fahey G C Jr. AIN-93 purified diets
for laboratory rodents: final report of the American Institute
of Nutrition ad hoc writing committee on the reformulation
of the AIN-76A rodent diet. J Nutr 1993;123:1939-1951
36. Vatassery GT, Brin MF, Fahn S, et al. Effect of high doses of
dietary vitamin E on the concentrations of vitamin E in several
brain regions, plasma, liver and adipose tissue of rats. J Neurochem 1988;51:62 1-623
37. Clarkson PM. Antioxidants and physical performance. Crit Rev
Food Sci Nurr 1995;35:131-I41
38. Yim MB, Chock PB, Stadtman ER. Copper,zinc superoxide
dismutase catalyzes hydroxyl radical production from hydrogen
peroxide. Proc Natl Acad Sci USA 1990;87:5006-5010
39. Beckman JS, Carson M , Smith C D , Koppenol WH. ALS,
S O D and peroxynitrite. Nature 1993;364:584 (Letter)
40. Halliwell B, Gutteridge JMC. Free radicals in biology and
medicine. New York: Oxford University Press, 1989:1-543
41. Vatassery G T , Angerhofer CK, Knox CA. Effect of age on
vitamin E concentrations in various regions of the brain and
a few selected peripheral tissues of the rat, and on the uptake
of radioactive vitamin E by various regions of rat brain. J Neurochem 1984;43:409-412
42. Hall ED, Yonkers PA, Horan KL, Braughler JM. Correlation
between attenuation of post-traumatic spinal cord ischemia and
preservation of spinal tissue vitamin E by the 21-aminosteroid
lipid peroxidation inhibitor U-74006F: evidence for an in vivo
antioxidant mechanism. J Neurotrauma 1989;6:169-176
43. Hall ED, Pazara KE, Braughler JM. Effects of tirilazad mesylate
on post-ischemic brain lipid peroxidation and recovery of extracellular calcium in gerbils. Stroke 1991;22:361-366
44. Pelligrini-Giampietro DE, Cherici G , Alesiani M , et al. Excitatory amino acid release and free radical formation may cooperate in the genesis of ischemia-induced neuronal damage. J Neurosci 1990;lO: 1035-1041
45. Boisvert DPJ, Schreiber C. Interrelationship of excitotoxic and
free radical mechanisms. In: Krieglstein J, Oberpichler H, eds.
Pharmacology of cerebral ischemia. Stuttgart: Wissenshaftliche
Verlaggesellschaft mbH, 1992:l-10
46. Tsuda T , Munthasser S, Fraser PE, et al. Analysis of the functional effects of a mutation in SOD1 associated with familial
amyotrophic lateral sclerosis. Neuron 1994;I3:727-736
47. Pardo CA, Xu 2, Borchelt DR, et al. Superoxide dismutase is
an abundant component in cell bodies, dendrites, and axons
of motor neurons and in a subset of other neurons. Proc Natl
Acad Sci USA 1995;92:954-958
48. Braughler JM, Duncan LA, Chase RL. Interaction of lipid peroxidation and calcium in the pathogenesis of neuronal injury.
CNS Trauma 1985;2:299-304
49. Rohn T T , Hinds TR, Vincenzi FF. Ion transport ATPases as
targets for free radical damage: protection by an aminosteroid of Ca" pump ATPase and N a t / K t pump ATPase of
human red cell membranes. Biochem Pharmacol 1993;46:
50. Malis C D , Bonventure JV. Susceptibility of mitochondrial
membranes to calcium and reactive oxygen species: implications for ischemic and toxic tissue damage. In: Karnofsky ML,
Leaf A, Bolis LC, eds. Biological membranes: aberration in
membrane structure and function. New York: Alan R. Liss,
5 1. Vlessis AA, Mela-Riker L. The potential role of mitochondrial
calcium metabolism during reperfusion injury. Am J Physiol
52. Hollmann M, Heinemann S. Cloned glutamate receptors.
Annu Rev Neurosci 1994;17:31- 108
53. Reiner A, Medina L, Figuerdo-Cardenas G, Anfinson S. Brainstern motoneuron pools that are selectively resistant in amyotrophic lateral sclerosis are preferentially enriched in pawalbumin: evidence from monkey brainstem for a calciummediated mechanism in sporadic ALS. Exp Neurol 1995;131:
54. Martin D, Thompson MA, Nadler JV. The neuroprotective
agent riluzole inhibits release of glutamate and aspartate from
slices of hippocampal area CAI. Eur J Pharmacol 1993;250:
55. Benoit E, Escande D. Riluzole specifically blocks inactivated
Na channels in myelinated nerve fiber. Pflugers Arch Eur J
Pharmacol 1991;419:603-609
56. Hebert T, Drapeau P, Pradier L, Dunn RJ. Block of the rat
brain 11A sodium channel alpha subunit by the neuroprotective
drug riluzole. Mol Pharmacol 1994;45:1055-1060
57. Malgouris C, Daniel M, Doble A. Neuroprotective effects of
riluzole on N-methyl-D-aspartate or veratridine-induced neurotoxicity in rat hippocampal slices. Neurosci Lett 1994;177:
58. Wahl F, Allix M, Plotkine M, Boulu RG. Effect of riluzole on
focal cerebral ischemia in rats. Eur J Pharmacol 1993;230:209214
59. Pratt J, Rataud J , Bardot F, et al. Neuroprotective actions of
riluzole in rodent models of global and focal cerebral ischaemia.
Neurosci Lett 1992; 140:225-230
GO. Boireau A, Dubedat P,Bordier F, et al. Riluzole and experimental parkinsonism: antagonism of MPTP-induced decrease
in central dopamine levels in mice. Neuroreport 1994;5:26572660
61. Welty DF, Schielke GP, Kothstein JD. Potential treatment of
amyotrophic lateral sclerosis by the anticonvulsant gabapentin:
a hypothesis. Ann Pharmacother 1996 (In press)
62. Goldlust A, Su T-Z, Welty DF, et al. Effects of the anticonvulsant drug gabapentin on the enzymes in the metabolic pathways of glutamate and GABA. Epilepsy Res 1995;22:1-11
63. Loescher W, Hoenack D, Taylor CP. Gabapentin increases
aminooxyacetic acid-induced GABA accumulation in regions
of rat brain. Neurosci Lett 1991;128:150-154
64. Rothstein JD, Kuncl RW. Neuroprotective strategies in a
model of chronic glutamate-mediated motor neuron toxicity.
J Neurochem 1995;65:643-651
Gurney et al: Transgenic Model of FALS
Без категории
Размер файла
1 065 Кб
mode, vitamins, transgenic, benefits, lateral, familiar, sclerosis, amyotrophic, riluzole, gababapentin
Пожаловаться на содержимое документа