вход по аккаунту


CGS-19755 A competitive NMDA receptor antagonist reduces calcium-calmodulin binding and improves outcome after global cerebral ischemia.

код для вставкиСкачать
CGS-19755, A Competitive NMDA
Receptor Antagonist, Reduces
Calcium-Cahoddn Binding and Improves
Outcome after Global Cerebral Ischemia
J. C. Grotta, MD," C. M. Picone, MD,I* P. T. Ostrow, MD," R. A. Strong, BS," R. M. Earls, BS,"
L. P. Yao, MD," H. M. Rhoades, PhD," and J. R. Dedman, PhDi
We evaluated several doses of cis-4-(phosphonomethyl)-2-piperidine-carboxylic
acid (CGS-19755),a potent competitive
N-methyl-D-aspartate (NMDA) receptor antagonist, systemically administered either before or after 20 to 30 minutes
of global ischemia in rats. We measured outcome by mortality, histological damage by light microscopy, and learning
ability on an eight-arm maze, and determined the drug's mechanism of action by an immunohistochemical assay of
calcium-calmodulin binding. High-dose treatment begun prior to ischemia resulted in reduced cellular damage in
severely ischemic hippocampal tissue, but also caused high mortality due to respiratory depression. Treatment begun
30 minutes after ischemia resulted in little histological protection but significantly improved learning ability when
tested 1 month after ischemia, and did not increase mortality. Furthermore, CGS-19755, 10 mglkg intraperitoneally,
begun either before or after ischemia substantially reduced calcium influx into ischemic neurons as evidenced by
reduced calcium-calmodulin binding. We conclude that CGS-19755 prevents calcium entry into ischemic neurons and
may be effective therapy for very acute cerebral ischemia.
Grotta JC, Picone CM, Ostrow PT, Strong RA, Earls RM, Yao LP, Rhoades HM, Dedman JR.
CGS-19755, a competitive NMDA receptor antagonist, reduces calcium-calmodulin binding and improves
outcome after global cerebral ischemia. Ann Neurol 1990;27:612-619
Activation of the N-methyl-D-aspartate (NMDA) excitatory amino acid receptor by glutamate after ischemia results in substantial entry of ionic calcium into
neurons, consequent progressive irreversible neuronal
damage by calcium-activated enzyme systems, and delayed cell death 11-51. The observation that antagonists of excitatory neurotransmission can protect neurons against ischemic injury has raised hopes that
effective therapy for stroke may be possible [6].
However, experimental results have varied, depending
on the choice of drug, the model used, and preischemic versus postischemic administration. The most
consistently positive results have been found with noncompetitive antagonists such as MK-801 (ditocilpine
maleate) in models of focal ischemia. These compounds directly block the ion channel associated with
the NMDA receptor and have been effective when
given before or after ischemia 17-91. More recently
MK-801 has also proved effective in models of global
ischemia 110, 11). Its disadvantages, however, include
its possible side effects on higher cortical function,
which might be compounded by its prolonged binding
characteristics. A recent study of MK-801 demonstrated no benefit in function or histologic damage in a
primate global ischemia model 1121.
Competitive antagonists of the NMDA receptor
have been somewhat less well studied. Most have limited penetration into the brain, so that there are few
published studies evaluating their efficacy after systemic administration [b]. Their ability to block calcium
influx through the NMDA-associated ion channel has
not been proved, and their side effects are still largely
We evaluated cis-4-(phosphonomethyl)-2-piperidine-carboxylic acid (CGS-19755) because it is among
the most potent competitive antagonists of the
NMDA receptor in the central nervous system studied
to date 111, 131 and preliminary studies have suggested its efficacy after systemic administration 111,
141.We designed our study to evaluate several doses
begun either before or after ischemia, and used a
global ischemia model of delayed neuronal damage in
From the Departments of "Neurology and tphysiology, The University of Texas Medical School at Houston, Houston, TX.
Address correspondence to Dr Grotta, University of Texas Medical
School, Department of Neurology, 6431 Fannin, MSB 7.044, Houston, TX 77030.
Received Jul 7, 1989, and in revised form Oct 3 and Dec 7. Accepted for publication Dec 11, 1989.
Copyright 0 1990 by the American Neurological Association
the hippocampus and cortex where NMDA receptors
are concentrated 121. With this model, it has been
difficult to demonstrate neuronal protection with other
classes of therapies such as calcium channel blockers
115, 161 and inhibitors of glutamate release 1171. We
monitored adverse effects on behavior by evaluating
learning and memory, and because our animals were
unanesthetized, w e could observe any sedative effects
of t h e d r u g o n ventilarion. Finally, we directly measured calcium binding to calmodulin (CaM) in hippocampal a n d cortical neurons, thereby indirectly determining t h e ability of CGS-19755 to prevent calcium
entry through the NMDA-associated ion channel.
of Ischemia
Male Wistar rats weighing 250 gm were subjected to global
ischemia without anesthesia using an adaptation of the fourvessel occlusion model Cl6, 181. With the rats under chloral
hydrate anesthesia, the vertebral arteries were coagulated
and the carotid arteries loosely tagged for subsequent rapid
access and occlusion. Twenty-four hours later, after a few
seconds of ether inhalation, an atraumatic aneurysm clip was
placed across both carotid arteries and the cervical muscles
ligated for 20 minutes. For the studies of calcium-CaM binding immediately and 2 hours after ischemia, because longterm survival was not important, a 30-minute period of ischemia was used to assure widespread ischemia to multiple
areas of the brain.
With this technique, animals were not provided with ventilatory support and no anesthetic was required during ischemia or reperfusion. Animals were fasted for 24 hours before ischemia, and were monitored for blood pressure,
arterial blood gases, body temperature, head temperature,
plasma glucose levels, hematocrit, and electroencephalographic (EEG) changes during the ischemic period.
Throughout the operative procedures and during ischemia,
animals were kept under a heating lamp, and a warming
blanket used as necessary to maintain rectal temperature at
37°C. Animals were included for analysis only if they lost
their righting reff ex and all EEG activity for the entire period
of ischemia, and did not have clinical seizures during the
postischemic period.
Several dose regimens of CGS-19755 were employed. In
a previous study in gerbils, doses lower than 10 mg/kg were
ineffective and an intraperitoneal dose resulted in glutamate
antagonism lasting approximately 2 hours [ 141. Therefore
doses of 10 to 30 mg repeated every 2 hours were chosen for
the present study. In Group 1, 30 mg/kg was administered
intraperitoneally immediately prior to ischemia and repeated
2 , 4 , and 6 hours after terminating ischemia. In Group 2, 10
mg/kg was administered intraperitoneally immediately prior
to ischemia and repeated 2, 4, and 6 hours after terminating
ischemia. In Group 3, 10 mg/kg was administered intraperitonealiy 30 minutes after terminating ischemia and repeated
after 2,4, and 6 hours.
Seventy-two hours after 2 0 minutes of ischemia, while under
light ether anesthesia, 6 untreated rats and 6 rats from each
of the three treatment groups were infused with phosphatebuffered 10% formalin and decapitated. The brains were
removed, embedded in paraffin, sectioned, and stained with
hematoxylin and eosin.
Ischemic changes in cells were quantitated by light microscopy in the entire parietal cortex and hippocampus of both
hemispheres on a single coronal section. For data analysis,
the entire hippocampus was divided into seven regions (subiculum, medial cornu ammonis (CA)l, lateral CA1, lateral
CA3, ventral CA3, CA4, and dentate) and the entire parietal
cortex was graded as a single region. Each region was given
a score of 0 to 4 based on the percentage of cells with
shrunken eosinophilic cytoplasm and pyknotic nuclei (0 =
normal, 1 = 1 to 25%, 2 = 26 to 50%, 3 = 51 to 75%, and
4 = 76 to 100%). We used a Kruskal-Wallis one-way analysis of variance (ANOVA) for evaluating group differences
between all eight regions in treatment versus control animals.
Subsequent analyses of individual group pairs were performed to identify the specific regional location of differences between treatment groups.
Five rats given 10 mg/kg of CGS-19755 30 minutes after
terminating 20 minutes of ischemia (Group 3) were tested
on an eight-arm radial maze as previously described { 16, 191,
and compared to 4 untreated, sham operated control rats and
5 untreated ischemic control rats. Testing was begun 1
month after ischemia. All rats were given 1 daily trial for 50
days, but the first 20 trials were considered a conditioning
period. Five of eight arms were baited and the number of
working errors (choosing the same arm twice) were recorded. The data were analyzed by a two-way repeated
ANOVA. After 50 trials, animals were sacrificed and hippocampi examined by light microscopy as previously described.
Calcium-CaM Binding
Ionic calcium binds primarily with intracellular CaM before
activating target enzymes such as kinases, proteases, and
phospholipases. We used an immunohistochemical assay of
calcium-CaM binding in order to study indirectly the distribution and time course of increases in free ionic calcium
levels in neurons and to measure directly its subsequent
binding to CaM [51. We used a labeled sheep anti-CaM
antibody (CaM-Ab) which recognizes only CaM unbound to
calcium or target proteins. In a previous study, we found that
decreased staining of brain sections by CaM-Ab after ischemia was due to calcium-CaM binding and not to changes in
the quantity of CaM or CaM target proteins [ 5 ] . While most
brain regions demonstrated increased calcium-CaM binding
immediately after ischemia before there was any lightmicroscopic evidence of cellular damage, this binding returned to normal in brain regions resistant to ischemic damage such as the cortex and dentate, but persisted beyond 24
hours in selectively vulnerable CA1 and partially vulnerable
CA3 which were destined to undergo irreversible damage in
this model [ 5 ] .
Animals were sacrificed immediately and 2 hours after 30
minutes of ischemia, and 24 hours after 20 minutes of ischemia. Brains were hand perfused and fixed with 4% paraformaldehyde in situ, and 40-pm-thick sections through the
Grotta et al: CGS 19755 for Cerebral Ischemia
Physiological Variables during Ischemia (Mean for Each Group)
Ischemic control rats
CGS-19755-treated rats
30 mglh before ischemia
10 mglkg before ischemia
10 mg/kg after ischemia
Hematocrit Level
(mmHg) (mmHg) (96)
(mmoVliter) (mm Hg)
(mm Hg)
Temperature Temperature
mean arterial blood pressure; PcoZ = carbon dioxide pressure; Poz
hippocampus were incubated in Cam-Ab. A diaminobenzadene-labeled anti-sheep secondary antibody was added to
stam the Cam-Ab. Staining in the endal limb of the dentate,
dorsal CA1, CA3, and parietal cortex was graded on a 4point scale (0 = no staining; 1 = minimal staining; 2 =
some staining; 3 = extensive staining, but normal neuronal
soma not distinguishable; 4 = extensive staining of normal
neuronal soma). These regions were selected because in our
previous studies they demonstrated changes in calcium-CaM
binding which correlated with histological damage.
We compared untreated ischemic control rats (n = 15)
with rats that received different dose regimens of CGS19755: 10 mg/kg begun immediately prior to ischemia and
repeated %, 2%, and 4?h hours after the completion of ischemia (n = 16); and 10 mg/kg begun % hour after ischemia
and repeated 1*/2,4*/2,and 6% hours after ischemia (n = 12).
These dosing schemes were slightly altered from those used
for the histological and learning studies in order to allow the
administration of at least two doses before the 2-hour postischemic immunohistochemical studies. Groups of 5 to 6
control rats and preischemic treated animals were compared
at all three time intervals (immediately, 2 hours, and 24
hours after ischemia) while control animals and animals
treated after ischemia were compared only 2 hours and 24
hours after ischemia. Statistical analysis was performed using
the Kruskal-Wallis one-way ANOVA, similar to that used
for the histological analysis.
Physiological Parameters and Mortality
There were no differences between ischemic control
animals and any of the three CGS-19755-treated
groups for any of the physiological variables during
ischemia, including rectal or head temperature, arterial
blood gases, hematocrit, and glucose levels (Table). All
animals demonstrated a substantial increase in mean
arterial blood pressure during ischemia, but this increase was comparable in all treatment groups and control animals.
Mortality during the first 72 hours after ischemia
was significantly increased in both groups treated with
CGS-19755 prior to ischemia (Groups I and 2) compared to ischemic control rats, but mortality of animals
begun on CGS-19755 after ischemia (Group 3 ) was no
different than that of control animals (Fig 1). Pre614 Annals of Neurology Vol 27 No 6 June 1990
oxygen pressure.
60 50
40 -
p = ,025
p = .01
Compared to untreated ischemia
Fig I . Mortality within 72 hours after ischemia in untreated
control animals compared to the three groups treated with CGS1975.5. The ordinate represents % mortality.
ischemic treatment with CGS-19755 clearly resulted in
sedation which usually occurred after the second or
third dose (i.e., 2 to 4 hours after ischemia) and lasted
for about 6 hours after the final dose (i.e., until about
12 hours after ischemia). Most of the excessive deaths
of CGS- 19755-treated animals were due to respiratory depression occurring during this period. Animals
begun on CGS-19755 after ischemia were usually less
2 tountreated
12 hours after
had no during
ratory difficulty and were indistinguishable from control animals by the second day.
Histological Findings
In this model, ischemic neuronal changes in untreated
animals as detected by light microscopy become progressively more evident for up to 72 hours of reperfusion after ischemia, and are most prominent in the
CA1 and subiculum regions of the hippocampus. The
cortex and regions of CA3 adjacent to CA1 (called
lateral CA3 in the present study) are less severely
damaged 116, 18, 201. Animals that received 30 mg/kg
of CGS-19755 prior to ischemia (Group 1) and that
survived for 72 hours after ischemia had better scores
in all eight brain regions examined, compared to untreated ischemic control rats. These differences were
significant in the subiculum (F = 5.8, 0.01 < p <
0.05) and the lateral CA1 (F = 4.3, p = 0.05) (Fig
2A). No protection was seen in any region in the
Group 2 animals treated with CGS-19755, 10 mg/kg,
begun prior to ischemia (Fig 2B). In animals treated
with 10 mg/kg of CGS-19755 begun after ischemia
(Group 3), there was no protection in the CA1, but
less damage was found in the cortex (F = 4.6, p =
0.05). Treated animals had worse damage in the lateral
CA3 (F = 5.3, 0.01 < p < 0.05) (Fig 2C).
4 1
Calcium-CaM Binding
CaM staining in CGS-19755-treated ischemic rats was
significantly greater (and therefore calcium-CaM binding was less) than in untreated ischemic animals. In the
group started on CGS-17755 prior to ischemia, calcium-CaM binding was significantly less than in control
animals at all three time intervals after ischemia when
tested across all four brain regions examined (immediate postischemia: F = 8.7, p = 0.01; 2 hours postischemia: F = 7.0, p < 0.05; 24 hours postischemia: F
= 8.3, p = 0.01). Blockade of calcium-CaM binding
was most prominent in CA1 and CA3 immediately @
< 0.001 and p < 0.005, respectively) and 24 hours (p
< 0.005 in CA3) after ischemia (Fig 4A). No
significant blockade was seen in the dentate, and only a
transient mild effect was seen in cortex 2 hours after
;&emia @ = 0.05).
Working Memory
This analysis was confined to animals treated after ischemia only (Group 3) because long-term survival was
needed for memory testing and preischemic treatment
with CGS-19755 (Groups 1 and 2) caused increased
mortality. Animals treated with 10 mg/kg of CGS1975 5 begun after ischemia made significantly fewer
working errors than did untreated ischemic control animals after the 20-day conditioning period (F = 7.5,
0.01 < p < 0.05) (Fig 3). There was no difference
in hippocampal damage between CGS-1775 5-treated
and ischemic control rats at the conclusion of the learning studies, both groups showing moderate to severe
gliosis in all animals.
p s
Fig 2. Grading (mean -+ SEM) of histological damage, demonstrated by light microscopy, in the cortex and seven hippacampal
regions in ischemic control animals compared to (A) animals
given CGS-I 9755, 30 mglkg, immediately prior to ischemia;
(B)animals given CGS-I 9755, I0 mglkg, immediately prior to
ischemia; (C) animals given CGS-19755, 10 mglkg, 30 minutes following ischemia. See text for scoring of histological
Grotta et al: CGS 19755 for Cerebral Ischemia 615
IwhemicConlrol n.5
0 soam
lomphgprlirchemia n-5
.r 3
05 -
Fig 3. Working ewors (mean % SEM) in ischemic control and
sham surgery groups, and in animals treated with CGS-19755,
10 mglkg, 30 minutes following ischemia. After a conditioning
period of 20 trials, the CGS-1 9755-treated animals made
signifcantty fewer errors than did ischemic control animals
(0.01 < p < 0.05).
In the group treated only after ischemia, calciumCaM binding was significantly less than in control animals 2 hours after ischemia when tested across all four
brain regions examined (F = 10.1, p = 0.01), but this
effect was gone by 24 hours. Again, blockade of calcium-CaM binding was most prominent in the CA1
and CA3 2 hours (p < 0.01 andp < 0.05, respectively)
and 24 hours (p < 0.05 in CA3) after ischemia (Fig
4B). Transient reduction of calcium-CaM binding was
also found in the dentate 2 hours after ischemia (p =
0.01). N o significant blockade was seen in the cortex.
The main finding in this study is that systemic administration of the competitive NMDA receptor antagonist
CGS-1975 5 had some beneficial effect by histological
protection and preserved learning ability in a global
ischemia model. Furthermore, even when administration was delayed until 30 minutes after ischemia, the
drug strikingly reduced calcium-CaM binding in severely ischemic brain regions, indicating effective inactivation of the NMDA-associated ion channel. The latter finding attests to the availability of CGS-19755 at
neuronal NMDA receptors after systemic administration.
The neuronal protection provided by CGS-19755
was at best modest, and was probably limited by the
sedative effects of the drug which limited the doses
that could be used in this model. Preischemic administration of the drug resulted in protection of hippocampal neurons, but also in high mortality from respiratory
depression. Respiratory depression did not occur during ischemia and therefore did not augment the standard ischemic insult since blood gases during the period of carotid occlusion were no different in treated
0CGS 19755 pretreated
* p5.05
** p 5 ,005
p 5 ,001
. .
r 3
$ 1
I=I CGS 19755 post treated
p 5 05
* * P 5 01
Fig 4. Grading of calmodulin (CaM) staining by an anti-CaM
antibody (which recognizes only CaM unbound to calcium ions
and target protein) in the C A I , CA3, dentate, and cortex immediately after ischemia, and after 2 and 24 hours of repe.f.sion (mean ? SEM). Results in ischemic controls and animals
treated with CGS-19755, 10 mglkg, immediately before (A)or
30 minutes after (B) ischemia, and repeated every 1 to 2 hours.
See text for explanation of grading.
compared to control animals, and treated animals usually awakened after removal of the carotid clips. In all
CGS-19755-treated animals, sedation occurred after
several doses had been received, indicating a cumulative effect of the drug. The plasma half-life of CGS19755 is approximately 30 minutes, but the half-life
in brain tissue could be considerably longer (CIBAGEIGY Corporation, unpublished data). A previous
study using similar doses of this drug in gerbils did not
report data on mortality or side effects [14].The sedative properties of CGS-19755 seen in our study may
have been due to generalized cortical andor brainstem
depression since glutamate receptors are widespread
throughout the brain [2].
Successful production of a standard ischemic insult
with this model depends on inclusion only of animals
that lose their righting reflex and have no cortical EEG
activity throughout the ischemic period {16, 181.
Could the sedative properties of CGS-19755 have led
to EEG suppression and loss of righting reflex with
lesser degrees of ischemia, and could this explain the
protection found with preischemic administration of
CGS-19755? We think that this is unlikely since arterial blood gases in treated animals did not differ from
those in control animals during ischemia, and most
treated animals awoke immediately after release of the
carotid clips, indicating that the drug was not causing
appreciable sedation at the time of ischemia. Furthermore, our observations in all three treatment groups
indicated that sedation occurred only after repetitive
doses. Finally, some protection was seen even when
administration was delayed until after the ischemic period was completed.
We observed no side effects of CGS-19755 other
than sedation. There were no effects of the drug on
physiological parameters during ischemia. As expected
in unanesthetized animals, blood pressure rose in both
treated and control groups during ischemia. In anesthetized animals during ischemia, MK-80 1 results
in lowering of blood pressure {77, and this might
also o c c u with CGS-19755. Although CGS-19755treated animals remained sedated for several hours after the completion of dosing, we found no evidence of
long-term effects on behavior. In fact, in animals
treated after ischemia, signhcantly better learning ability was observed, compared to untreated ischemic control animals. Of course, more subtle but important behavioral side effects could not be evaluated in this
small animal model.
The improved learning ability found with postischemic administration of CGS-1975 5 was impressive,
and was similar in magnitude to what we found in the
same model with nicardipine, a dihydropyridine antagonist of the calcium L channel {lb], and significantly
better than learning ability after treatment with baclofen, an inhibitor of presynaptic glutamate release { 177.
Working errors are a validated measurement of outcome and the most sensitive test of memory dysfunction in this model El6, 191. Preservation of working
memory has been observed in the absence of hippocampal histological protection 115, 161 and implies
that nonhippocampal integrative circuits are involved
in these learning tasks.
Preischemic administration of CGS-197 55 resulted
in significant but limited protection of CA1 neurons.
In this model, 20 minutes of ischemia produces severe
damage to the CA1 and moderate damage in other
hippocampal regions and cortex. Blood flow falls so
low in CA1, and the consequent ischemic insult is so
severe {lb, IS}, that it should be difficult to demon-
strate neuronal protection by any pharmacotherapy.
Previously, we were unable to find protection of severely affected CA1 neurons with either nicardipine or
baclofen 115-177, so we were impressed by these results with CGS-19755. Because the number of animals
(n = 6) in each group was small, power analysis based
on data in CA1 reveals that the chance of a type I1
error in our analysis was large, ranging from 20 to
70% in various CA1 regions with an a level of 0.05.
With the variability found in our measurements, we
would need 3 to 11 animals in each group to detect a
25 % reduction in light-microscopic damage in various
CA1 regions with 80% certainty at a = 0.05.
Because of the high mortality found with preischemic administration of CGS-19755, our positive results must be interpreted with caution. We think that
excess mortality was caused by respiratory depression
from CGS-197 55, and had those animals received ventilatory support and survived, they also might have
demonstrated hippocampal neuronal protection. However, it is certainly possible that those animals died in
part because they had more severe ischemic cerebral
damage, which would have nullified the beneficial results found in surviving animals. Additional studies of
animals dosed prior to ischemia and receiving ventilatory support will be needed to clarify this point.
We found only slight histological protection in cortical neurons with postischemic administration of
CGS-19755. Other investigators have reported more
substantial neuronal protection with postischemic administration { 141 especially when treatment was begun
within minutes after ischemia { 111. It is possible that
we began treatment too late after ischemia or used too
low a dose, but we found that doses that were of
greater magnitude or were given closer to the ischemic
period resulted in higher mortality. It is also interesting that postischemic dosing resulted in protection
of cortical neurons only. The posterior region of the
cerebral cortex is the region where MK-801 has its
greatest effect {7, 91. Since relatively little cortical
damage is produced in our model, it might be difficult
to discover more marked protection in that region.
The most striking finding in our study was that after
either preischemic or postischemic administration of
CGS- 19755, calcium-CaM binding was reduced for up
to 24 hours after ischemia. This suggests that a substantial amount of calcium enters neurons and binds to
CaM via the NMDA ion channel, and that this flux can
be partially prevented by a systemically administered
NMDA receptor antagonist begun up to 30 minutes after ischemia. In a simultaneous study, we found
that nicardipine, a dihydropyridine calcium L channel
blocker, resulted in only a transient reduction of
calcium-CaM bindmg in severely ischemic hippocampal
neurons 1217. CGS-19755 begun prior to ischemia
seemed to render protection from calcium entry for up
Grotta et al: CGS 19755 for Cerebral Ischemia 617
to 24 hours after ischemia, while administration begun
30 minutes after ischemia resulted in calcium blockade
during the period of drug administration ( 2 hours after
ischemia) but loss of effect by 24 hours. In general, the
distribution of histological protection found when
CGS-19755 was begun prior to ischemia corresponded to the distribution of blockade of calciumCaM binding found with the same dosing regimen of
CGS-19755. It is likely, therefore, that the histological
protection afforded by CGS-197 55 occurs in regions
where there is prominent calcium entry into neurons
which is significantly blocked for up to 24 hours.
Other mechanisms may also play a role, but studies to
date indicate that various glutamate antagonists do not
act by increasing cerebral blood flow E22, 231.
These findings may explain why CGS-19755 did not
result in more striking histological protection of neurons. One possibility is that the protective effect seen
with antagonism of excitatory amino acid receptors is
due in part to cerebral hypothermia E24, 251 and that
we found little protection because head temperature
was kept above 36°C. However, this explanation does
not account for the significant protection we found in
pretreated animals compared to control animals even
though head temperature was the same in both groups.
We believe that more striking histological protection
was not found in our study because to be effective
CGS-19755 must permanently prevent substantial calcium influx into neurons. In this model, persistent
marked calcium-CaM binding in control animals occurs only in the CA1 and CA3, and calcium-CaM
binding was not totally or permanently prevented in
these regions by either preischemic or postischemic
administration of CGS-19755. Intraneuronal ionic calcium levels can increase through voltage-operated
channels and by release of intracellular stores via second messenger pathways, and these mechanisms
would not be inhibited by CGS-19755. Generally, histological protection seemed to correlate with reduction
of calcium-CaM binding 24 hours after ischemia. For
example, 24 hours after ischemia in the animals treated
before ischemia, staining in protected CA1 was graded
1.6, compared to 0.2 in ischemic control animals and
4.0 in normal animals. Histological protection may
have been only modest because this represents only
partial reduction of calcium-CaM binding. Regions
such as the dentate and cortex, which were not protected, had less calcium-CaM binding in ischemic control animals, and there was no difference between control and treated animals 24 hours after ischemia. In
postischemic treated animals, which had little histological protection in any region, there was transient reduction of calcium-CaM binding 2 hours after ischemia
but no significant difference 24 hours after ischemia
compared to untreated animals, indicating that more
prolonged administration may be needed if treatment
is withheld until after ischemia.
618 Annals of Neurology
Vol 27
No 6 June 1990
We conclude that CGS-19755 may be an effective
therapy for very acute cerebral ischemia but recommend further study in a model in which more complex
behavioral side effects can be recognized. The optimal
dose and duration of therapy remains to be determined, but very early and prolonged treatment may be
This study was funded by National Institutes of Health, National
Institute of Neurological and Communicative Disorders and Stroke
grant NS 23979. CGS-19755 was supplied by the CIBA-GEIGY
Corporation, Summit, NJ.
1. Choi DW. Glutamate neurotoxicity in cortical cell culture is
calcium-dependent. Neurosci Lett 1985;58:293-297
2. Rothman SM, OlneyJW. Glutamate and the pathophysiology of
hypoxic-ischemic brain damage. Ann Neurol 1986;19:105-111
3. Goldberg MP, Giffard RG, Kurth MC, Choi DW. Role of extracellular calcium and magnesium in ischemic neuronal injury
in vitro. Neurology 1989;39(suppl 1):217 (Abstract)
4. Kurth MC, Weiss JH, Choi DW. Relationship between glutamate-induced *%a+ + influx and resultant neuronal injury in
cultured cortical neurons. Neurology 1989;39(suppl 1):217
5. Picone CM, Grotta JC, Earls R, et al. Immunohistochemical
determination of calcium-calmoddin binding predicts neuronal
damage after global ischemia. J Cereb Blood Flow Metab
1989;9:805-81 I
6. Simon RP, Swan JH, Griffiths T, Meldrum BS. Blockade of
NMDA receptors may protect against ischemic damage in the
brain. Science 1984;226:850-852
7. Park CK, Nehls DG, Graham DI, et al. The glutamate antagonist MK-801 reduces focal ischemic brain damage in the rat.
Ann Neurol 1988;24:543-551
8. Kochhar A, Zivin JA, Lyden PD, Mazzarella V. Glutamate antagonist therapy reduces neurologic deficits produced by focal
central nervous system ischemia. Arch Neurol 1988;45:148153
9. Park CK, Nehls DG, Graham DI, et al. Focal cerebral ischaemia
in the cat: treatment with the glutamate antagonist MK-801
after induction of ischaemia. J Cereb Blood Flow Metab 1988;
10. Gill R, Foster AC, Woodruff GN. Neuroprotective actions of
MK-801 in rat global ischaemia model. J Cereb Blood Flow
Metab 1989;9(suppl 1):S629 (Abstract)
11. Swan JH, Meldnun BS. Early, not late, administration of
NMDA receptor antagonists protects against selective ischaemic cell loss. J Cereb Blood Flow Metab 1989;9(suppl 1):
SJ58 (Abstract)
12. Lanier WL, Perkins WJ, Karlsson BR, et al. Effect of the excitatory amino acid antagonist MK-801 on cerebral injury following complete ischemia in primates. J Cereb Blood Flow Metab
1989;9(suppl 1):S744 (Abstract)
13. Lehmann JC, Hurchison AJ, McPherson SE, et al. CGS 19755,
a selective and competitive NMDA-type excitatory amino acid
receptor antagonist. J Pharmacol Exp Ther 1988;246:65-75
14. Boast CA, Gerhardt SC, Pastor G, et al. The N-methyl-Daspartate antagonists CGS 19755 and CPP reduce ischemic
brain damage in gerbils. Brain Res 1988;442:345-348
15. Grotta J, Spydell J, Pettigrew C, et al. The effect of nicardipine
on neuronal function following ischemia. Stroke 1986;17:2132 19
16. Grotta JC, Pettigrew LC, Rosenbaum D, et al. Efficacy and
mechanism of action of a calcium channel blocker after global
cerebral ischemia in rats. Stroke 1988;19:447-454
17. Rosenbaum DM, Grotta JC, Pettigrew LC, et al. Baclofen does
not protect against cerebral ischemia. Stroke 1990;21:138-140
18. Pulsinelli WA, Brierley JB. A new model of bilateral hemispheric ischemia in the unanesthetized rat. Stroke 1979;lO:
19. Volpe BT, Pulsinelli WA, Tribuna J, Davis HP. Behavioural
performance of rats following transient forebrain ischemia.
Stroke 1984;15:558-562
20. Pulsinelli WA, Brierley JB, Plum F. Temporal profile of
neuronal damage in a model of transient forebrain ischemia.
Ann Neurol 1982;11:491-498
21. Picone CM, Grotta JC, Earls R, et al. Calcium calmodulin bind-
ing after calcium channel blocker therapy. Stroke 1990 (in
Nehls DG, Park CK, McCulloch J. Cerebral circulatory effects
of Dizocilipine (MK-801) in the rat. J Cereb Blood Flow Metab
1989;9(suppl 1):S376 (Abstract)
Hardebo JE, Wieloch T, Kahrstrom J. Excitatory amino acids
and cerebrovascular tone. J Cereb Blood Flow Metab 1989;
9(suppl 1):S441 (Abstract)
Minamisawa H, Smith ML, Siesjo BK. The effect of moderate
hyperthermia (39°C) and hypothermia (35°C) on brain damage
following 10 and 15 minutes of forebrain ischemia. J Cereb
Blood Flow Metab 1989;9(suppl 1):S265 (Abstract)
Busto R, Dietrich WD,Globus MYT, et al. Postischemic moderate hypothermia inhibits CA hippocampal ischemic neuronal
injury. J Cereb Blood Flow Metab 1989;9(suppl 1):S266 (Abstract)
Grotta et al: CGS 19755 for Cerebral Ischemia 619
Без категории
Размер файла
786 Кб
nmda, global, competition, improve, reduced, cerebral, cgs, outcomes, ischemia, antagonisms, calcium, receptov, binding, calmodulin, 19755
Пожаловаться на содержимое документа