The measurement of cortical to 3 (peripheral-type benzodiazepine binding) site densities provides an accurate index for the detection and quantification of ischaemic brain lesions following middle cerebral artery occlusion (MCAO) in mice 6. Here, we have used this marker to assess the neuroprotective activity of potential anti-ischaemic drugs belonging to several chemical classes. In untreated mice, the mean infarcted volume measured 96 h after unilateral coagulation of the middle cerebral artery was 27.9 _+ 4.3 mm 3 (17.5% of the hemisphere volume) and ~o 3 site densities (measured by incubation with 3H-PK 11195) were increased by 107.3 _+ 4.8% (cortical homogenates) or by 81% (coronal brain sections). The administration of the anti-ischaemic agent SL 82.0715 (10 mg/kg i.p.), 5 min, 6 h and 18 h after the occlusion and then twice daily until sacrifice evoked a decrease of similar magnitude (ca. 60-70%) in the volume of the infarction and in the proliferation of to 3 sites. The constant tissue sparing effect of SL 82.0715 allowed the examination of the window of therapeutic opportunity. A significant diminution of cortical w 3 sites was still noted when the first administration was delayed until 3 h post-occlusion. Moreover, the protective effect of SL 82.0715 was enhanced by repeated treatment for the first 36 h but not thereafter. Based on the histological, autoradiographic and homogenate binding results obtained with SL 82.0715, we studied the protective effects of several competitive and non-competitive NMDA receptor antagonists. When administered according to the above-described standard protocol, these drugs reduced o~ 3 site levels in cortical homogenates from MCAO mice in a dose-dependent manner. The dose preventing by 50% the increase in ~o 3 site levels (in mg/kg i.p.) and the maximal inhibition were respectively: MK-801 (0.2, 93%); TCP (1.6, 66%); kynurenate (260, 58%); ifenprodil (7.0, 58%); SL 82.0715 (1.1, 72%); CGS 19755 (46% at 10 mg/kg); dextromethorphan (46% at 30 mg/kg). In contrast, agents acting preferentially upon sigma (a) opiate receptors ((+)-3PPP, 1-10 mg/kg i.p. and haloperidol, 0.3-3 mg/kg i.p.) did not provide a significant protection. In general, calcium channel blockers (nimodipine, flunarizine, verapamil, perhexiline, diltiazem) were devoid of a clear neuroprotective potential when administered at non-toxic doses after the coagulation of the middle cerebral artery. Diltiazem (3 and 10 mg/kg i.p.) provided a significant protection when the first administration was performed 10 min prior to the occlusion. Limited protection was observed with adenosine A~ receptor agonists (N~-cyclohexyladenosine and 2-chloro-adenosine). No protection was afforded by the platelet activating factor (PAF) receptor antagonists, brotizolam and alprazolam at the dose of 10 mg/kg or by a Gingko biloba extract at 30 mg/kg. The anti-inflammatory agent, dexamethasone, and the minor tranquillizer, clonazepam, also failed to alter a~ 3 site densities when administered at the dose of 10 mg/kg. Some putative cerebroprotective compounds also prevented the increases in to 3 site densities in MCAO mice with 50% effective doses (in mg/kg i.p.) and maximal effects as follows: vincamine (0.2, 74%); cyclandelate (15, 63%) and piribedil (4, 66%) although many others were inactive (eg. naftidrofuryl, pyrithioxine, papaverine). In conclusion, the results with both NMDA receptor-channel blockers (MK-801, TCP) and atypical non-competitive NMDA receptor antagonists (ifenprodil, SL 82.0715) suggest that NMDA antagonists constitute the first coherent pharmacological class for the treatment of focal cerebral ischaemia. The mouse model in which tissue loss and sparing are indirectly assessed by the quantification of oJ 3 sites further allows not only the rapid evaluation of putative neuroprotective agents but possibly also the determination of an optimal treatment regimen for active compounds.
INTRODUCTION
The most valid parameter for the quantification of the
extent of cerebral ischaemic damage is the measurement
of total neuronal loss. As such techniques are onerous
and t ime-consuming many alternative approaches have
been proposed to assess the cerebroprotective potential
of drugs in experimental brain ischaemia. However,
many of these approaches rely on the measurement of
indirect and possibly invalid indices of neuronal death
such as cerebral blood flow, water content or concentra-
tions of glycolytic metabolites 46.
It is now well established that the neuronal loss
subsequent to a cerebral ischaemic insult is followed by
a dramatic increase in the density of 0) 3 (peripheral-type
benzodiazepine 45) binding sites 6'z4. This increase is a
reflection of the glial reaction and macrophage invasion
of the ischaemic brain tissue. These cells are richly
endowed with 0)3 sites when compared to neurones 73.
The accuracy and reliability of this indirect marker of
neuronal loss have been demonstrated in several models
of focal ischaemia, excitotoxic lesions and in stroke-prone
spontaneously hypertensive rats 6'24, as well as in human
proliferative, demyelinat ing and ischaemic disorders 7.
We have developed and discussed in our companion
paper 6 a model of focal ischaemia induced by electro-
coagulation of the middle cerebral artery in the mouse.
This procedure results in a reproducible ischaemic lesion
of the temporoparietal cortex and, in some animals, of
the dorsolateral striatum which is accompanied by an
increase in (93 site densities mainly in the interface between the infarcted and the healthy tissue. Moreover, the measurement of (93 site densities in cortical homog-
enates demonstrated that binding densities are maximally
elevated between days 4-8 post-occlusion 6. The first objective of the present investigation was to
determine if a reduction, by pharmacological means, of the volume of the infarcted tissue is reflected by a
decrease in (93 site densities in middle cerebral artery occluded (MCAO) mice. To this end, we have compared the effect of the anti-ischaemic compound, SL 82.0715 ((+)-a-(4-chlorophenyl)-4-[(4-fluorophenyl)methyl]-l-pi- peridine ethanol; a novel N M D A antagonist15"16), on the
consolidated volume of infarction with the effect of the
same drug treatment on (93 binding site densities mea- sured autoradiographically or in brain homogenates. SL
82.0715 was employed because this compound exerts
considerable neuroprotection in rat and cat models of focal cerebral ischaemia 35.
The second objective of our study was to evaluate, by
the measurement of (93 site levels in cortical homogenates of occluded mice, the cerebroprotective activity of drugs
(and drug classes) potentially able to protect against the ischaemia-induced tissue damage. Recently and based
largely on cell culture studies, a number of therapeutic
rationales for the mitigation or prevention of ischaemic neuronal damage have been proposed18'3a'32; the devel-
opment of the mouse model has greatly facilitated the
evaluation of these theoretical approaches in vivo. It is now well recognized that an influx of extracellular
calcium with subsequent intracellular overload is a major pathogenic mechanism in ischaemic neuronal death 18.
This cellular accumulation of calcium results from the
activation of receptor- and possibly voltage-operated
calcium channels. Indeed, an excessive activity of the N M D A receptor (coupled to a cationic channel highly permeable to calcium 48) has been strongly implicated in
neuronal death in chronic and acute neuropathological processes 38. Accordingly, we have initially focused our
interest on drugs that block voltage-operated and N M D A
receptor-operated calcium channels. We have also eval- uated the neuroprotective potential of compounds that inhibit other (non-NMDA) excitatory amino acid recep-
tors, of 0 receptor ligands and of some adenosine receptor agonists that may inhibit the liberation of excitatory amino acids via activation of presynaptic adenosine A 1 receptors 22,59. To fully characterize the
pharmacology of this focal ischaemia model, we have also examined the effects of several compounds with proposed anti-ischaemic activity but with poorly defined mecha- nisms of action (e.g. vincamine and cyclandelate), or of
291
debatable clinical utility (e.g. clonazepam and dexameth-
asone). Finally, several putative PAF antagonists were
also tested. The prognosis of cerebrovascular accident could be
greatly improved if drugs were unequivocally able to
reduce the ischaemic damage when given after the ictal
event. Thus, the third objective of this work was to determine the time period, after the middle cerebral
artery occlusion, during which pharmacological interven- tion remains effective in order to establish the window of
therapeutic opportunity in this focal ischaemia model. Overall, the present study demonstrates that the
prevention of the increase in brain (93 site densities in M C A O mice provides an accurate and sensitive estimate
of the anti-ischaemic potential of drugs. As the surgical
procedure and the quantification of (93 binding site
densities are much less time consuming than those previously employed, such as the histological determina-
tion of cerebroprotective activity in experimental focal
ischaemia in the rat or in the cat, this novel approach is well adapted to the initial identification and character-
ization of putative cerebral anti-ischaemic and neuropro-
tective agents.
MATERIALS AND METHODS
All experiments were performed on 4-week old, male Swiss mice (OF 1, Iffa Credo, France) with body weights between 18 and 22 g at the time of surgery. The animals were fed standard laboratory chow (AO4, UAR, France) and were not starved prior to anaesthesia. The animal house was under a 12 h light/12 h dark cycle (lights on: 07.00 h) and maintained at 22 + 0.5 °C and 50% humidity.
Middle cerebral artery coagulation Mice were anaesthetized with chloral hydrate (500 mg/kg i.p.).
The exposure of the left middle cerebral artery was based on the description by Welsh et al. TM. The temporoparietal region was shaved and a skin incision was made between the lateral aspect of the orbit and the external acoustic meatus. The superior pole of the parotid gland was reflected downwards as was the temporalis after partial resection of its cranial insertion. The distal course of the middle cerebral artery was then visible through the translucent skull.
Under a low-power magnifying glass (×4), a burr-hole craniect- omy was performed with a dental drill and then enlarged with fine synovectomy rongeurs. The middle cerebral artery was coagulated by bipolar diathermy. The temporalis and parotid gland were replaced, the incision was lightly dusted with virginiamycin (Staphy- lomycin, Smith Kline & French), the scalp was sutured and a 2% solution of iodine in alcohol was topically applied. Sham-operated mice were subjected to simple exposure of the middle cerebral artery. After surgery, the animals were maintained in a warmed environment by using a heating blanket and a lamp. In these conditions, rectal temperature was maintained within physiological limits (37.5 + 1.0 °C). After recovery from anaesthesia (about 1 h), mice were housed 5 per cage with free access to food and water.
Measurement o f to 3 site densities Our previous studies 6 have demonstrated that coagulation of the
mouse middle cerebral artery induces a marked enhancement of the total cortical % site densities which is near-maximal 4 days after ischaemia and thereafter remains constant. Therefore, in the
292
present investigation, all determinations of 0) 3 site densities were performed 96 h after occlusion or sham-operation. Mice were decapitated under haiothane anaesthesia, and both ipsilateral and contralateral cortices were dissected in their entirety, weighed, frozen on dry ice and stored at -80 °C until binding assay. Cortices were Polytron homogenized (setting 6 for 30 s) in 10 vols. of a 120 mM NaCI, 50 mM Tris-HC1 (pH = 7.4) buffer. Aliquots of 50ktl of homogenate were incubated with 1 nM 3H-PK 11195 (75 Ci/mmol, NEN, F.R.G.) in 1 ml (final volume) of Tris-saline buffer for 1 h at 25 °C. The incubate was decanted on ground glass filtres (GFB) followed by vacuum filtration; the filtres were rinsed 3 times with 3.5 ml of cold Tris-saline and the bound radioactivity was estimated by liquid scintillation spectrometry 9. Non-specific binding was defined by 1/~M PK 11195 added to duplicate samples. Binding was measured in triplicate. In some experiments, 0)3 site densities were assessed autoradiographically in coronal sections as detailed in our companion paper 6.
Calculations and statistical analyses None of the compounds significantly altered 0)3 site levels in the
ipsilateral cortex of sham-operated animals when administered according to the standard treatment schedule (data not shown).
Accordingly, unless otherwise stated, the drug-induced percentage changes in 0)3 site densities were calculated as follows:
A0)3% = {(1-(treated MCAO-sham)/(vehicle MCAO-sham)} x 100
The calculation was based on the mean values for w3 site densities obtained at each dose of the agent investigated. The statistical significance of the lesion-induced changes in 0)3 site levels was performed using the one-way analysis of variance followed by the Student's t-test or Dunnett's test for multiple comparison or the Cochran test (for heterogeneous variances).
Treatment schedule Unless otherwise stated, drugs were injected i.p. in 0.2 ml saline.
When an additional solvent was necessary, appropriate controls with the solvent were carried out. The usual administration schedule consisted in the administration of the tested drug 5 min, 6 h, 18 h, 24 h, 36 h, 48 h, 60 h, 72 h and 84 h after surgery.
Fig. 1. Chemical structure of compounds with cerebroprotective activity in MCAO mice.
293
hexiline, alprazolam, brotizolam, vincamine, nicergoline, pyrithio- xine, naftidrofuryl, piribedil, clonazepam, haloperidol and cyclan- delate were synthetized by the Chemistry Dept., SyntMlabo Recherche (L.E.R.S.), Paris. TCP, MK-801 and CGS 19755 were gifts of Dr. Chicheportiche and Kamenka (Ecole Sup6rieure de Chimie, Montpellier, France), Dr. Woodruff (Merck Sharp and Dohme, Harlow, England and Ciba-Geigy (Summit, N J)), respec- tively. MLV 5860, (+)- and (-)-3PPP, nimodipine and the Gingko biloba extract were gifts from Dr. Shinagaki (The Tokyo Metro- politan Institute of Medical Science, Japan), Astra (Sfdert/ilje, Sweden), Bayer (FRG) and Dr. Etienne (Institut Henri Beaufour, Le Plessis-Robinson, France), respectively. The other drugs were purchased from Sigma Chemistry Co. (St-Louis, MO, U,S.A.).
TCP, MK-801, ifenprodil hydrochloride, CGS 19755, dextrome- thorphan, MLV 5860, (+)- and (-)-3PPP, diltiazem, flunarizine hydrochloride, verapamil, theophylline, 2-chloroadenosine, nicer- goline, naftidrofuryl oxalate and papaverine were dissolved in water. SL 82.0715, riluzole, haloperidol, perhexiline, Gingko biloba extract and vincamine were dissolved in methanesulphonic acid and kynurenate was solubilized in diluted NaOH. Nimodipine, N 6- cyclohexyladenosine, alprazolam, brotizolam, pyrithioxine, clona- zepam, cyclandelate and dexamethasone were suspended by soni- cation in a 0.2% Tween 80 solution. The nimodipine suspension was kept in the dark. Solutions or suspensions were neutralized with either HCI or NaOH.
The chemical structures of representative drugs are shown in Fig. 1.
RESULTS
Relationship between the infarcted vo lume and o) 3 site
levels
As reported in the companion paper 6, middle cerebral
artery occlusion in mice does not result in any significant
mortality (approximately 1% of all operated mice died)
or atypical behaviour. When mice were sacrificed 96 h
after the occlusion, a reproducible increase in 0) 3 site
levels in ipsilateral cortex homogenates was observed.
The increase in binding densities ranged from 76% to
dent experiments as compared to control mice which
displayed o93 site levels ranging from 18.4 to 27.0 pmol/g
wet tissue. Simple exposure of the middle cerebral artery
(sham-operated mice) evoked a slight increase (2-29%)
in o93 site levels (Table I).
The validity of o)3 site levels as an index of the
protective activity of anti-ischaemic agents has been
examined by comparing the effects of the anti-ischaemic
compound, SL 82.0715, on the infarcted volume 6 and on
°)3 site densities. When administered i.p. at a dose of 10
mg/kg following a standard post-occlusion schedule (see
Materials and Methods), SL 82.0715 reduced the infarc-
ted surface of coronal sections at all rostrocaudal planes
(Figs. 2 and 3). The infarcted volume (computed from
the rostrocaudal integration of the infarcted surfaces) was
reduced from 27.9 + 4.6 mm 3 (17.5% of the hemispheric
volume) to 10.4 + 3.8 mm 3.
The reduction in the infarcted volume was paralleled
by a decrease in o93 site densities. In sections adjacent to
those used for stereometric reconstruction, the distribu-
tion of o)3 sites faithfully reflected the tissue-sparing
activity of SL 82.0715 (Fig. 2). Moreover, quantitative
autoradiographic studies demonstrated that the percent-
age decreases in binding densities in the whole hemi-
sphere (68%) or in the cortex (59%) were of the same
magnitude as the reductions in the infarcted volume
(Table II). In further experiments in which o)3 site levels
were measured in cortex homogenates from M C A O
mice, the reduction in (-03 site levels elicited by SL 82.0715
(10 mg/kg i.p.) was again similar to that observed
autoradiographically both in the cortex and in the whole
hemisphere (Table II). In contrast, the same dosage of SL
82.0715 failed to significantly decrease o93 site levels in
TABLE I
Comparison of cortical w 3 site densities in control (unoperated), sham-operated and lesioned groups of mice
Results are means _+ S.E.M. of data obtained in contemporaneous groups of mice (number of animals, n, in parentheses). Sham-operated and lesioned mice were sacrificed 4 days after exposure of the middle cerebral artery and the to 3 site density was determined in membrane preparations.
Experimental co 3 site density (pmol/g wet tissue) group
*P < 0.05; **P < 0.01 vs respective controls; + +P < 0.01 vs sham-operated.
294
STRIATUM
VEHICLE
SL 82.0715 10 mg/kg ip
THALAMUS
VEHICLE
SL 82.0715 10 mg/kg ip
r ¸
r
Fig. 2. Illustrative correlations between the localization of the increase in 093 site densities (left side) and the histologically defined limits of the infarction (right side) as shown in coronal sections adjacent to those used for autoradiography. The upper panel, taken at the level of the striatum, compares a vehicle- with a SL 82.0715-treated brain; the same is true for the lower panel, taken at the level of the thalamus. For autoradiography, brain sections were incubated in the presence of 1 nM 3H-PK 11195 and apposed to tritium-sensitive film. The adjacent, frozen sections were post-fixed and stained with haemalun-eosin.
295
sham-opera ted mice (data not shown).
The magni tude of the t issue-sparing effects of SL
82.0715 was reproducible ; the decreases seen in 7
independent exper iments in which binding densities were
measured in cortical homogenates were: 79, 67, 70, 62,
63, 75 and 60% (mean = 68 + 6%).
N M D A receptor antagonists
Six compounds known to antagonize N M D A receptors
were evalua ted for their potent ia l protect ive effects
against the ischaemia-induced increase in ta 3 site densities
(Table III) . Al l compounds were active al though with
widely varying potencies. The most potent was MK-801,
an N M D A receptor-channel b locker 53-s5, which induced
a dose- re la ted decrease in binding densities, significant at
a dose of 0.3 mg/kg i.p. At 1 mg/kg i .p. , the ischaemia-
induced increase in 093 binding was almost abolished.
However , this compound evoked a behavioural syn-
d rome (including s tereotypies and respira tory distress),
at doses of 0.3 mg/kg and higher, similar to that described
for the dissociative anaesthet ic agent, phencyclidine. The
phencycl idine analogue, TCP 1° also provided a significant
protec t ion at doses of 3 and 10 mg/kg i.p.
Two phenyle thanolamine derivatives, SL 82.0715 and
ifenprodil (which are antagonists at the polyamine mod-
• MCAO 10
STRIATUM O MCAO + SL 82.0715 I I (10 mg/kg i.p.)
~ s
0
Antero-posterior distance (turn)
Fig. 3. Infarcted surface areas in vehicle- and SL 82.0715-treated mice following middle cerebral artery occlusion. The surface of the ischaemic lesion was measured in haemalun-eosin stained serial sections using a Quantimet 920 Image Analysis System. Points represent the mean + S.E.M. of 5 (MCAO group) or 4 mice (MCAO + SL 82.0715 group).
ulatory site in the N M D A recep tor complexlS'16), also
afforded substantial protect ion. SL 82.0715 (at doses of
1-10 mg/kg i .p. , significantly reduced to 3 binding by at
least 50%. Ifenprodi l was inactive when the s tandard
adminis trat ion schedule was employed (da ta not shown);
TABLE II
SL 82.0715-induced decrease in the volume of the ischaemic lesions and 0)3 binding site densities (by autoradiography or in homogenates) in the brains of MCA 0 mice
SL 82.0715 (10 mg/kg i.p.) was administered 5 min, 6 h and 18 h after the occlusion and then twice a day for three days. Mice were sacrificed 96 h after surgery. In coronal sections, the volume of the infarcted region was calculated, according to the procedure described by Duverger et al. 26 from the infarcted surfaces measured by planimetry in haemalun-eosin stained coronal sections at 20 rostrocaudal levels (see Fig. 2). Adjacent sections were processed for 0) 3 site autoradiography and binding densities were measured by quantitative autoradiography. Integrated binding densities were calculated by the equation D = Z (D x S)/~,S in which D is the mean binding density, D the binding density in a given section, S the surface of that section and 27S the total surface. Contralateral hemispheric values were taken as control as they were not significantly altered by the occlusion. Values are the means + S.E.M. of data obtained on the number of mice given between parentheses. No significant differences in to 3 site densities were observed in the contralateral side between untreated and SL 82.0715 treated animals. The % binding densities in untreated MCAO animals were significantly different from controls (P < 0.001, whole hemisphere; P < 0.05, cortex). No significant difference was observed between MCAO mice treated with SL 82.0715 and controls.
Control MCA 0 MCA 0 + Difference between SL 82.0715 vehicle- and drug-
*P < 0.05 versus occluded, untreated mice. In brain homogenates, the 0) 3 binding site densities were measured in cortical homogenates. The increase in 0)3 site density elicited by middle cerebral artery occlusion was statistically significant (P < 0.01).
a Increase with respect to controls, b SL 82.0715 induced decrease compared to untreated, occluded mice.
296
however , when the initial frequency of administrat ion
was increased ( injection every 90 rain for the first 6 h),
then injected at 12 h, fol lowed by the s tandard protocol ,
a significant protect ive effect was observed at 10 mg/kg
i.p.
A protect ive effect was observed at the doses of 10 and
30 mg/kg with dex t romethorphan , a non-compet i t ive
N M D A antagonist that also possesses an affinity for the
a binding sites 8°. A significant protect ion was also
afforded by the competi t ive N M D A antagonist , CGS
197555°. This protec t ion was 46% at 10 mg/kg. Higher
dosages could not be tested because of insufficient
quant i ty of this product .
The dose - r e sponse curves allowing the interpolat ion
of the doses that reduce by 50% the ischaemia-induced
prol i fera t ion of cortical o) 3 sites are presented in Fig. 4.
The rank order of potency with the EDso values (fol-
lowing i.p. adminis t ra t ion) is as follows: MK-801, 0.2
3PPP, failed to al ter o)3 site levels in M C A O mice. The
neurolept ic compound , haloperido121, had a l imited
protect ive effect at high doses (Table IV).
Antagonists o f other excitatory amino acid (quisqualate
and kainate) receptors (Table V)
Kynurenic acid, a kainate receptor antagonist 77 that
also blocks the glycine modula tory site in the N M D A
recep tor complex 41, decreased significantly ~o 3 site levels
in M C A O mice at 260 mg/kg i.p. No significant reduction
in this pa rame te r was observed with MLV 5860, a
putat ive quisqualate recep tor antagonist 68. The non-
specific b locker of exci tatory amino acid neurotransmis- sion, ri luzole (PK 261245), was also inactive.
Adenosine receptor agonists and antagonists
Two adenosine receptor agonists, 2-chloroadenosine
and N6-cyclohexyladenosine 64, and an adenosine recep-
tor blocker , theophyl l ine did not al ter to 3 binding in the
cortex of M C A O mice (Table VI). The only significant
effect was seen with the highest dose (1 mg/kg i .p.) of
N6-cyclohexyladenosine which reduced the ischaemia-
TABLE III
Effects of NMDA receptor antagonists (administered following occlusion) on the ischaemia-induced increase in w 3 site densities in the m o u s e cor tex
Results are means + S.E.M. of data obtained on the number (n) of mice indicated. Mice were sacrificed 4 days following middle cerebral artery coagulation and to 3 site density in the entire ipsilateral cerebral cortex was determined in membrane preparations. The treatment regimen is described in Materials and Methods.
Drug n Dose to 3 Site density % Change (mg/kg (prnol/g tissue) vs vehicle i.p.)
Fig. 4. Antagonism by NMDA receptor antagonists, of the ischaemia-induced augmentation of cortical 0) 3 site densities in the mouse. The results are expressed as a percentage of corresponding groups of vehicle-treated mice similarly subjected to coagulation of the middle cerebral artery 4 days prior to sacrifice. (*P < 0.05; **P < 0.01). For the sake of clarity, standard errors have been omitted (cf. Table III for absolute values). The values for kynurenate are taken from Table V.
induced 0)3 proliferation by 36%. The highest doses of the adenosine agonists induced a pronounced peripheral vasodilatation.
Calcium channel blockers Five calcium channel blockers, each representative of
a different class 75, were studied with the standard post-occlusion regimen. Significant changes in 0)3 binding in the cortex of MCAO mice were not observed with diltiazem, flunarizine and verapamil (Table VII).
With nimodipine, a significant prevention of the ischaemia-induced increase in cortical 0) 3 sites was ob- served only at the highest dose tested (5 mg/kg i.p.). A
seemingly dose-related protection was noted with perhe- xiline though a significant delayed mortality (4 out of 10 mice) was associated with the highest dose (20 mg/kg i.p.) of this agent (Table VII).
As nimodipine may be active in the pre-treatment, but not post-treatment, of experimental focal cerebral isch- aemia 34'49, we used a preventive treatment schedule
injecting nimodipine or diltiazem 10 min prior to the induction of ischaemia. The following administration was effected 6 h after occlusion and the standard dosing schedule was employed thereafter. Under these condi- tions, nimodipine was no more efficacious than when the standard post-occlusion regimen was used. In contrast,
DECREASE IN to 3 DENSITIES (% VEHICLE)
0
- 2 0
-40
- 6 0
- 8 0
SL 82.0715 ~ . . . . . . . 10 mg/kg p
- 1 0 0 --/,~ I I I I I I
5 15 45 90 180 360
DELAY log [mini
Fig. 5. Influence of the delay between the coagulation of the middle cerebral artery and the first administration of SL 82.0715 on the final augmentation of % site densities in the mouse cortex as assessed at 4 days post-occlusion. Relative increases in (,o 3 densities were calculated in comparison to non-occluded mice. The values are expressed as percentage decreases (+ S.E.M.) compared to vehicle-treated, occluded mice (*P < 0.05; **P < 0.01).
298
DECREASE IN co 3 DENSITIES (% VEHICLE)
- 2 0
- 4 0 ,~r
- 6 0
- 80
- 100 I 0.08
I st_ 82.o715 I 10 mg/kg ip
I I I I I
6 18 36 60 84
DURATION log [hi
Fig. 6. Influence of repeated SL 82.0715 administration (10 mg/kg i.p.) for various durat ions (6-84 h post-occlusion) on the final, ischaemia-induced increase in cortical to 3 site densities, measured 4 days after middle cerebral artery coagulation. Relative increases in to 3 sites were calculated in comparison to non-occluded mice. The values are percentage decreases ( + S .E.M.) compared to vehicle-treated, occluded mice (*P < 0.05; **P < 0.01).
doses of 3 and 10 mg/kg diltiazem provided a significant protection when the first drug administration was ef- fected prior to the coagulation of the middle cerebral artery (Table VIII).
PAF antagonists Two triazolodiazepine derivatives (brotizolam and
T A B L E IV
Effects o f various o receptor ligands (administered following occlu- sion) on the ischaemia-induced increase in 03 site densities in the mouse cortex
Results are means + S.E.M. of data obtained on the number (n) of mice indicated. Mice were sacrificed 4 days following middle cerebral artery coagulation and to 3 site density in the entire ipsilateral cerebral cortex was de termined in membrane preparations. The t reatment regimen is described in Materials and Methods.
Drug n Dose ~3 Site density % Change (mg/kg (pmol/g tissue) vs vehicle i.p.)
Sham occlusion 15 - 30.5 + 1.4 M C A O + vehicle 26 - 56.4 + 1.9 M C A O + (-)-3PPP 9 1.0 55.0 + 2.8 - 6
10 3.0 56.6 + 1.8 + 1 10 10.0 53.5 + 1.8 -11
M C A O + (+) -3PPP l0 1.0 63.5 + 1.9 +28 9 3.0 67.4 + 1.7"* +43 9 10.0 63.2 + 2.2 +27
Sham occlusion 14 - 31.0 + 0.6 M C A O + vehicle 19 - 53.6 + 1.9 M C A O + haloperidol 11 0.3 52.3 + 1.5 - 6
9 1.0 52.3 + 2.1 - 6 9 3.0 45.8 + 1.6" -35
alprazolam) with reported PAF antagonist activity 17 have been tested. These compounds were unable to prevent the increase in to 3 site densities even at doses that induced a long-lasting sedation (10 mg/kg). A Gingko biloba
T A B L E V
Effects o f excitatory amino acid receptor antagonists (administered following occlusion) on the ischaemia-induced increase in to 3 densities in the mouse cortex
Results are means _+ S.E.M. of data obtained on the num b er (n) of mice indicated. Mice were sacrificed 4 days after middle cerebral artery coagulation and % site density in the entire ipsilateral cerebral cortex was determined in membrane preparat ions. The t rea tment regimen is described in Materials and Methods .
Drug n Dose to 3 Site density % Change (mg/kg (prnol/g tissue) vs vehicle i.p.)
Sham occlusion 10 - 27.9 + 1.1 M C A O + vehicle 9 - 53.6 + 1.9 M C A O + kynurenate 10 30 47.2 + 3.4 -25
10 100 47.4 + 1.9 -24 10 300 38.8 + 1.0"* -58
Sham occlusion 18 - 18.7 + 0.7 M C A O + vehicle 17 - 32.5 + 1.3 M C A O + riluzole 10 33.9 + 1.1 +10
10 33.0 + 1.4 +3 10 36.5 + 4.0 +29
Sham occlusion 23 - 24.0 + 0.7 M C A O + vehicle 26 - 40.3 + 1.4 M C A O + MLV5860 11 3 38.0 + 1.3 -14
11 10 37.7 + 3.1 -1 6 9 30 36.0 + 3.6 -2 6
*P < 0.05; **P < 0.01 versus M C A O + vehicle. **P < 0.01 vs M C A O + vehicle.
ext rac t , f rom which cer ta in molecu les have been isolated
that possess P A F an tagonis t act ivi ty t3, also fai led to
r educe ~o 3 site densi t ies in M C A O mice (Table IX).
M i s c e l l a n e o u s ' a n t i - i s c h a e m i c ' d r u g s
A wide range of chemica l ly un re l a t ed c o m p o u n d s that
have been a d v o c a t e d for the t r e a t m e n t of ce reb rovascu la r
d i sorders was eva lua ted . With the s tandard pos t - t rea t -
m e n t r eg imen , v i n c a m i n e 81 exh ib i t ed a c lear-cut p reven-
t ive effect , wi th doses of 0.3 to 10 mg/kg i.p. be ing
e q u i p o t e n t (Table X). A slight and not dose - re la ted
dec rease was seen with n icergol ine . Pir ibedi l , at the
h ighes t dose (5 mg/kg i .p.) e f fec ted a highly significant
dec rease in the cor t ical to 3 site p ro l i fe ra t ion p r o v o k e d by
focal ce rebra l i schaemia .
C l o n a z e p a m , a specific agonis t at the central (091,2) b e n z o d i a z e p i n e recep to r s was unable to r educe co 3 site
densi t ies in M C A O mice , even at doses that induced
cent ra l depress ion (10 mg/kg) . A l so inact ive was the
g lucocor t i co id an t i - in f l ammato ry and i m m u n o d e p r e s -
sant , d e x a m e t h a s o n e . In contras t , the puta t ive anti-
i schaemic agent , cyc lande la te TM, r educed significantly ~o 3
levels at 10 and 30 mg/kg. No significant effect , at any
dose , was n o t e d for the fo l lowing compounds : pyr i th io-
xine, naf t id rofury l and papave r ine (Table X).
TABLE VI
Effects o f agents that interfere with adenosine-mediated transmission (administered fol lowing occlusion) on the ischaemia-induced increase in ~o 3 site densities in the mouse cortex
Results are means + S.E.M. of data obtained on the number (n) of mice indicated. Mice were sacrificed 4 days following middle cerebral artery coagulation and w 3 site density in the entire ipsilateral cerebral cortex was determined in membrane preparations. The treatment regimen is described in Materials and Methods.
Drug n Dose a~ 3 Site density % Change (mg/kg (pmol/g tissue) vs vehicle i.p.)
The backg round to the p resen t s tudy was based on our
p rev ious au to rad iog raph ic inves t iga t ions in which we
d e m o n s t r a t e d that increase in ~o 3 b inding sites, a ref lec-
t ion of the glial r eac t ion and m a c r o p h a g e invas ion that
fol lows brain in jury , cons t i tu tes a sensi t ive m a r k e r of
neu rona l dea th . In these s tudies , increases in ce rebra l w 3
sites were seen fo l lowing exc i to tox ic and focal i schaemic
lesions in rats 24 as well as in p ro l i f e ra t ive , i schaemic and
demye l ina t ing lesions in h u m a n bra in t issue 7. In ou r
c o m p a n i o n pape r 6, we address the ques t ion of the
sensit ivi ty and reproduc ib i l i ty of ~o 3 label l ing fo l lowing a
var ie ty of i schaemic and neu ro tox i c insults in the roden t
brain. We obse rved that , in mice , the coagu la t ion of the
midd le ce rebra l a r te ry e v o k e d a r ep roduc ib l e increase in
~o 3 site densi ty in the ipsi la teral hemisphe re . Paral le l
TABLE VII
Effects o f calcium channel blockers (administered fol lowing occlu- sion) on the ischaemia-induced increase in 0) 3 site densities in the mouse cortex
Results are means + S.E.M. of data obtained on the number (n) of mice indicated. Mice were sacrificed 4 days following middle cerebral artery coagulation and % site density in the entire ipsilateral cerebral cortex was determined in membrane preparations. The treatment regimen is described in Materials and Methods.
Drug n Dose to 3 Site density % Change (mg/kg (pmol/g tissue) vs vehicle i.p.)
**P < 0.01 vs MCAO + vehicle. *P < 0.05; **P < 0.01 vs MCAO + vehicle.
300
T A B L E VIII
Effects o f pretreatment with calcium channel blockers on the ischaemia-induced increase in oL~ site densities in the mouse cortex
Results are means + S.E.M. of data obtained on the number (n) of mice indicated. Mice were sacrificed 4 days following middle cerebral artery coagulat ion and ~3 site density in the entire ipsilateral cerebral cortex was de termined in membrane preparations. The t reatment regimen is described in Materials and Methods.
Drug n Dose to 3 Site density % Change (mg/kg (pmol/g tissue) vs vehicle i.p.)
Sham occlusion 10 - 26.2 + 1.9 M C A O + vehicle 10 - 45.5 + 1.6 M C A O + diltiazem 10 1.0 42.1 + 1.5 -18
10 3.0 33.5 + 1.6"* -62 10 10.0 33.9 + 1.4"* -60
Sham occlusion 10 - 27.9 + 1.1 M C A O + vehicle 18 - 52.1 + 1.3 M C A O + nimodipine 10 0.5 45.9 + 1.7 -26
10 1.5 45.1 + 0.6** -29 10 5.0 47.1 + 1.5 -21
*P < 0.01 vs M C A O + vehicle.
a u t o r a d i o g r a p h i c s t u d i e s d e m o n s t r a t e d t h a t t h e g r e a t e s t
b i n d i n g i n c r e a s e s w e r e l o c a l i z e d a t t h e i n t e r f a c e b e t w e e n
n e c r o t i c a n d h e a l t h y t i s s u e . T h e s e d a t a s u p p o r t t h e
a p p l i c a b i l i t y o f t h i s m a r k e r t o a s s e s t h e e x t e n t o f f o c a l
i s c h a e m i c l e s i o n s in m i c e in w h i c h t h e r e is a g o o d s p a t i a l
c o r r e s p o n d e n c e b e t w e e n t h e i n f a r c t e d c o r e a n d t h e
s u r r o u n d i n g '0 )3- r ich ' t i s s u e . B a s e d o n t h e s e o b s e r v a -
T A B L E IX
Effects o f various PAF antagonists (administered following occlusion) on the ischaemia-induced increase in o93 site densities in the mouse cortex
Results are means + S.E.M. of data obtained on the number (n) of mice indicated. Mice were sacrificed 4 days following middle cerebral artery coagulation and 093 site density in the entire ipsilateral cerebral cortex was de termined in membrane preparations. The t reatment regimen is described in Materials and Methods.
Drug n Dose o~ 3 Site density % Change (mg/kg (pmol/g tissue) vs vehicle i.p.)
Sham occlusion 15 - 27.3 + 0.8 M C A O + vehicle 18 - 52.9 + 1.5 M C A O + Gingko
Sham occlusion 18 - 20.8 + 0.9 M C A O + vehicle 22 - 35.2 + 1.8 M C A O + brotizolam 11 1.0 34.7 + 1.0 - 4
9 3.0 36.2 + 1.1 +6 9 10.0 34.3 + 1.1 - 7
M C A O + alprazolam 10 1.0 37.2 + 1.8 +13 9 3.0 37.5 + 1.9 + 15 8 10.0 31.0 + 1.0 -29
t i o n s , t h e a i m o f t h e p r e s e n t s t u d y w a s t o e x a m i n e t h e
p o t e n t i a l n e u r o p r o t e c t i v e a c t i v i t y o f a v a r i e t y o f a g e n t s as
r e f l e c t e d b y a b l o c k a d e o f t h e i n c r e a s e in 0) 3 s i t e d e n s i t y
in t h i s m u r i n e m o d e l o f f o c a l c e r e b r a l i s c h a e m i a .
T A B L E X
Effects of various 'anti-ischaemic' agents (administered following occlusion) on the ischaemia-induced increase in 0) 3 site densities in the mouse cortex
Results are means + S.E.M. of data obtained on the num b er (n) of mice indicated. Mice were sacrificed 4 days following middle cerebral artery coagulation and ~o 3 site density in the entire ipsilateral cerebral cortex was determined in membrane preparat ions. The t rea tment regimen is described in Materials and Methods .
Drug n Dose o93 Site density % Change (mg/kg (pmol/g tissue) vs vehicle i.p.)
Sham occlusion 30 - 25.3 + 0.8 M C A O + vehicle 39 - 47.7 + 1.6 M C A O + vincamine 10 0.1 41.1 + 0.9 -3 0
Validation of the approach Throughout the studies here reported (more than 3000
occlusions in 20 independent experiments), we have observed that the ischaemia-induced increases in hemi- spheric 0)3 site density were highly reproducible. Basal levels ranged from 18 to 27 fmol/mg tissue, and the relative increases after middle cerebral occlusion varied from 76 to 128%; the relative increases were not apparently related to the basal levels. A reproducible though small elevation of 0; 3 site levels was observed in sham-operated animals. As this increase was not pre- vented by anti-ischaemic agents, it may reflect a non- specific, mechanical lesion. Accordingly, the percent protection by anti-ischaemic agents was calculated by considering the 0) 3 site densities in occluded mice as 100% and the density in sham-operated animals as 0%.
Previous studies have demonstrated that the increase in 0)3 levels faithfully reflects the neuronal loss in a variety of experimental and human brain injuries. Fur- thermore, this approach to the evaluation of anti- ischaemic drugs in MCAO mice is justified as the reduction in 0)3 site levels was accompanied by a parallel decrease in the infarcted area as assessed histologically for the anti-ischaemic compound, SL 82.0715. This compound exhibits cytoprotective activity in both rat and feline focal ischaemia in which the volume of infarction was assessed histologically 35.
Although the infarction was mainly located in the parietal cortex an increase in 0) 3 site density, of a magnitude similar to that measured in the cortex, was seen in subcortical regions. These regional binding increases are, most likely, the result of the glial reaction subsequent to the ischaemic insult as in the SL 82.0715 treated mice subcortical and cortical binding increases were reduced to a similar extent (as assessed both autoradiographically and in homogenates obtained from dissected tissues).
The reproducibility of 0)3 site levels as a marker of the neuroprotective activity is illustrated by the fact that our reference compound, SL 82.0715, antagonized in a dose-dependent fashion, the 0)3 site increases and dem- onstrated a constant degree of protection in 7 different experiments carried out within a 12 month period. Although some variability was observed for the 0) 3 site densities in the control groups, percent increases in 0)3 sites for MCAO mice in each individual experiment were not dependent on basal levels. Moreover, this variability is not likely to affect the neuroprotective potency since for compounds tested several times the protective effect was independent of the basal 0)3 site levels.
Overall, these data indicate that the drug-induced decrease in 0)3 site densities in MCAO mice is an accurate reflection of the reduction of the infarction and thus
301
provides a reliable estimate of the anti-ischaemic poten- tial of differing therapeutic interventions. As the to 3 site quantification in homogenates is advantageous in terms of rapidity, this approach was selected for the subsequent pharmacotherapy studies.
NMDA antagonists The pivotal role played by excitatory amino acid
neurotransmitters in the neuronal death occasioned by a variety of neuropathological states is supported by ever-growing experimental evidence 18"63"66'7°. It is well
established that elevated concentrations of glutamate and/or other natural agonists at the NMDA receptor, effect a delayed neuronal death; these excitatory trans- mitters accumulate massively in the extracellular space in experimental models of global ischaemia 12'39. While the NMDA receptor subserves defined functions in the normal brain, its excessive activation results in an irreversible neurotoxicity. The results obtained in the focal ischaemia model herein developed attest to the major and nefarious importance of excitatory amino acids since all the NMDA receptor antagonists examined decreased the 0)3 levels in cortical homogenates from MCAO mice. The specificity of this protective effect in ischaemic tissue is demonstrated by the failure of these compounds to significantly decrease 0)3 site levels in sham-operated mice.
The most potent compound was the non-competitive NMDA receptor antagonist, MK-801, which reduced, by 50%, the elevated 0)3 site levels at 0.2 mg/kg. Moreover, an almost complete protection was obtained at a dose of 1 mg/kg. The lower in vivo potency and efficacy in the mouse model of TCP (a phencyclidine analogue which, like MK-801, blocks the NMDA receptor associated channel 1°) is in agreement with its lower in vitro affinity as a NMDA receptor antagonist and, perhaps, due to the fact that the channel blockade by TCP is less use- dependent than that of MK-80128. The effectiveness of these compounds in ischaemia compares well with their potency as antagonists of the NMDA receptor in several biochemical paradigms in vivo '5 and as inhibitors of in vivo 3H-MK-801 binding 6° in the rodent.
The use-dependency of MK-801 and TCP at the NMDA receptor associated channel, well established in vitro but still controversial in vivo 19, may explain the signal neuroprotective activity of these blockers in con- ditions in which NMDA receptors are maximally acti- vated by an excessive release of excitatory amino acids. If, by analogy with the experiments performed with SL 82.0715, the almost complete MK-801-induced reduction of 0)3 site proliferation is paralleled by an equivalent decrease in infarcted volume, the implication would be that the ischaemia-provoked neuronal loss in mice is
302
almost entirely mediated by an excessive activation of
NMDA receptors. The direct neuroprotective effects,
evaluated by quantitative histology, of MK-801 in focal cerebral ischaemia have been amply demonstrated in cat and rat studies 53 5.5; similarly, TCP reduces infarction volume in rats 25.
Marked protection was observed with ifenprodil and SL 82.0715, two non-competitive NMDA receptor an- tagonists, that act via the polyamine modulatory site within the NMDA receptor complex 15'16. Although both
phenylethanolamine derivatives display a similar activity in vivo as antagonists of NMDA receptor-mediated responses 15, SL 82.0715 is more potent than ifenprodil in MCAO mice. This differential potency may relate to the rapid metabolization of ifenprodil in rodents 35 which would explain why, to demonstrate the cerebroprotective activity of ifenprodil in mice, an increased initial fre- quency of administration was needed.
In our model, dextromethorphan and CGS 19755 afforded a limited protection against ischaemia-induced glial proliferation. The weak neuroprotective activity of dextromethorphan is consistent with its low potency as NMDA receptor antagonist 8°. The low potency of CGS 19755 at preventing 0) 3 site elevation is paradoxical in view of its nanomolar affinity for the NMDA receptor but may relate to the competitive nature of its blockade 5° which can be overcome by excessive release of glutamate in ischaemic tissue.
Accumulating experimental evidence indicates that, in global ischaemia, a delayed neuronal loss may be the consequence of an excessive activation of NMDA recep- tors by excitatory amino acids leaking from metabolically injured or dying neurones 3°'3z'7°. Experiments in which the first administration of SL 82.0715 was delayed, support the existence of late (or progressive) neuronal death, even in focal ischaemia, since a reduction in 0) 3 site density was still detectable when this compound was first administered between 45 min and 3 h after the occlusion. However, this critical period is limited since, if the first administration is delayed by more than 6 h, then significant neuroprotection cannot be evidenced in this experimental model. This delay (approx. 3-6 h) for effective therapeutic intervention in experimental focal ischaemia is shorter than that noted for MK-801-induced neuroprotection in global ischaemia (6-24 h) 27 but correlates well with two studies in which irreversible occlusion of the middle cerebral artery for 6 h or more is essential to obtain maximal infarct size 11'79.
Studies with SL 82.0715 suggested that the cytopro- tective effects, obtained with administrations limited to 5 min, 6 h and 18 h after occlusion of the middle cerebral artery, could not be markedly enhanced by subsequently repeating the administration of the compound. Thus,
other than the timing of the maximal delay of pharma-
cological intervention, there is the question of the
optimal duration of treatment. Overall, the window of therapeutic opportunity in stroke paradigms may be a function of the drug, or drug class, under study but, at least, this model of middle cerebral artery occlusion in the mouse allows such questions to be examined.
The now recognized psychostimulant effects of the blockers of the NMDA receptor-operated channel such as MK-801 ~2'57'58 and phencyclidine have cast some doubt
on the applicability of these compounds to the therapy of stroke. However, these unwanted effects are not a constant feature of NMDA receptor antagonists since the phenylethanolamine derivatives, ifenprodil and SL 82.0715, which readily block NMDA receptor-mediated responses in vitro and in vivo 15A6 do not display notice- able psychostimulant effects in the rodent 65 and are not psychotomimetic in man. As these compounds provide an effective neuroprotection in feline, rat 35 and murine (present results) focal ischaemia models, yet are devoid of dissociative properties, they hold considerable promise for the prevention and treatment of cerebrovascular accidents. A preliminary account of these results has already been given 8.
cr Receptor ligands The receptor or receptors that mediate the psychoto-
mimetic effects of phencyclidine analogues have long been controversial. Phencyclidine labels two binding sites: the o and the phencyclidine binding sites 61'71. Whereas the latter seems to be associated with the NMDA receptor channel, the former is in search of a function. Despite the fact that some of the NMDA antagonists tested have a certain affinity for a sites, their neuroprotective effects do not seem to be mediated by these sites because the specific o receptor ligand, (+)- 3PPP 21, not only failed to prevent the increase in 0) 3 binding site densities in MCAO mice but, at 3 mg/kg, even exacerbated this increase. In contrast, haloperidol, a D2 receptor antagonist that possesses a high affinity for the o receptor 21 partially prevented 093 site increases, though only at cataleptogenic and highly sedative and hypothermic doses. Whether or not these effects are mediated by the o receptors, or are related to alterations of dopaminergic (and other monoaminergic) neurotransmis- sion, remains to be established. In any case, these results speak strongly against the involvement of o receptors in the evolution of ischaemic brain damage. Again, the cytopro- tective potency of MK-801 and phencyclidine in the present investigation (despite their relatively low affinities in vitro for the o site) would militate against a role for the o sites in the aetiology of neuronal death, an observation coherent with the in vitro data of Goldberg et al. 31.
Antagonists of other excitatory amino acid (quisqualate and kainate) receptors
The almost complete protection afforded by the NMDA receptor channel blocker, MK-801 in MCAO mice, suggests that pathological activation of this recep- tor subtype is the primary cause of the ischaemic neuronal loss. However, excitatory amino acids also activate receptors with a different pharmacological spec- ificity such as the quisqualate and kainate subtypes 77. An excessive stimulation of the kainate receptor has been implicated in some neurodegenerative disorders and the overactivation of both quisqualate and kainate receptors (via a depolarization) will relieve the magnesium block- ade of the NMDA receptor thus favouring its use 47"62.
However, the quisqualate and kainate receptors do not appear to be of pathogenetic importance in the MCAO- induced neuronal death, since the selective quisqualate receptor antagonist, MLV 586068, was inactive. Further- more, kynurenic acid, a non-specific antagonist of excit- atory amino acid neurotransmission 77 that blocks both the kainate receptor and the glycine regulatory site in the NMDA receptor complex 41 was only neuroprotective at very high doses. The correspondence between the doses of kynurenic acid required to decrease the infarcted surface in rat brain focal ischaemia 29 and those that prevent the increase in to 3 site densities in MCAO mice further validate our approach to assess the importance of neuronal damage. The lack of involvement of non- NMDA excitatory amino acid receptors is also supported by the inefficacy of riluzole, a weak and non-specific antagonist of excitatory amino acid neurotransmission 5, to significantly decrease infarction volume, as assessed by the proliferation of ~03 sites.
Adenosine receptor agonists and antagonists Metabolic insults to the nervous tissue (seizures,
ischaemia, hypoglycaemia) not only result in an accumu- lation of excitatory amino acid neurotransmitters but also in an accelerated formation and massive release of adenosine 4°'56. Adenosine may play a protective role in ischaemia by hyperpolarizing neurones and decreasing their activity through the activation of the adenosine A 1
receptors. Indeed, adenosine A~ agonists decrease, and adenosine A 1 antagonists exacerbate, the loss of hippo- campal neurons in a model of global ischaemia in the gerbil 64. This protective effect seems to be mediated, at least in part, by the inhibition of glutamate release subsequent to the activation of presynaptic adenosinergic A 1 receptors located on glutamatergic fibres 22. Such presynaptic mechanisms seem inoperative in focal isch- aemia in the mouse since the adenosine A1 receptor agonists employed failed to prevent the ischaemia- induced increase in cortical o~ 3 site densities. A possible
303
explanation for this failure may be related to the fact that, after MCAO, the release of excitatory amino acids is due to a non-specific alteration of cellular permeability (leakage) and is not susceptible to presynaptic modula- tion. Although improbable in the cortex, such protective mechanisms may be operative in the hippocampus. Indeed, it has been recently reported that much higher concentrations of adenosine A t agonists are necessary to block epileptiform activity in the cerebral cortex than in the hippocampus 52, a difference explained by the much higher adenosine A 1 receptor density observed in the latter structure. In addition, the presence of presynaptic adenosine A 1 receptors on glutamatergic terminals has been demonstrated in the hippocampus and cerebellum but not in cortical areas 33. The failure of A 1 receptors to
modulate the release of glutamate in the ischaemic tissue is also suggested by the fact that the adenosine receptor antagonist, theophylline, did not modify cortical w3 site levels in ischaemic mice though this xanthine derivative exacerbates hippocampal neuronal damage following global ischaemia 2°.
Calcium channel blockers It is now well recognized that a massive intraneuronal
accumulation of Ca 2÷ is of cardinal importance in the mechanisms that lead to irreversible cell damage 69. Intracellular calcium can be increased by a variety of mechanisms such as stimulation of receptor- or voltage- operated calcium channels, operation of Na+/Ca 2÷ anti- ports, release from intracellular stores TM. The ionic channel associated with the NMDA receptor is highly permeant to this cation 48. The recognition of the cardio- protective activity of voltage-operated calcium channel blockers was at the origin of the proposed cerebropro- tective activity of compounds belonging to this therapeu- tic class. However, the neuroprotective potential of calcium channel blockers is controversial and experimen- tal findings are incongruent 46. Our studies in MCAO mice cast doubt on the applicability of existing calcium channel blockers to prevent the neuronal loss in focal ischaemia. When given following the occlusion, perhexi- line caused a decrease in ~03 site levels only at doses that in themselves were lethal or sublethal, and nimodipine afforded only a limited protection at cardiotoxic doses; this minimal protection was not improved when the dihydropyridine derivative was given prior to the occlu- sion. The weak protective activity of nimodipine is consistent with previous studies in middle cerebral artery occluded r a t s 34'49. In contrast, moderate doses of diltia- zem, when given prior to (but not after) the occlusion prevented the increase in o93 site levels to an extent similar to that seen with NMDA receptor antagonists.
The limited protective activity of calcium channel
304
blockers in focal ischaemia is in contrast with the well recognized protection of hippocampal neurones afforded by these drugs in global ischaemia 2"3. Several explana- tions can be proposed for this discrepancy. Firstly, dihydropyridine-sensitive calcium channels may not be involved in the intracellular calcium accumulation in the ischaemic cortex. It has been observed that, consistent with the higher densities of dihydropyridine-sensitive calcium channels in the hippocampus, Ca 2÷ influx into hippocampal neurones is more sensitive to blockade by calcium channel antagonists than in neurones from other brain regions 48. It is also possible that a neuroprotective effect could be compromised by a deleterious effect at the level of the vascular bed. Vasodilatation in healthy tissue may aggravate the ischaemic damage by provoking a steal phenomenon in the periphery of the necrotic fOCUS 46. Finally, peripheral effects of calcium antagonists (hypo- tension, negative inotropic effects) may be incompatible with the administration of the doses necessary to achieve therapeutic concentrations in the brain.
Although several reasons could be proferred for the inefficacy of calcium antagonists in focal ischaemia, it is clear that their potency, as a class, is considerably less than that of NMDA antagonists though the neuropro- tective effects (in pretreatment) of diltiazem deserve further attention. The clinical usefulness of calcium channel blockers in the treatment of cerebrovascular accidents is also much debated except in the treatment of the delayed neurological deficit that can follow subarach- noid haemorrhage 1.
PAF antagonists Some evidence has been provided for the involvement
of PAF in the post-ischaemic neuronal loss. PAF, secreted by cells of the monocytic and polymorphonu- clear lineages, may contribute to the total neuronal loss through an aggravation of secondary brain oedema. Moreover, PAF seems to have a direct neurotoxic effect as its application can increase intracellular Ca 2+ to cytotoxic concentrations 43.
Support for the involvement of PAF in the evolution of the ischaemic damage is adduced from the observed protective effect of several PAF antagonists in the gerbil global ischaemia model 72. However, our studies do not support the involvement of PAF in the neuronal loss observed in MCAO mice since two triazolodiazepines 17'44 and a Gingko biloba extract that possess PAF antagonist properties ~3 were equally inefficacious.
Our results are at variance with a recent report in which the PAF antagonists Gingkolide B and WEB 2107, when first administered prior to MCAO occlusion, afforded a significant protection against the neuronal death in a model of focal cerebral ischaemia in the rat 76.
Several experimental differences (antagonists selected, pre- versus post-administration schedule and animal species) may account for this discrepancy. It has been hypothesized that PAF antagonists mainly protect against the deleterious effects taking place during the reperfusion period TM and the failure of PAF antagonists, administered post-occlusion, to protect against the neuronal loss observed in MCAO mice would be compatible with this view.
Miscellaneous drugs It has been hypothesized that the lack of neuropro-
tective effects in a feline model of focal ischaemia of certain compounds advocated for the treatment of cere- brovascular diseases (naftidrofuryl, papaverine) was re- lated to their relaxatory properties; tissue sparing was observed only with agents that directly contract isolated cerebral arteries 36. Our studies in MCAO mice to some degree support these observations. The protective activ- ity of vincamine in ischaemic cats has been related to the vasoconstriction of isolated pial vessels evoked by this agent sl. However, the protective effect of vincamine might also be explained by other mechanisms such as the prevention of the loss of the respiratory control (uncou- pling) of brain mitochondria 51. Nonetheless, the involve- ment of these vascular and mitochondrial effects in the cerebroprotective activity of vincamine is put into doubt by the lack of protective activity of nicergoline, a vasoconstrictor compound with mitochondrial stabilizing properties in vitro 36.
The dose-dependent reduction of 093 site densities in MCAO mice evoked by cyclandelate is consistent with its presumed clinical activity in cerebrovascular disorders TM. The neuroprotective activity of cyclandelate might be related to a blockade of C a 2+ in f lux TM. However, its low affinity for L-type calcium channels (as labelled with [3H]nitrendipine)4 suggests a mechanism of action differ- ent from that proposed for most known calcium channel blockers.
Clonazepam was selected as a prototype agonist for central benzodiazepine (091,2) receptors because it does not possess significant affinity for o93 sites67. Although benzodiazepines are currently used as therapeutic agents in cerebrovascular accidents because of their brain depressant activity, clonazepam does not appear to provide any protection in MCAO mice.
The rationale for the use of o93 site density as an estimate of cerebroprotective activity relies upon the assumption that drugs will inhibit o93 site increases by preventing the neuronal death and thus the subsequent gliotic and macrophagic reaction. However, the possibil- ity remains that an immunosuppressive drug, by prevent- ing macrophage invasion or immune influences on glial
305
cells, could inhibit the increase in ~03 binding densities without providing any neuronal protection, thus giving
rise to false-positive results. This does not appear to be the case since immunosuppressive dosages of dexameth-
asone (which decrease 0)3 levels in the mouse thymus by 80%) 23 failed to alter 0)3 site densities in M C A O or
control mice. The failure of this agent to decrease co 3 site
levels speaks against the involvement of brain oedema in
the neuronal loss and the glial proliferation subsequent to M C A O in mice.
CONCLUSIONS
The present study demonstrates that the measurement
of 0)3 site levels in cortical homogenates of M C A O mice constitute a rapid, sensitive and accurate technique to
determine the anti-ischaemic potential of putative neu- roprotective agents. The increases in 0) 3 site levels 4 days
post occlusion faithfully reflect the extent of the consol-
idated infarction and its reduction by cerebroprotective
agents. Furthermore, this technique renders feasible the determination of the cerebroprotective potency of a drug
at several doses and thus the calculation of a value (eg.
EDso ) which permits the comparison of efficacy in a given pharmacological class.
The assessment of the ischaemic damage in M C A O
mice by the 0)3 radioligand technique provides a robust
method to detect active compounds acting on the threatened neuropil. The results in mice are highly
consistent with those obtained by quantitative histology
in rat and cat models of focal cerebral ischaemia for MK-80153-55 , TCP 23, kynurenate 29, SL 82.071535 , nimodi- pine 34"49, piribedi137, pyrithioxine 37 and vincamine 37.
These data, although limited, support the validity of our approach.
Using this method, we have demonstrated the anti- ischaemic potential of N M D A receptor antagonists in
M C A O mice; no systematic statement can be made about
any of the other drug classes studied. However , it must be
accepted that with a situation as complicated as human stroke (only partially modelled in experimental animals)
optimal results are likely to be eventually achieved by a
combination of pharmacological approaches. Nonethe- less, the coherence of the present findings provides
further support for the involvement of excitatory amino
acid neurotransmitters in the ischaemic neuronal loss.
Finally, our results underline the applicability of N M D A receptor antagonists in the treatment of stroke for which
a feasible window of therapeutic opportunity exists in the hours that follow the ictal event.
REFERENCES
1 Allen, G.S., Ahn, H.S., Preziosi, T.J., Battye, R., Boone, S.C., Chou, S.N., Kelly, D.L., Weir, B.K., Crabbe, R.A., Lavik, P.J., Rosenbloom, S.B., Dorsey, EC., Ingram, C.R., Mellits, D.E., Bertsch, L.A., Boisvert, D.P.J., Hundley, M.B., John- son, R.K., Strom, J.A. and Transou, C.R., Cerebral arterial spasm - - a controlled trial of nimodipine in patients with subarachnoid hemorrhage, N. Engl. J. Med., 308 (1983) 619- 624.
2 Alps, B.J., Calder, C., Hass, W.K. and Wilson, A.D., Com- parative protective effects of nicardipine, flunarizine, lidoflazine and nimodipine against ischaemic injury in the hippocampus of the Mongolian gerbil, Br. J. Pharmacol., 93 (1988) 877-883.
3 Alps, B.J. and Hass, W.K., The potential beneficial effect of nicardipine in a rat model of transient forebrain ischaemia, Neurology, 37 (1987) 809-814.
4 Bast, A., Leurs, R. and Timmerman, H., Cyclandelate as a calcium modulating agent in rat cerebral cortex, Drugs, 33 (1987) 67-74.
5 Benavides, J., Camelin, J.C., Mitrani, N., Flamand, E, Uzan, A., Legrand, J.J., Gu6r6my, C. and Le Fur, G., 2-Amino- 6-trifluoromethoxy benzothiazole, a possible antagonist of ex- citatory amino acid neurotransmission. II. Biochemical proper- ties, Neuropharmacology, 24 (1985) 1085-1092.
6 Benavides, J., Capdeville, C., Dauphin, E, Dubois, A., Du- verger, D., Fage, D., Goni, B., MacKenzie, E.T. and Scatton, B., The quantification of brain lesions with an % ligand: a critical analysis of animal models of cerebral ischaemia and neurodegeneration, Brain Research, 522 (1990) 275-289.
7 Benavides, J., Cornu, P., Dennis, T., Dubois, A., Hauw, J.J., MacKenzie, E.T., Sazdovitch, V. and Scatton, B., Imaging of human brain lesions with an w3 site radioligand, Ann. Neurol., 24 (1988) 708-712.
8 Benavides, J., Cornu, P., Dubois, A., Goni, B., MacKenzie,
E.T. and Scanon, B., % Binding sites as a tool for the detection and quantification of brain lesions: application to the evaluation of neuroprotective agents. In J. Kriegelstein (Ed.), Pharmacol- ogy of Cerebral lschaemia 1988, CRC Press, Boca Raton, 1989, pp. 187-196.
9 Benavides, J., Quarteronet, D., Imbault, E, Malgouris, C., Uzan, A., Renault, C., Dubroeucq, M.C., Gueremy, C. and Le Fur, G., Labelling of 'peripheral-type' benzodiazepine binding sites in the rat brain by using 3H-PK 11195, an isoquinoline carboxamide derivative: kinetic studies and autoradiographic localization, J. Neurochem., 41 (1983) 1744-1750.
10 Benavides, J., Rivy, J.P., Carter, C. and Scatton, B., Differen- tial modulation of [3H]TCP binding to the NMDA receptor by L-glutamate and glycine, Eur. J. Pharmacol., 149 (1988) 67-72.
11 Bederson, 3.B., Pitts, L.H., Tsuji, M., Nishimura, M.C., Davis, R.L. and Bartkowski, H., Rat middle cerebral artery occlusion: evaluation of the model and development of a neurologic examination, Stroke, 17 (1986) 472-476.
12 Benveniste, H., Drejer, J., Schousboe, A. and Diemer, N.H., Elevation of the extracellular concentrations of glutamate and aspartate in rat hippocampus during transient cerebral iscbemia monitored by intracerebral microdialysis, J. Neurochem., 43 (1984) 1369-1374.
13 Braquet, P., Involvement of PAF-acether in various immune disorders using BN 52021 (ginkgolide B): a powerful PAF- acether antagonist isolated from Gingko biloba L. In Advances in Prostaglandin, Thromboxane and Leukotriene Research, 1/ol. 16, Raven, New York, 1986, pp. 179-198.
14 Braquet, P., Touqui, L., Shen, T.Y. and Vargaftig, B.B., Perspectives in platelet-activating factor research, Pharmacol. Rev., 39 (1987) 97-142.
15 Carter, C., Benavides, J., Legendre, P., Vincent, J.D., NoEl, E, Thuret, F., Lloyd, K.G., Arbilla, S., Zivkovic, B., MacKenzie, E.T., Scatton, B. and Langer, S.Z., lfenprodil and SL 82.0715 as cerebral anti-ischemic agents. II. Evidence for N-methyl-
16 Carter, C., Rivy, J.P. and Scatton, B., Ifenprodil and SL 82.0715 are antagonists at the polyamine site of the N-methyl-D-aspartate (NMDA) receptor, Eur. J. Pharmacol., 164 (1989) 611-612.
17 Casals-Stenzel, J. and Weber, K.H., Triazolodiazepines: disso- ciation of their PAF (platelet-activating factor) antagonistic and CNS activity, Br. J. Pharmacol., 90 (1987) 139-146.
18 Choi, D.W., Calcium-mediated neurotoxicity: relationship to specific channel types and role in ischemic damage, Trends Neurosci., 11 (1988) 465-469.
19 Davies, S.N., Martin, D., Millar, J.D., Aram, J.A., Church, J. and Lodge, D., Differences in results from in vivo and in vitro studies on the use-dependency of N-methylaspartate antagonism by MK-801 and other phencyclidine receptor ligands, Eur. J. Pharmacol., 145 (1988) 141-151.
20 De Leo, J., Schubert, P. and Kreutzberg, G.W., Propentofylline (HWA 285) protects hippocampal neurons of Mongolian gerbils against ischemic damage in the presence of an adenosine antagonist, Neurosci. Lett., 84 (1988) 307-311.
21 Downes, C.P., Lewis, P.E. and Stone, M.A., A comparison of the binding of o opioids and phencyclidine, and the interaction with antipsychotic drugs in rat brain membranes, Br. J. Pharmacol., 88 (1986) 231-237.
22 Dragunow, M. and Faull, R.L.M., Neuroprotective effects of adenosine, Trends Pharmacol. Sci., 9 (1988) 193-194.
23 Dubois, A., Etude Autoradiographique de la Localisation Ana- tomique et Cellulaire et de la Plasticit~ des Sites de Liaison de Type P&iph~rique pour les Benzodiaz~pines (Sites to3) dans le Systdme lmmunitaire et dans le Systdme Nerveux Central chez le Rongeur, PhD Thesis, Paris VI University, 1988.
24 Dubois, A., Benavides, J., Peny, B., Duverger, D., Fage, D., Gotti, B., MacKenzie, E.T. and Scatton, B., Imaging of primary and remote ischaemic and excitotoxic brain lesions. An autora- diographic study of peripheral type benzodiazepine binding sites in the rat and cat, Brain Research, 445 (1988) 77-90.
25 Duverger, D., Benavides, J., Cudennec, A., MacKenzie, E.T., Scatton, B., Seylaz, J. and Verrecchia, C., A glutamate antagonist reduces infarction size following focal cerebral isch- aemia independently of vascular and metabolic changes, J. Cereb. Blood Flow Metabol., 7 (1987) S144.
26 Duverger, D., Lecoffre, C. and MacKenzie, E.T., Histological quantification of cerebral infarction following middle cerebral artery occlusion in various rat strains, J. Cereb. Blood Flow Metab., 5 (1985) S145.
27 Foster, A.C., Gill, R. and Woodruff, G.N., Neuroprotective effects of MK-801 in vivo: selectivity and evidence for delayed degeneration mediated by NMDA receptor activation, J. Neu- rosci., 8 (1988) 4745-4754.
28 Foster, A.C. and Wong, E.H.E, The novel anticonvulsant MK-801 binds to the activated state of the N-methyl-D-aspartate receptor in rat brain, Br. J. Pharmacol., 91 (1987) 403-409.
29 Germano, I.M., Pitts, L.H., Meldrum, B.S., Bartkowski, H.M. and Simon, R.P., Kynurenate inhibition of cell excitation decreases stroke size and deficits, Ann. Neurol., 22 (1987) 730-734.
30 Gill, R., Foster, A.C. and Woodruff, G.N., Systemic adminis- tration of MK-801 protects against ischemia-induced hippo- campal neurodegeneration in the gerbil, J. Neurosci., 7 (1987) 3343-3349.
31 Goldberg, M.P., Viseskul, V. and Choi, D.W., Phencyclidine receptor ligands attenuate cortical neuronal injury after N- methyI-D-aspartate exposure or hypoxia, J. Pharrnacol. Exp. Ther., 245 (1988) 1081-1087.
32 Goldberg, M.P., Weiss, J.H., Pham, P.C. and Choi, D.W., N-methyl-D-aspartate receptors mediate hypoxic neuronal injury in cortical culture, J. Pharmacol. Exp. Ther., 243 (1987) 784-791.
33 Goodman, R.R., Kuhar, M.J., Hester, L. and Snyder, S.H., Adenosine receptors: autoradiographic evidence for their loca-
tion on axon terminals of excitatory neurons, Science, 220 (1983) 967-969.
34 Gotoh, O., Mohamed, A.A., McCulloch, J., Graham, D.I., Harper, A.M. and Teasdale, G.M., Nimodipine and the hae- modynamic and histopathologicai consequences of middle cere- bral artery occlusion in the rat, J. Cereb. Blood Flow Metabol., 6 (1986) 321-331.
35 Gotti, B., Duverger, D., Bertin, J., Carter, C., Dupont, R., Frost, J., Gaudilli~re, B., MacKenzie, E.T., Rousseau, J., Scatton, B. and Wick, A., Ifenprodil and SL 82.0715 as cerebral anti-ischemic agents. I. Evidence for efficacy in models of focal cerebral ischaemia, J. Pharmacol. Exp. Ther., 247 (1988) 1211-1221.
37 Gotti, B., MacKenzie, E.T. and Young, A.R., The pharmaco- therapy of focal cerebral ischaemia - - development of an experimental model, J. Chron. Dis. Ther. Res., 8 (1984) 44-61.
38 Greenamyre, J.T., The role of glutamate in neurotransmission and in neurologic disease, Arch. Neurol., 43 (1986) 1058-1063.
39 Hagberg, H., Andersson, P., Kjellmer, I., Thiringer, K. and Thordstein, M., Extracellular overflow of glutamate, aspartate, GABA and taurine in the cortex and basal ganglia of fetal lambs during hypoxia-ischaemia, Neurosci. Lett., 78 (1987) 311-317.
40 Hagberg, H., Andersson, P., Lacarewicz, J., Jacobson, I., Butcher, S. and Sandberg, M., Extracellular adenosine, inosine, hypoxanthine, and xanthine in relation to tissue nucleotides and purines in rat striatum during transient ischemia, J. Neurochem., 49 (1987) 227-231.
41 Kemp, J.A., Foster, A.C., Leeson, P.D., Priestly, T., Tridgett, R., Iversen L.L. and Woodruff, G.N., 7-Chlorokynurenic acid is a selective antagonist at the glycine modulatory site of the N-methyl-D-aspartate receptor complex, Proc. Natl. Acad. Sci. U.S.A., 85 (1988) 6547-6552.
42 Koek, W., Woods, J.H. and Winger, G.D., MK-801, a proposed noncompetitive antagonist of excitatory amino acid neurotrans- mission, produces phencyclidine-like behavioral effects in pi- geons, rats and rhesus monkeys, J. Pharmacol. Exp. Ther., 245 (1988) 969-974.
43 Kornecki, E. and Ehrlich, Y.H., Neuroregulatory and neuro- pathological actions of the ether-phospholipid platelet-activating factor, Science, 240 (1988) 1792-1794.
44 Kornecki, E., Ehrlich, Y.H. and Lenox, R.H., Platelet-acti- vating factor-induced aggregation of human platelets specifically inhibited by triazolobenzodiazepines, Science, 226 (1985) 1454- 1456.
45 Langer, S.Z. and Arbilla, S., Imidazopyridines as a tool for the characterization of benzodiazepine receptors: a proposal for a pharmacological classification as omega receptor, Pharmacol. Biochem. Behav., 29 (1988) 763-766.
46 MacKenzie, E.T., Duverger, D., Gotti, B. and Nowicki, J.P., Cerebral anti-ischaemic agents: problems and approaches. In J. Krieglstein (Ed.), Pharmacology of Cerebral lschemia, Elsevier, 1986, pp. 167-179.
47 Mayer, M.L. and Westbrook, G.L., The physiology of excitatory amino acids in the vertebrate central nervous system, Prog. Neurobiol., 28 (1987) 197-276.
49 Mohamed, A.A., Gotoh, O., Graham, D.I., Osborne, K.A., McCulloch, J., Mendelow, A.D., Teasdale, G.M. and Harper, A.M., Effect of pretreatment with the calcium antagonist nimodipine on local cerebral blood flow and histopathology after middle cerebral artery occlusion, Ann. Neurol., 18 (1985) 705-711.
50 Murphy, D.E., Hutchison, A.J., Hurt, S.D., Williams, M. and Sills, M.A., Characterization of the binding of [3H]CGS 19755: a novel N-methyl-D-aspartate antagonist with nanomolar affinity in rat brain, Br. J. Pharmacol., 95 (1988) 932-938.
51 Nowicki, J.P., MacKenzie, E.T. and Spinnewyn, B., Effects of agents used in the pharmacotherapy of cerebrovascular disease on the oxygen consumption of isolated cerebral mitochondria, J. Cereb. Blood Flow Metabol., 2 (1982) 33-40.
52 O'Shaughnessy, C.T., Aram, J.A. and Lodge, D., A~ adenosine receptor-mediated block of epileptiform activity induced in zero magnesium in rat neocortex in vitro, Epilepsy Res., 2 (1988) 294-301.
53 Ozyurt, E., Graham, D.I., Woodruff, G.N. and McCulloch, J., Protective effect of the glutamate antagonist, MK-801 in focal cerebral ischemia in the cat, J. Cereb. Blood Flow Metabol., 8 (1988) 138-143.
54 Park, C.K., Nehls, D.G., Graham, D.I., Teasdale, G.M. and McCulloch, J., Focal cerebral ischaemia in the cat: treatment with the glutamate antagonist MK-801 after induction of ischaemia, J. Cereb. Blood Flow Metabol., 9 (1988) 757-762.
55 Park, C.K., Nehls, D.G., Graham, D.I., Teasdale, G.M. and McCulloch, J., The glutamate antagonist, MK-801 reduces focal ischaemic brain damage in the rat, Ann. Neurol., 24 (1988) 543-551.
56 Park, T.S., Van Wylen, D.G.L., Rubio, R. and Berne, R.M., Increased brain interstitial fluid adenosine concentration during hypoxia in newborn piglet, J. Cereb. Blood Flow Metabol., 7 (1987) 178-183.
57 Piercey, M.F., Hoffmann, W.E. and Kaczkofsky, P., Functional evidence for PCP-like effects of the anti-stroke candidate MK-801, Psychopharmacology, 96 (1988) 561-562.
58 Piercey, M.F. and Ray, C.A., Dramatic limbic and cortical effects mediated by high affinity PCP receptors, Life Sci., 43 (1988) 379-385.
59 Prestwich, S.A., Forda, S.R. and Dolphin, A.C., Adenosine antagonists increase spontaneous and evoked transmitter release from neuronal cells in culture, Brain Research, 405 (1987) 130-139.
60 Price, G.W., Ahier, R.G., Middlemiss, D.N., Singh, L., Tricklebank, M.D. and Wong, E.H,F., In vivo labelling of the NMDA receptor channel complex by [3H]MK-801, Eur. J. Pharrnacol., 158 (1988) 279-282.
61 Quirion, R., Chicheportiche, R., Contreras, P.C., Johnson, K.M., Lodge, D., Tam, S.W., Woods, J.H. and Zukin, S.R., Classification and nomenclature of phencyclidine and sigma receptor sites, Trends Neurosci., 10 (1987) 444-446.
62 Robinson, M.B. and Coyle, J.T., Glutamate and related acidic excitatory neurotransmitters: from basic science to clinical application, FASEB J., 1 (1987) 446-455.
63 Rothman, S.M. and Olney, J.W., Glutamate and the patho- physiology of hypoxic-ischemic brain damage, Ann. Neurol., 19 (1986) 105-111.
64 Rudolphi, K.A., Keil, M. and Hinze, H.J., Effect of theophyl- line on ischemically induced hippocampal damage in mongolian gerbils: a behavioral and histopathological study, J. Cereb. Blood Flow Metabol., 7 (1987) 74-81.
65 Sanger, D.J. and Jackson, A., The effects of phencyclidine and
307
other NMDA antagonists on the schedule-controlled behavior of rats, J. Pharmacol. Exp. Ther., 248 (1989) 1215-1221.
66 Scatton, B. and Carter, C., R61e des acides amines excitateurs dans la pathophysiologie des dommages isch6miques cEr6braux, Circ. Metab. Cerv., 6 (1989) 185-210.
67 Schoemaker, H., Boles, R.G., Horst, W.D. and Yamamura, H.I., Specific high affinity binding sites for [3H]Ro5-4864 in rat brain and kidney, J. Pharmacol. Exp. Ther., 225 (1983) 61-69.
68 Shinozaki, H. and Ishida, M., A new potent channel blocker: effects on glutamate responses at the crayfish neuromuscular junction, Brain Research, 372 (1986) 260-268.
69 Siesjo, B.K., Historical overview: calcium, ischaemia, and death of brain cells, Ann. N.Y. Acad. Sci., 522 (1988) 638-661.
70 Simon, R.P., Swan, J.H., Griffiths, T. and Meldrum, B.S., Blockade of N-methyl-D-aspartate receptors may protect against ischemic damage in the brain, Science, 226 (1984) 850-852.
71 Sonders, M.S., Keana, J.F.W. and Weber, E., Phencyclidine and psychotomimetic sigma opiates: recent insights into their bio- chemical and physiological sites of action, Trends Neurosci., 11 (1988) 37-40.
72 Spinnewyn, B., Blavet, N., Clostre, F., Bazan, N. and Braquet, P., Involvement of platelet-activating factor (PAF) in cerebral post-ischemic phase in mongolian gerbils, Prostaglandins, 34 (1987) 337-349.
73 Syapin, P.J. and Skolnick, P., Characterization of benzodiaze- pine binding sites in cultured cells of neuronal origin, J. Neurochem., 32 (1979) 1047-1051.
74 Timmerman, H., Modes and mechanisms of action of vasoactive drugs and especially of cyclandelate, Br. J. Clin. Pract., 38 Suppl. (1984) 10-19.
75 Vanhoutte, P.M. and Paoletti, R., The WHO classification of calcium antagonists, Trends Pharmacol. Sci., 8 (1987) 4-5.
76 Wagener, G. and Bielenberg, G.W., PAF antagonists reduce infarct size in focal ischaemia in the rat brain. In Second International Symposium on Pharmacology of Cerebral Ischa- emia, Marburg, 1988.
77 Watkins, J.C. and Evans, R.H., Excitatory amino acid trans- mitters, Ann. Rev. Pharmacol., 21 (1981) 165-204.
78 Welsh, F.A., Sakamoto, T., McKee, A.E. and Sims, R.E., Effect of lactacidosis on pyridine nucleotide stability during ischemia in mouse brain, J. Neurochem., 49 (1987) 846-851.
79 Weinstein, P.R., Anderson, G.G. and Telles, D.A., Neurolog- ical deficit and cerebral infarction after temporary middle cerebral artery occlusion in unanesthetized cats, Stroke, 17 (1986) 318-324.
80 Wong, B.Y., Coulter, D.A., Choi, D.W. and Prince, D.A., Dextrorphan and dextromethorphan, common antitussives, are antiepileptic and antagonize N-methyl-D-aspartate in brain slices, Neurosci. Lett., 85 (1988) 216-266.
81 Young, A.R~, Bouloy, M., Boussard, J.F., Edvinsson, L. and MacKenzie, E.T., Direct vascular effects of agents used in the pharmacotherapy of cerebrovascular disease on isolated cerebral vessels, J. Cereb. Blood Flow Metabol., 1 (1981) 117-128.
![[Intestinal ischaemia caused by acute mesenteric vein thrombosis]](https://static.documents.pub/doc/80x56/635a2bbf9d85dc43cb06d7e9/intestinal-ischaemia-caused-by-acute-mesenteric-vein-thrombosis.jpg)
