Brain, Vol. 125, No. 1, 102-112,
January 1, 2002
© 2002 Oxford University Press
Endothelin-1 potently induces Leão’s cortical spreading depression in vivo in the rat
A model for an endothelial trigger of migrainous aura?
Departments of 1Experimental Neurology, 2Neurology and 4Neurophysiology, Charité, Humboldt-University and 3Institute of Clinical Pharmacology and Toxicology, Benjamin Franklin Hospital, Free University of Berlin, Berlin, Germany Correspondence to: Jens P. Dreier, Department of Neurology, Charité, Humboldt Universität, Schumannstr. 20/21, 10117 Berlin, Germany E-mail: jens.dreier{at}charite.de
Received March 2, 2001. Revised May 22, 2001. Second revision August 13, 2001. Accepted August 30, 2001. .
 |
Summary |
|---|
 Top Summary IntroductionMaterial and methodsResultsDiscussionReferences |
|---|
According to the ‘neuronal’ theory, cortical spreading depression (CSD) is the pathophysiological correlate of migrainous aura. However, the ‘vascular’ theory has implicated altered vascular function in the induction of aura symptoms. The possibility of a vascular origin of aura symptoms is supported, e.g. by the clinical observation that cerebral angiography frequently provokes migrainous aura. This suggests that endothelial irritation may somehow initiate one of the pathways resulting in migrainous aura. Up to now, an endothelium-derived factor has never been shown to trigger CSD. Here, for the first time, we demonstrate and characterize the ability of the vasoconstrictor and astroglial/neuronal modulator endothelin-1 to trigger Leão’s ‘spreading depression of activity’ in vivo in rats. At a concentration range between 10 nM and 1 µM, endothelin-1 induced changes characteristic of CSD with regard to the rate of propagation, steady (direct current) potential and extracellular K+-concentration. A spreading hyperaemia followed by oligaemia was observed similar to those in K+-induced CSD. Endothelin-1 did not provoke changes characteristic of a terminal depolarization. The mechanism by which endothelin-1 generated CSD involved the N-methyl-D-asparate receptor. Cerebral blood flow decreased slightly, but significantly, before endothelin-1 generated CSD. A vasodilator (NO·-donor) shifted the threshold for CSD induction to higher concentrations of endothelin-1. Endothelin-1, in contrast to K+, did not induce CSD in rat brain slices suggesting indirectly that endothelin-1 may require intact perfusion to exert its effects. In conclusion, endothelin-1 was found in the experiment to be the most potent inducer of CSD currently known. We propose endothelin-1 as a possible candidate for the yet enigmatic link between endothelial irritation and migrainous aura.
Keywords: migraine; aura; cortical spreading depression; vasospasm; endothelin-1
Abbreviations: ACSF= artificial cerebrospinal fluid; CSD = cortical spreading depression; [K+]ACSF = potassium concentration in the artificial cerebrospinal fluid; [K+]0 = extracellular potassium concentration; NMDA = N-methyl-D-aspartate
|
Introduction |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
The hypothesis that cortical spreading depression (CSD) represents the pathophysiological correlate of migrainous aura is based on the characteristic form and development of migrainous visual and sensory disturbances, and measurements of CBF (Lauritzen, 1994
). CSD is a depolarization wave that propagates at a rate of 2–6 mm/min in the cerebral cortex. The phenomenon is triggered electrically, mechanically or by various toxic factors such as a high extracellular K+ concentration disturbing astrocytes and neurones. Pronounced alterations of ionic homoeostasis are associated with CSD, such as a rise of the extracellular potassium concentration ([K+]0) from 3 mM to 
60 mM. The spreading depolarization leads to secondary changes of CBF consisting of hyperaemia followed by oligaemia under physiological conditions.
Based on these characteristics of CSD, it has been proposed that the pathogenesis of migrainous aura is related to primary neuronal or astroglial abnormalities rather than to a vasospastic condition. On the other hand, for a fraction of patients, there are clinical data that favour vascular triggers for migrainous aura, e.g. the induction of migrainous aura by cerebral angiography (Janzen et al., 1972) and in the presence of vascular disease (Olesen et al., 1993; Dichgans et al., 1998). There are also angiographic and Doppler-sonographic observations of short-term vasospasm in large cervicocephalic vessels related to migraine attacks (Call et al., 1988).
The induction of migrainous aura by cerebral angiography raises the question of whether endothelial irritation may somehow provoke CSD. Hypothetically, an endothelial factor may mediate this effect. An interesting candidate in this context is endothelin-1. This peptide is one of the most potent vasoconstrictors currently known (Yanagisawa et al., 1988). It is also reported to increase neuronal excitability and metabolism in both in vivo and in vitro models, to induce ‘convulsive’ behaviour in vivo, and to trigger interastrocytic Ca2+ waves in cell culture (Gross et al., 1992; Venance et al., 1997; Shihara et al., 1998). It has been proposed that interastrocytic Ca2+ waves are critical in the mechanism of CSD (Nedergaard et al., 1995). For these reasons, we investigated the ability of endothelin-1 to induce Leão’s spreading depression of activity.
|
Material and methods |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
In vivo experiments
All animal experiments were approved by the Landesamt für Arbeitsschutz, Gesundheitsschutz und technische Sicherheit, Berlin (G 0346/98). Male Wistar rats (n = 44; 280–380 g) were anaesthetized with 100 mg/kg thiopental sodium intraperitoneally (Trapanal, BYK Pharmaceuticals, Konstanz, Germany), tracheotomized and ventilated artificially (Effenberger Rodent Respirator, Effenberger Med.-Techn. Gerätebau, Pfaffing/Attel, Germany). The left femoral artery and vein were cannulated and a saline solution was infused continuously at 1 ml/h. Body temperature was maintained at 38.0 ± 0.5°C using a heating pad. Systemic arterial pressure (RFT Biomonitor, Zwönitz, Germany) and endexpiratory carbon dioxide pressure (Heyer carbon dioxide monitor EGM I, Bad Ems, Germany) were monitored. Arterial partial pressure of oxygen, arterial partial pressure of carbon dioxide and pH were measured serially using a Compact 1 Blood Gas Analyser (AVL Medizintechnik GmbH, Bad Homburg, Germany). Since the rats were not paralysed, the adequacy of the level of anaesthesia was assessed by testing motor responses to tail pinch. Changes of blood pressure in response to tail pinch were also used to control anaesthesia. Further thiopental doses (25 mg/kg intraperitoneally) were applied when necessary.
A craniotomy was performed over the somatosensory cortex using a saline-cooled drill. In 18 animals (Groups 4–6), a closed cranial window was implanted as described previously (Dreier et al., 1998). The dura mater was removed. The craniotomy site was covered with a piece of glass cut from a cover slip. Inflow and outflow tubes allowed the brain cortex to be superfused with artificial cerebrospinal fluid (ACSF) at the closed window (Fig. 1).
|
Intracranial pressure and ACSF parameters such as partial pressure of oxygen, partial pressure of carbon dioxide and pH can be better controlled in a closed cranial window preparation than in an open cranial window preparation. However, only open cranial windows allow the use of ion-selective microelectrodes. In addition, if ACSF of different composition is superfused at two different windows, the solutions at the two cortical areas are better separated in the open cranial window setting. Therefore, in the remaining 26 animals, one or two open cranial windows were implanted (Groups 1–3).
Unless otherwise specified, the composition of the ACSF was Na+ 152 mM, K+ 3 mM, Ca2+ 1.5 mM, Mg2+ 1.2 mM, HCO3– 24.5 mM, Cl– 135 mM, glucose 3.7 mM and urea 6.7 mM. The ACSF was equilibrated with a gas mixture containing 6.6% oxygen, 5.9% carbon dioxide and 87.5% nitrogen. A partial oxygen pressure of 90–130 mmHg, a partial carbon dioxide pressure of 35–45 mmHg and a pH of 7.35–7.45 were accepted as physiological. CBF was continuously monitored by one to three laser Doppler flow probes (Perimed AB, Järfälla, Sweden). The DC potential was measured either by an Ag–AgCl electrode in the subarachnoid space or by K+-sensitive microelectrodes in the cortex. The latter also recorded [K+]0. Ion-selective/reference microelectrodes were manufactured and tested as reported previously (Dreier et al., 1991). Electrodes were connected to a differential amplifier (Jens Meyer, Munich, Germany). CBF, DC potential and [K+]0 were recorded continuously using a PC and a chart recorder (DASH IV, Astro-Med, Inc., West Warwick, RI, USA). Animals were killed immediately after the experiment by intravenous administration of KCl solution.
Drugs and solutions
Endothelin-1 was purchased from Sigma Chemicals (Deisenhofen, Germany), the NO-donor (spermine/NO·) and the N-methyl-D-aspartate (NMDA) receptor antagonist, MK-801, from Sigma/RBI (Deisenhofen, Germany).
Brain slices
The experiments were performed using 13 slices from 13 different male Wistar rats (150–200 g). Combined entorhinal cortex–hippocampal slices were prepared as reported previously (Dreier and Heinemann, 1991). The brain was removed and washed in cold (5–8°C) ACSF after decapitation under deep ether anaesthesia. Near horizontal slices (400 µm) were cut using a vibratome (752 M vibroslice, Campden Instruments, Loughborough, UK). The slices included ventral hippocampal formation, entorhinal and neocortex.
Slices were transferred into an interface recording chamber and perfused continuously with prewarmed (35–36°C), carbogenated ACSF containing Ca2+ 2 mM, Mg2+ 2 mM, HCO3– 26 mM, Cl– 133 mM, glucose 10 mM, SO22– 2 mM, H2PO4– 1.25 mM (pH 7.4). The K+ concentrations (5 and 25 mM) determined the Na+ concentrations (151 and 131 mM, respectively). A warmed, humidified 95% oxygen and 5% carbon dioxide mixture was directed over the surface of the slices. The recordings were started after >1 h of slice equilibration in the ACSF. To test slice viability, glass-insulated bipolar platinum wire stimulation electrodes were placed into the Schaffer collaterals and a microelectrode measuring the extracellular field potential was inserted into the pyramidal layer of area CA1 of the hippocampus. Slices were accepted when they responded to a paired pulse at 50 ms interstimulus interval, with single population spikes of >3 mV amplitude displaying frequency potentiation. A K+-selective microelectrode was then positioned in the neocortex and another in the entorhinal cortex. Endothelin-1 was applied at increasing concentrations from 1 nM to 100 nM (n = 4) and 10 nM to 1 µM (n = 6) to test whether CSDs are induced. Equilibration time for each concentration was 1 h. Thereafter, ACSF containing endothelin-1 was replaced by ACSF with no endothelin-1 but a K+ concentration ([K+]ACSF) of 25 mM to trigger CSD. In three slices, a droplet application into the perfusion medium at a distance of 100 µm near layer I of the medial entorhinal cortex was performed using a pipette filled with either endothelin-1 (100 µM) or KCl (200 mM).
Data analysis
Data were analysed by comparing relative changes of CBF and absolute changes of the DC potential and [K+]0. CBF changes were calculated in relation to the baseline at the onset of the experiment (= 100%). All data are given as mean ± standard deviation. Statistical comparisons were performed as specified in the text using the paired t-test, Wilcoxon signed rank test, Mann–Whitney U-test, Fisher’s exact test, analysis of variance for repeated measures with Scheffé’s post hoc test or Kruskal–Wallis H-test with post hoc Newman–Keuls test. P < 0.05 was accepted as statistically significant.
|
Results |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Systemic variables
The systemic variables remained within physiological limits throughout the experiments.
Endothelin-1 induces CSD in vivo—experiments with open cranial window
Endothelin-1 induces changes in CBF, DC potential and [K+]0 typical of CSD (Group 1)
To simultaneously measure CBF, intracortical DC potential and [K+]0 during brain topical superfusion of endothelin-1, an open cranial window was implanted over the parietal cortex using rats under thiopental anaesthesia (n = 7). Two K+-sensitive microelectrodes were positioned 2 mm apart at a cortical depth of 300 µm. A laser Doppler flow probe was placed close to the caudal microelectrode. Superfusion of ACSF, containing endothelin-1 at 100 nM, initially decreased CBF to 88 ± 10% of baseline. This was accompanied by a slow increase of [K+]0 from 3.0 to 4.4 ± 0.6 mM at three of the 14 electrode positions. In five out of seven animals, one to three transient increases of CBF to 139 ± 32% lasting for 91 ± 35 s occurred 21 ± 3 min after wash-in of endothelin-1 (Fig. 2A). The first transient increase was followed by a CBF decrease to 63 ± 14%. A sharp negative DC shift by –21 ± 3 mV and steep rise of [K+]0 from 3 to 58 ± 10 mM, lasting for 84 ± 47 s, simultaneously occurred with the transient CBF increase. The rise of [K+]0 was followed by an undershoot to 2.4 ± 0.3 mM. Between the two microelectrode positions, the onsets of the negative DC shift were separated by 22 ± 16 s. In summary, the pattern of the spreading DC negativity, rise and undershoot of [K+]0, short initial hyperaemia followed by oligaemia fulfilled the criteria typical of Leão’s cortical spreading depression (Lauritzen, 1994).
|
The ACSF concentration of endothelin-1 was then increased to 1 µM after 60 min. Starting at a CBF level of 74 ± 15%, all animals developed one to five CSDs under this condition (Fig. 2B). The CSDs were preceded by a slow [K+]0 increase to 4.5 ± 0.6 mM at seven of the 14 electrode positions. When endothelin-1 was superfused at 10 µM, CSDs started at a CBF level of 63 ± 9%. Slow rises of [K+]0 to 4.6 ± 1.0 mM were found, before CSD, at 11 of the 14 electrode positions. In summary, with increasing endothelin-1 concentrations, the CBF level decreased significantly and [K+]0 increased slightly, but significantly, before the occurrence of CSD (P < 0.05, analysis of variance for repeated measures with Scheffé’s post hoc test). Endothelin-1 was a robust stimulus for CSD. Application of endothelin-1 at 1 µM elicited a CSD in every experiment.
Endothelin-1 induced depolarizations propagate like CSD (Group 2)
To calculate the rate of propagation of endothelin-1 induced CSDs, a second open cranial window was implanted. Subarachnoid DC potential and cortical CBF were measured simultaneously at each window using Ag–AgCl electrodes and laser Doppler flow probes. The two recording sites were separated by 8 mm. While ACSF containing endothelin-1 (10 nM–1 µM) was superfused at window 1, physiological ACSF was applied at window 2. Endothelin-1 generated CSD at a concentration of 100 nM in two out of six cases and only at 1 µM in the rest. The rate of propagation of the first CSD was 3.9 ± 0.2 mm/min. Before the first CSD, CBF reached a level of 74 ± 20% at window 1. At the same time, a level of 111 ± 14% was detected at the control window. The difference was statistically significant (P < 0.01, paired t-test). The cortical spreading hyperaemia was significantly smaller when endothelin-1 was present in the ACSF compared with the response at the window where physiological ACSF was superfused (171 ± 50% versus 265 ± 123%, P < 0.05, Wilcoxon signed-rank test). The negative shift of the subarachnoid DC potential was not larger (4.9 ± 1.9 mV versus 4.0 ± 1.3 mV), but it was significantly longer (104 ± 30 s versus 87 ± 24 s, P < 0.01, paired t-test) under endothelin-1.
K+ and endothelin-1 induced CSD propagate at a similar rate (Group 3)
To compare the threshold and propagation of endothelin-1- and K+-induced CSD, high [K+]ACSF was applied instead of endothelin-1 in 13 animals. Nine of the 13 animals developed CSD in response to [K+]ACSF at 100 mM, the remaining four only in the presence of [K+]ACSF at 250 mM. The rate of CSD propagation was similar to that under endothelin-1 (3.8 ± 0.5 mm/min). Before the first CSD, CBF increased significantly in response to high [K+]ACSF compared with the control window (123 ± 37% versus 102 ± 15%, P < 0.05, paired t-test). The cortical spreading hyperaemia was not significantly different (high [K+]ACSF 250 ± 91% versus control 277 ± 87%). The negative DC shift was both significantly larger (6.1 ± 1.6 mV versus 3.8 ± 1.3 mV, P < 0.01, paired t-test) and longer (113 ± 40 s versus 85 ± 23 s, P < 0.05, paired t-test) compared with the response in the control window. A similar delay of repolarization in presence of high baseline [K+]ACSF is known with rat brain slices (Dreier et al., 2001).
Endothelin-1 induces CSD in vivo—experiments with closed cranial window
Control group (Group 4)
In the following experiments, a closed cranial window technique was used (cf. Fig. 1A and Fig. 1B). An Ag–AgCl electrode was inserted into the window to measure the subarachnoid DC potential. Two laser Doppler flow probes were positioned over the caudal and rostral thirds of the cranial window. Different fibre separations in the laser probes (500 µm and 140 µm in the caudal, and 250 µm in the rostral) allowed us to measure CBF in different cortical depths. This gave estimated maximal measurement depths of 2 mm (laser probe with 500 µm fibre separation), 1 mm (laser probe with 250 µm fibre separation) and 0.5 mm (laser probe with 140 µm fibre separation) (Fabricius et al., 1997). Brain topical superfusion of ACSF, containing endothelin-1 at concentrations from 10 nM to 1 µM, decreased CBF similarly in the three tissue volumes (n = 6). CBF was reduced to 79 ± 16% by endothelin-1 (1 µM) at the caudal window area/fibre separation 500 µm (P < 0.05, analysis of variance for repeated measures with Scheffé’s post hoc test), to 89 ± 33% at the caudal window area/fibre separation 140 µm (P < 0.05) and to 90 ± 15% at the rostral window area/fibre separation 250 µM (P < 0.05). Thus, the average CBF decline at different cortical sites of the window area appeared to be rather homogeneous.
Endothelin-1 did not induce CSD at 10 nM in any of the six animals. Five of the six animals generated CSD at 100 nM endothelin-1. All the six animals developed CSD when endothelin-1 was increased to 1 µM (Figs 3 and 4). The CSD-induced hyperaemia showed a delay of 42 ± 19 s between the caudal and rostral recording sites (3 mm distance). Examples of original traces from such an experiment are given in Fig. 3A and B. While Fig. 3A shows a survey for the complete experiment, Fig. 3B demonstrates a higher temporal resolution of one of the CSDs. Note the spreading of the CBF response from Laser-Doppler I (caudal laser probe, fibre separation 500 µm) to Laser-Doppler II (frontal laser probe) in Fig. 3B. The Group 4 experiments also served as controls for Groups 5 and 6.
|
The vasodilator spermine/NO· shifts the threshold for generation of endothelin-1 induced CSDs to higher concentrations of endothelin-1 (Group 5)
To further investigate whether the vasoconstrictive property of endothelin-1 might be responsible for the generation of CSDs, we co-applied the vasodilator spermine/NO· at a saturating concentration of 100 µM together with increasing concentrations of endothelin-1 (10 nM to 1 µM) (n = 6). Only one laser Doppler probe was used (fibre separation 250 µm). Co-application of the NO·–donor significantly increased CBF levels in response to administration of endothelin-1 compared with the controls (P < 0.01, Mann–Whitney U-test and Kruskal–Wallis H-test with post hoc Newman–Keuls test, respectively, Fig. 4). Following the application of endothelin-1 at 1 µM and spermine/NO·, CBF was 223 ± 94% of baseline. The NO·-donor also shifted significantly the generation of CSDs to higher concentrations of endothelin-1 (P < 0.01, Fisher’s exact test, Fig. 4). However, in response to endothelin-1 at 1 µM, all animals generated CSDs despite significantly increased CBF levels induced by spermine/NO· (Fig. 4).
|
The NMDA receptor antagonist MK-801 inhibits the generation of endothelin-1 induced CSDs (Group 6)
NMDA receptors are involved in the generation and propagation of CSDs (Lauritzen, 1994
). To test whether NMDA receptors also play a role in the generation of CSD by endothelin-1, we administered MK-801 intravenously (5 mg/kg body weight), together with topical application of increasing concentrations of endothelin-1 (10 nM to 1 µM) (n = 6). This NMDA receptor antagonist completely blocked the generation of CSD in five out of six animals (P < 0.01, Fisher’s exact test), while CBF was not altered significantly (Fig. 4).
Endothelin-1 does not induce CSD in rat brain slices
Our in vivo data did not provide unequivocal evidence for either a vascular or a primarily neuronal/astroglial mechanism by which endothelin-1 generated CSD. Brain slices can be used to separate experimentally the direct effects on the neuronal/astroglial network from secondary effects in response to vascular alterations. Therefore, we tested whether endothelin-1 would induce CSD in rat brain slices containing parts of the temporal neocortex, entorhinal cortex and hippocampal formation (Dreier and Heinemann, 1991). A K+-selective microelectrode was placed in the neocortex and a DC-electrode in the entorhinal cortex. Using a similar protocol to the in vivo experiments and under standard conditions, endothelin-1 was applied at concentrations from 1 nM to 100 nM in four slices and 10 nM to 1 µM in another six slices. CSDs were not triggered by endothelin-1 in any of the 10 slices. After endothelin-1 had been washed out, the baseline K+ concentration of the ACSF had increased from 5 mM to 25 mM. This yielded typical CSDs in all 10 slices (negative DC shift: –12 ± 6 mV; rise of [K+]0 to 53 ± 8 mM) (Fig. 2C). In our in vivo experiments, the threshold for the generation of CSD by increased baseline [K+]ACSF was 100 mM. This supported the hypothesis that the failure of endothelin-1 to induce CSD in slices was not due to a higher threshold for CSD compared with the in vivo situation. The data suggested that the direct effects of endothelin-1 on the neuronal astroglial network were not sufficient to induce CSD. In three additional slices, droplet application of endothelin-1 at 100 µM did not provoke CSD. This was in contrast to a droplet application of K+ at 200 mM.
|
Discussion |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
We found that endothelin-1 induced CSD in rats. The assumption that the short-lasting depolarizations induced by endothelin-1 represented CSDs is based on the propagation and the typical patterns of CBF, DC potential, rise and undershoot of [K+]0, and inhibition by the NMDA receptor antagonist MK-801 (cf. Lauritzen, 1994
). K+-induced CSDs started from a slightly increased CBF level and endothelin-1 induced CSDs from a slightly decreased level. A discriminating feature was that endothelin-1, in contrast to K+, did not induce CSD in brain slices.
We found that endothelin-1 is the most potent inducer of CSD currently known. Endothelin-1 started to elicit CSD between 10 nM and 1 µM. Concentrations in this range and higher have been applied in the majority of earlier animal studies using endothelin-1. Hence, CSD was probably a significant, but unidentified, confounding factor in these investigations. Indeed, signs of CSD are encountered in previous reports, e.g. CBF changes typical of CSD are shown in a figure by Willette and Sauermelch (1990) in response to cortical microinjection of endothelin-1 in rats.
Endothelin-1 has been implicated in the pathogenesis of ischaemic stroke and vasospasm after subarachnoid haemorrhage (Ferro and Webb, 1996). Possibly, endothelin-1 is involved in the generation of peri-infarct depolarizations after ischaemic stroke (Strong et al., 2000) and cortical spreading ischaemia when red blood cell products are present in the subarachnoid space (Dreier et al., 1998, 2000).
Endothelin-1 modifies the function of endothelial cells, vascular smooth muscle, neurones and astrocytes. In our study, a vascular hypothesis and an astroglial/neuronal hypothesis were tested with regard to the cellular targets responsible for endothelin-1 induced CSDs.
The vascular hypothesis
Others have shown that endothelin-1 can reduce CBF below the ischaemic threshold. For example, Macrae et al. (1993) observed significant ischaemic damage at concentrations of 1 µM and higher with direct application of endothelin-1 to the middle cerebral artery, but found that endothelin-1 was not effective at a concentration of 100 nM. When they applied endothelin-1 topically to the frontoparietal cortex, Fuxe et al. (1997) identified a threshold concentration of 40 µM for a significant decrease in CBF and ischaemic damage.
Our finding that topical superfusion of endothelin-1 at concentrations of 1 µM and below did not decrease average CBF to a level typical of an ischaemic core region was consistent with the report by Fuxe et al. (1997). An anaemic terminal depolarization was correspondingly not induced by endothelin-1 in our experiments. However, comparison of the endothelin-1 concentration ranges applied in the two studies is limited by the fact that the experiments by Fuxe et al. (1997) were performed under halothane anaesthesia, whereas our experiments were performed under barbiturate anaesthesia. Halothane not only directly antagonizes the astroglial and vascular actions of endothelin-1 (Boillet et al., 1995; Venance et al., 1997), but also increases baseline CBF, which barbiturate is known to reduce (Todd and Weeks, 1996; Linde et al., 1999). Absolute CBF levels under barbiturate anaesthesia have been reported to range between 50 and 70 ml/100 g/min in rats (Todd and Weeks, 1996; Linde et al., 1999). Hence, the baseline CBF level is reduced by 40% under barbiturate anaesthesia.
Starting from this low CBF baseline level (e.g. in Group 2 of our study), endothelin-1 reduced CBF to 74% before the first CSD occurred. On a rough estimation, this corresponds to an absolute CBF level of 40–50 ml/100 g/min immediately before the first CSD. In cerebral ischaemia, protein synthesis is inhibited by 50% at 55 ml/100 g/min and is completely suppressed at <35 ml/100 g/min. Energy metabolism begins to be disturbed at 20 ml/100 g/min and an anaemic terminal depolarization occurs at flow rates of 6–15 ml/100 g/min. Hossmann (1994) defined the ischaemic penumbra as a region of constrained blood supply in which the energy metabolism is preserved. According to this definition of penumbra, the estimated CBF value in response to endothelin-1 immediately before the first CSD was possibly in the upper range of that found in an ischaemic penumbra. It is not clear whether a CBF level of this magnitude is sufficiently low to induce, by itself, CSD-like depolarizations similar to those observed in the surrounding of an ischaemic core. Such a hypothesis is questioned by the fact that NO· synthase inhibition by topical superfusion with nitro-L-arginine at 1 mM reduces CBF to a similar level as endothelin-1 at 1 µM, but does not provoke CSDs. However, the K+-threshold for the induction of CSD declines in the presence of NO· synthase inhibition (Dreier et al., 2001).
Another argument against a penumbra-related cause of the endothelin-1 induced CSDs is related to the barbiturate-induced reduction of the CBF baseline, as this occurs in parallel with a decline in glucose metabolism and neuronal activity. Therefore, this fraction of the CBF reduction is not due to a constrained blood supply in the true sense of Hossmann’s penumbra definition.
An argument supporting the presence of energy compromise induced by endothelin-1 in our experiments is related to the small increases of [K+]0 preceding the CSDs. Such small increases of [K+]0 were also observed preceding peri-infarct depolarizations in the penumbra after middle cerebral artery occlusion (Nedergaard and Hansen, 1993), or preceding so-called hypoxic CSD-like depolarizations in rat brain slices (Müller and Somjen, 2000). In hypoxic brain slices, the small K+ increases arose locally due to the release of K+ from neurones via activated K+ channels in response to ATP deficiency. This was accompanied by neuronal hyperpolarization and astroglial depolarization, while the subsequent large rise in [K+]0 at the time of the CSD-like slow potential shift was associated with depolarization of both neurones and astrocytes (Müller and Somjen, 2000). However, it is also possible that the opening of K+ channels by endothelin-1 was due to a mechanism unrelated to the decline of CBF, e.g. direct activation of neuronal ion channels by endothelin-1.
The hyperaemic response to CSDs in the presence of endothelin-1 was only slightly smaller than that of a normal CSD. In contrast, in peri-infarct depolarizations after middle cerebral artery occlusion, the flow response is strongly suppressed because the reduced haemodynamic capacity of the collateral system prevents adequate coupling of the blood supply to the metabolic demand of the tissue (Hossmann, 1994). This argues against an ischaemic mechanism for the endothelin-1 induced CSDs. However, the value of this argument is somewhat limited because, in the endothelin-1 model, the CBF decline is not caused by a mechanical obstacle but via increase of vascular tone. Endothelin receptors are not involved in the hyperaemic response to CSD (Goadsby et al., 1996). Hence, it is conceivable that the CSD related mechanism of vasodilation just manages to overcome the endothelin-1 induced rise of vascular tone for a short period.
In addition, the effects of the NO· donor were ambiguous with respect to the question of whether the CSDs were induced by a primarily vascular or a primarily astroglial/neuronal effect of endothelin-1. A primarily vascular mechanism of endothelin-1 induced CSDs is supported by our observation that the NO· donor shifted the generation of CSDs to higher concentrations of endothelin-1. Neither elicitation nor propagation of CSD triggered by needle stab was found to be altered by an NO·-donor (Kaube et al., 1999). This discordance between endothelin-1 and trauma-induced CSD suggests that the modulating effect of NO· on the endothelin-1 induced CSDs was more related to vasodilation than to a direct effect of NO· on the neuronal/astroglial network such as inhibition of NMDA receptors (Lipton et al., 1993). On the other hand, the vascular theory of endothelin-1 induced CSDs is clearly challenged by our observation that endothelin-1 at 1 µM induced CSDs in presence of NO·/spermine despite an average CBF level of >200% of base line. This finding tends to support the neuronal/astroglial hypothesis.
One proposal to differentiate between CSD and hypoxic CSD-like depolarizations in brain slices is the responsivity of CSD to NMDA receptor antagonists, which do not block hypoxic CSD-like depolarizations (Aitken et al., 1988). However, the blockade of endothelin-1 induced CSDs by MK-801 in our experiments does not contradict a vascular origin since, for example, peri-infarct depolarizations are also inhibited by this drug (Iijima et al., 1992). Possibly, only the propagation into the surrounding tissue was abolished and initiation at the ischaemic spot continued.
Energy compromise leads to a delay in repolarization after CSD (Leão, 1947). Indeed, we observed a slight but significant delay when CSDs at a window superfused with endothelin-1 were compared with those at a control window superfused with physiological ACSF (Group 2). However, we detected a similar delay of repolarization when CSDs in presence of high baseline [K+]ACSF were compared with those at a control window (Group 3). Such a delay of repolarization after CSD in response to high baseline [K+]ACSF was also found in rat brain slices and is possibly related to a down-regulation of Na+/K+-ATPase activity (Dreier et al., 2001). The observed analogy between endothelin-1- and K+-induced CSD makes it uncertain whether the source of the delayed repolarization elicited by endothelin-1 is related to the vascular supply.
The astroglial/neuronal hypothesis
Endothelin-1 has been found to augment neuronal responses to glutamate in brain slices (Shihara et al., 1998). Such effects might promote the generation of CSDs via NMDA receptors. It was also reported that intraventricular administration of small doses of endothelin-1 induced barrel rolling behaviour and a cerebral hypermetabolic state in rats in vivo (Gross et al., 1992). Like the CSDs characterized in our study, this ‘convulsive’ behaviour induced by endothelin-1 was inhibited by MK-801 (Chew et al., 1994). The question of whether processes related to CSD were involved in these behavioural abnormalities deserves further study.
Endothelin-1 is also known to stimulate arachidonic acid metabolism via activation of phospholipases. Using a scraped monolayer of cultured astrocytes, Tabernero et al. (1996) showed that arachidonic acid at 50 µM blocked metabolic trafficking of energy substrates such as glucose similarly to endothelin-1 at 100 nM within the astroglial syncytium. In vivo, but not in brain slices, inhibition of metabolic trafficking might interfere with neuronal nutrition since astrocytes constitute a physical barrier to the supply of metabolic substrates from blood to neurones (Pellerin and Magistretti, 1994). A decreased supply of energy substrates such as glucose is known to promote CSD (Zhang et al., 1990).
Although endothelin-1 is a gap junction inhibitor, it was also found to be a potent inducer of interastrocytic Ca2+ waves (Venance et al., 1997). It has been suggested that interastrocytic Ca2+ waves are involved in the propagation mechanism of CSD (Nedergaard et al., 1995; Kunkler and Kraig, 1998).
Endothelin-1 also seems to interfere with the activity of the Na+/K+-ATPase in the CNS (Shah and Jandhyala, 1993). This may be relevant in the pathogenesis of endothelin-1 induced CSDs since inhibition of the Na+/K+-ATPase is a potent trigger of CSD as demonstrated, for example, in rat brain slices using the Na+/K+-ATPase inhibitor ouabain (Balestrino et al., 1999).
Clinical implications
An association between migraine and vascular disease is supported by all recent population- and hospital-based epidemiological studies demonstrating a slight but significant increased risk of migraine patients developing ischaemic stroke (Tzourio et al., 2000). This risk was significantly higher for migraineurs with aura compared with those without. Although a CSD-induced secondary ischaemia was principally demonstrated in animal experiments, this mechanism could only explain the fraction of migraine-induced strokes (Dreier et al., 1998, 2001). In contrast, for example, the epidemiological association of migraine and extracranial cervicocephalic artery dissection is hardly related to a primary neuronal network dysfunction (D’Anglejan-Chatillon et al., 1989). In this setting, primary vascular changes are more likely to trigger a secondary CSD-like depolarization than vice versa (Olesen et al., 1993). Such a sequence is supported by case reports that migrainous aura occurred as a symptom of carotid artery dissection (Olesen et al., 1993). A primarily vascular process was also suggested by CBF studies using intracarotid 133Xe. The large number of recorded migrainous auras was because the procedure of catheterizing and injecting the carotid artery provoked aura in >50% of migraineurs (Lassen and Friberg, 1991). Similarly, cerebral angiography is known to induce migrainous aura (Janzen et al., 1972; Whitty, 1986).
There are several vascular diseases providing additional arguments for a vascular trigger of migrainous aura in a fraction of patients: CADASIL (cerebral autosomal dominant arteriopathy with subcortical infarcts and leucoencephalopathy) was associated with migraine in 38% of cases. Eighty-seven per cent of the CADASIL patients with migraine suffered from migraine with aura (Dichgans et al., 1998). Migraine patients also carry a significantly increased risk of suffering from Raynaud’s syndrome and symptoms of coronary artery spasm (O’Keefe et al., 1992), suggesting the existence of a generalized vasospastic disorder. There is evidence of an autosomal dominant disorder linking migraine, vascular retinopathy and Raynaud’s syndrome (Terwindt et al., 1998). Migraine has also been associated with hereditary endotheliopathy (Jen et al., 1997) and antiphospholipid antibody syndrome (Silvestrini et al., 1993). Involvement of endothelin-1 has been implicated in Raynaud’s syndrome, coronary artery spasm and antiphospholipid antibody syndrome (Ferro and Webb, 1996; Atsumi et al., 1998).
Our findings support the hypothesis that endothelin-1 is a potential candidate for a link between the endothelium and migrainous aura in those patients in whom an endothelial irritation is suspected to cause the aura, e.g. after cerebral angiography. Several earlier reports implicated the involvement of endothelin-1 in the pathogenesis of migraine based on increased plasma levels during migraine attacks (Gallai et al., 1994; Kallela et al., 1998). A link between migraine and endothelins has also been sugested in a recent population-based study demonstrating an association between migraine and an endothelin type A receptor gene polymorphism (Tzourio et al., 2001).
Conclusion
We found that nanomolar concentrations of endothelin-1 produced CSD in vivo in rats. The findings did not indicate that endothelin-1 induced an ischaemic core at concentrations triggering spreading depolarizations. However, a penumbra-like condition cannot be excluded. Direct neuronal/astroglial targets of endothelin-1 represent other candidates that may play a role in the cascade leading to CSD. Our results add evidence that endothelin-1 is possibly involved in the pathogenesis of migrainous aura. This may be particularly interesting for the subset of migraine patients with increased risk of stroke.
|
Acknowledgements |
|---|
This study was supported by grants DFG-SFB 507 A1 (J.P.D. and U.D.) and DFG EI 207/2–3 (U.D., U.L.). The support of the Hermann and Lilly Schilling Foundation (U.D.) is gratefully acknowledged.
|
References |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Aitken PG, Balestrino M, Somjen GG. NMDA antagonists: lack of protective effect against hypoxic damage in CA1 region of hippocampal slices. Neurosci Lett 1988; 89: 187–92.[ISI][Medline]
Atsumi T, Khamashta MA, Haworth RS, Brooks G, Amengual O, Ichikawa K, et al. Arterial disease and thrombosis in the antiphospholipid syndrome: a pathogenic role for endothelin-1. Arthritis Rheum 1998; 41: 800–7.[ISI][Medline]
Balestrino M, Young J, Aitken P. Block of (Na+,K+)ATPase with ouabain induces spreading depression-like depolarization in hippocampal slices. Brain Res 1999; 838: 37–44.[ISI][Medline]
Boillet A, Vallet B, Marty J, Auclerc A, Barale F. Effects of halothane, enflurane and isoflurane on contraction of rat aorta induced by endothelin-1. Br J Anaesth 1995; 75: 761–7.
Call GK, Fleming MC, Sealfon S, Levine H, Kistler JP, Fisher CM. Reversible cerebral segmental vasoconstriction. [Review]. Stroke 1988; 19: 1159–70.
Chew BH, Weaver DF, Balaban CD, Gross PM. NMDA-mediated metabolic activation of the cerebellar cortex in behaving rats by the neuropeptide endothelin-1. Brain Res 1994; 647: 345–52.[ISI][Medline]
D’Anglejan-Chatillon J, Ribeiro V, Mas JL, Youl BD, Bousser MG. Migraine – a risk factor for dissection of cervical arteries. Headache 1989; 29: 560–1.[ISI][Medline]
Dichgans M, Mayer M, Uttner I, Bruning R, Muller-Hocker J, Rungger G, et al. The phenotypic spectrum of CADASIL: clinical findings in 102 cases. Ann Neurol 1998; 44: 731–9.[ISI][Medline]
Dreier JP, Heinemann U. Regional and time dependent variations of low Mg2+ induced epileptiform activity in rat temporal cortex slices. Exp Brain Res 1991; 87: 581–96.[ISI][Medline]
Dreier JP, Körner K, Ebert N, Görner A, Rubin I, Back T, et al. Nitric oxide scavenging by hemoglobin or nitric oxide synthase inhibition by N-nitro-L-arginine induces cortical spreading ischemia when K+ is increased in the subarachnoid space. J Cereb Blood Flow Metab 1998; 18: 978–90.[ISI][Medline]
Dreier JP, Ebert N, Priller J, Megow D, Lindauer U, Klee R, et al. Products of hemolysis in the subarachnoid space inducing spreading ischemia in the cortex and focal necrosis in rats: a model for delayed ischemic neurological deficits after subarachnoid hemorrhage? J Neurosurg 2000; 93: 658–66.[ISI][Medline]
Dreier JP, Petzold G, Tille K, Lindauer U, Arnold G, Heinemann U, et al. Ischemia triggered by spreading neuronal activation is inhibited by vasodilators in rats. J Physiol (Lond) 2001; 531: 515–26.
Fabricius M, Akgören N, Dirnagl U, Lauritzen M. Laminar analysis of cerebral blood flow in cortex of rats by laser Doppler flowmetry: a pilot study. J Cereb Blood Flow Metab 1997; 17: 1326–36.[ISI][Medline]
Ferro CJ, Webb DJ. The clinical potential of endothelin receptor antagonists in cardiovascular medicine. [Review]. Drugs 1996; 51: 12–27.[ISI][Medline]
Fuxe K, Bjelke B, Andbjer B, Grahn H, Rimondini R, Agnati LF. Endothelin-1 induced lesions of the frontoparietal cortex of the rat. A possible model of focal cortical ischemia. Neuroreport 1997; 8: 2623–9.[ISI][Medline]
Gallai V, Sarchielli P, Firenze C, Trequattrini A, Paciaroni M, Usai F, et al. Endothelin-1 in migraine and tension-type headache. Acta Neurol Scand 1994; 89: 47–55.[ISI][Medline]
Goadsby PJ, Adner M, Edvinsson L. Characterization of endothelin receptors in the cerebral vasculature and their lack of effect on spreading depression. J Cereb Blood Flow Metab 1996; 16: 698–704.[ISI][Medline]
Gross PM, Wainman DS, Espinosa FJ, Nag S, Weaver DF. Cerebral hypermetabolism produced by intraventricular endothelin-1 in rats: inhibition by nimodipine. Neuropeptides 1992; 21: 211–23.[ISI][Medline]
Hossmann K-A. Viability thresholds and the penumbra of focal ischemia. Ann Neurol 1994; 36: 557–65.[ISI][Medline]
Iijima T, Mies G, Hossmann K-A. Repeated negative DC deflections in rat cortex following middle cerebral artery occlusion are abolished by MK-801: effect on volume of ischemic injury. J Cereb Blood Flow Metab 1992; 12: 727–33.[ISI][Medline]
Janzen R, Tanzer A, Zschocke S, Dieckmann H. Postangiographische Spätreaktionen der Hirngefässe bei Migräne-kranken. Z Neurol 1972; 201: 24–42.[ISI][Medline]
Jen J, Cohen AH, Yue Q, Stout JT, Vinters HV, Nelson S, et al. Hereditary endotheliopathy with retinopathy, nephropathy, and stroke (HERNS). Neurology 1997; 49: 1322–30.
Kallela M, Färkkilä M, Saijonmaa O, Fyhrquist F. Endothelin in migraine patients. Cephalalgia 1998; 18: 329–32.[ISI][Medline]
Kaube H, Knight YE, Storer RJ, Hoskin KL, May A, Goadsby PJ. Vasodilator agents and supracollicular transection fail to inhibit cortical spreading depression in the cat. Cephalalgia 1999; 19: 592–7.[ISI][Medline]
Kunkler PE, Kraig RP. Calcium waves precede electrophysiological changes of spreading depression in hippocampal organ cultures. J Neurosci 1998; 18: 3416–25.
Lassen NA, Friberg L. Cerebral blood flow measured by xenon 133 using the intraarterial injection method or inhalation combined with SPECT in migraine research. In: Olesen J, editor. Migraine and other headaches. The vascular mechanisms. New York: Raven Press; 1991. p. 5–13.
Lauritzen M. Pathophysiology of the migraine aura: the spreading depression theory. [Review]. Brain 1994; 117: 199–210.
Leão AAP. Further observations on the spreading depression of activity in the cerebral cortex. J Neurophysiol 1947; 10: 409–14.
Linde R, Schmalbruch IK, Paulson OB, Madesen PL. The Kety–Schmidt technique for repeated measurements of global cerebral blood flow and metabolism in the conscious rat. Acta Physiol Scand 1999; 165: 395–401.[ISI][Medline]
Lipton SA, Choi YB, Pan ZH, Lei SZ, Chen HS, Sucher NJ, et al. A redox-based mechanism for the neuroprotective and neurodestructive effects of nitric oxide and related nitroso-compounds. Nature 1993; 364: 626–32.[Medline]
Macrae M, Robinson MJ, Graham DI, Reid JL, McCulloch J. Endothelin-1 induced reductions in cerebral blood flow: dose dependency, time course, and neuropathological consequences. J Cereb Blood Flow Metab 1993; 13: 276–84.[ISI][Medline]
Müller M, Somjen GG. Na+ and K+ concentrations, extra- and intracellular voltages, and the effect of TTX in hypoxic rat hippocampal slices. J Neurophysiol 2000; 83: 735–45.
Nedergaard M, Hansen AJ. Characterization of cortical depolarizations evoked in focal cerebral ischemia. J Cereb Blood Flow Metab 1993; 13: 568–74.[ISI][Medline]
Nedergaard M, Cooper AJ, Goldman SA. Gap junctions are required for the propagation of spreading depression. J Neurobiol 1995; 28: 433–44.[ISI][Medline]
O’Keefe ST, Tsapatsaris NP, Beetham WP. Increased prevalence of migrain and chest pain in patients with primary Raynaud disease. Ann Intern Med 1992; 116: 985–9.
Olesen J, Friberg L, Olsen TS, Andersen AR, Lassen NA, Hansen PE, et al. Ischemia-induced (symptomatic) migraine attacks may be more frequent than migraine-induced ischemic insults. Brain 1993; 116: 187–202.
Pellerin L, Magistretti PJ. Glutamate uptake into astrocytes stimulates aerobic glycolysis: a mechanism coupling neuronal activity to glucose utilization. Proc Natl Acad Sci USA 1994; 91: 10625–9.
Shah J, Jandhyala BS. Physiological significance of Na(+)/K(+)-ATPase activity in the central nervous system and endogenous sodium-pump inhibitors in the neural regulation of arterial blood pressure. J Cardiovasc Pharmacol 1993; 22 Suppl 2: 51–5.
Shihara M, Hirooka Y, Hori N, Matsuo I, Tagawa T, Suzuki S, et al. Endothelin-1 increases the neuronal activity and augments the responses to glutamate in the NTS. Am J Physiol 1998; 275: R658–65.
Silvestrini M, Cupini LM, Matteis M, De Simone R, Bernardi G. Migraine in patients with stroke and antiphospholipid antibodies. Headache 1993; 33: 421–6.[ISI][Medline]
Strong AJ, Smith SE, Whittington DJ, Meldrum BS, Parsons AA, Krupinski J, et al. Factors influencing the frequency of fluorescence transients as markers of peri-infarct depolarizations in focal cerebral ischemia. Stroke 2000; 31: 214–22.
Tabernero A, Giaume C, Medina JM. Endothelin-1 regulates glucose utilization in cultured astrocytes by controlling intercellular communication through gap junctions. Glia 1996; 16: 187–95.
Terwindt GM, Haan J, Ophoff RA, Groenen SM, Storimans CW, Lanser JB, et al. Clinical and genetic analysis of a large Dutch family with autosomal dominant vascular retinopathy, migraine and Raynaud’s phenomenon. Brain 1998; 121: 303–16.
Todd MM, Weeks J. Comparative effects of propofol, pentobarbital and isoflurane on cerebral blood flow volume. J Neurosurg Anesthesiol 1996; 8: 296–303.[ISI][Medline]
Tzourio C, Kittner SJ, Bousser MG, Alperovitch A. Migraine and stroke in young women. [Review]. Cephalalgia 2000; 20: 190–9.[ISI][Medline]
Tzourio C, El Amrami M, Poirier O, Nicaud V, Bousser M-G, Alpérovitvh A. Association between migraine and endothelin type A receptor (ETA-231 A/G) gene polymorphism. Neurology 2001; 56: 1273–7.
Venance L, Stella N, Glowinski J, Giaume C. Mechanism involved in initiation and propagation of receptor-induced intercellular calcium signalling in cultured rat astrocytes. J Neurosci 1997; 17: 1981–92.
Venance L, Premont J, Glowinski J, Giaume C. Gap junctional communication and pharmacological heterogeneity in astrocytes cultured from the rat striatum. J Physiol 1998; 510: 429–40.
Whitty CWM. Familial hemiplegic migraine. In: Vinken PJ, Bruyn GW, Klawans HL, Clifford Rose F, editors. Headache, handbook of clinical neurology, Vol. 48. Amsterdam: Elsevier; 1986. p. 141–53.
Willette RN, Sauermelch CF. Abluminal effects of endothelin in cerebral microvasculature assessed by laser Doppler flowmetry. Am J Physiol 1990; 259: H1688–93.
Yanagisawa M, Kurihara H, Kimura S, Tomobe Y, Kobayashi M, Mitsui Y, et al. A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature 1988; 332: 411–15.[Medline]
Zhang ET, Hansen AJ, Wieloch T, Lauritzen M. Influence of MK-801 on brain extracellular calcium and potassium activities in severe hypoglycemia. J Cereb Blood Flow Metab 1990; 10: 136–9.[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. M. Hansen, L. L. Thomsen, J. Olesen, and M. Ashina Calcitonin gene-related peptide does not cause the familial hemiplegic migraine phenotype Neurology, September 9, 2008; 71(11): 841 - 847. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Strong, P. J. Anderson, H. R. Watts, D. J. Virley, A. Lloyd, E. A. Irving, T. Nagafuji, M. Ninomiya, H. Nakamura, A. K. Dunn, et al. Peri-infarct depolarizations lead to loss of perfusion in ischaemic gyrencephalic cerebral cortex Brain, April 1, 2007; 130(4): 995 - 1008. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Hermans, M. C Post, V. Thijs, M. Spaepen, and W. I H L Budts Increased prevalence of migraine in adult congenital heart disease Heart, March 1, 2007; 93(3): 361 - 362. [Full Text] [PDF]
|
|
|
|
 |
|
 J. P. Dreier, J. Kleeberg, M. Alam, S. Major, M. Kohl-Bareis, G. C. Petzold, I. Victorov, U. Dirnagl, T. P. Obrenovitch, and J. Priller Endothelin-1-Induced Spreading Depression in Rats Is Associated with a Microarea of Selective Neuronal Necrosis Experimental Biology and Medicine, February 1, 2007; 232(2): 204 - 213. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Kann and R. Kovacs Mitochondria and neuronal activity Am J Physiol Cell Physiol, February 1, 2007; 292(2): C641 - C657. [Abstract] [Full Text] [PDF]
|
|