Abstract
Berger M, Speckmann EJ, Pape HC, Gorji A. Spreading depression enhances human neocortical excitability in vitro. Cephalalgia 2008; 28:558–62.
To explore the effects of cortical spreading depression (CSD), an intrinsic brain event underlying migraine aura, on the excitability of the human neocortex, the authors conducted in vitro electrophysiological studies on slices from human temporal lobe that were obtained during epilepsy surgery. Neocortical slices were prepared from 13 different patients with intractable temporal lobe epilepsy. CSD was induced by KCl application to the sixth layer of neocortical tissue, and evoked extracellular field potentials and long-term potentiation (LTP) were induced by single or tetanic stimulation of white matter substance. Local application of KCl consistently induced spreading depressions (SDs). Baseline field excitatory postsynaptic potentials (EPSPs) temporarily disappeared during CSD for 3–5 min and recovered completely approximately 20 min later. Further recordings revealed that CSD significantly enhanced the amplitude of EPSPs for the next 90 min and increased the induction of LTP in the third layer of neocortical tissues. The authors concluded that CSD facilitates synaptic excitability and efficacy in human neocortical tissues, which can be assumed to contribute to hyperexcitability of the cerebral cortex in migraineurs.
COMMENTARY
The results of this study are valuable not only for being conducted in human neocortical structures, but also because they demonstrate that the relationship between hyperexcitability and CSD are mutual. CSD indicates an extreme excitability state of the grey matter with massive redistribution of ions that emerges in reaction to noxious stimulus. CSD occurs in different clinical conditions such as cerebral ischaemia, head trauma or subarachnoid haemorrhage and it is the pathophysiological event underlying migraine aura with the capability of triggering the trigeminovascular system (1). Recent studies have clearly indicated that familial hemiplegic migraine mutations (CACNA1A, ATP1A2, SCN1A) all influence the glutamatergic synapses in the cerebral cortex and lower the threshold for CSD (2). So the altered balance towards increased excitability is needed for generation of CSD, confirming extensive experience with pharmacological triggers such as potassium, glutamate, N-methyl-D-aspartate (NMDA) or the Na+/K+-ATPase inhibitor ouabain (3). During CSD, the spontaneous activity ceases for a few minutes; nonetheless, synaptic transmission is much more vulnerable and impairment in evoked responses could last up to half an hour. Those features, compatible with the clinical symptoms of migraine aura, have already been shown in different species.
The demonstration of hyperexcitability and sustained increase in the efficacy of synaptic transmission in the affected neocortex as a long-term complication of CSD in human tissues is remarkable. Whether CSD induces irreversible hyperexcitability in the symptomatic cortical area in migraineurs as concluded by the authors is debated. It is important to emphasize that long-term repetitive seizures that are intractable to medication in the patients enrolled could alter the excitability of the cortical tissue in a way that affects the results, although epilepsy is one of the significant comorbid disorders with migraine. Among various mechanisms leading to excitability change, selective suppression of GABAergic function by repetitive CSDs (4) or selective facilitation of glutamatergic synaptic efficacy in Ca(v)2.1 knock-in migraine mice (5) strengthen the aforementioned notion. It seems likely that occurrence of CSD could produce the state of vulnerability to subsequent auras in humans, considering particularly certain conditions such as frequent or persistent auras.
The study underscores the importance of prophylactic medication to confine the excitability change and could have implications for long-term prophylaxis in migraine patients with aura (MA). Based on the author's data, even a parenteral antiepileptic medication such as intravenous valproate or magnesium would be considered advisable for a prolonged or single aura attack. Could similar therapeutic deductions be made for migraine without aura (MoA) patients? Available scientific proofs are unsatisfactory to support the controversial hypothesis of ‘silent aura’ in MoA (6). In fact, the study demonstrates how hard is to observe a silent CSD by showing that CSD per se would cause significant symptoms due to absence of neuronal activity (3–5 min) and deficient evoked activity (21 min) to drive postsynaptic neurons in the affected neocortex, unless SD occurs in the cerebral cortex, where such a loss of function remains unnoticeable! The existing evidence only shows the CSD-attenuating capability of several effective antimigraine drugs in rodent brain (7). However, CSD-inhibiting properties are not confined to antimigraine drugs and the pharmaceutical agents that failed to demonstrate antimigraine efficacy such as clonidine could block CSD (8). In that sense, a recent clinical study with tonabersat is noteworthy (9). Tonabersat is a novel antiepileptic drug with well-known CSD-suppressive properties in experimental models and its effectiveness as a preventive migraine drug was tested for the first time. The results of the study did not show any efficacy over a placebo group, since the reduction of headache-free days, the primary end-point of the study, was not significantly different between tonabersat and placebo, though the efficacy was not investigated between MA and MoA groups separately. Accordingly, it is rational to consider that the CSD-blocking property could not be the primary mechanism of effective antimigraine preventive medication, but an epiphenomenon.
H Bolay
Gazi University, Department of Neurology & Neuropsychiatry Centre, Ankara, Turkey
References
Richter F, Bauer R, Lehmenkühler A, Schaible HG. Spreading depression in the brainstem of the adult rat: electrophysiological parameters and influences on regional brainstem blood flow. J Cereb Blood Flow Metab 2008; 28:984–94.
CSD is a well-described pathophysiological brain event in the cerebral cortex and characterized by initial massive depolarization of neurons and glial cells followed by prolonged inhibition of electrical and metabolic activity that propagates slowly over the cerebral cortex, affecting the whole hemisphere in the case of lissencephalic brain. The brainstem is one of the structures that are resistant to generation and propagation of SD waves. The authors were able to demonstrate for the first time that with the proper conditioning stimulus-reproducible CSD waves were generated in the brainstem of an adult rat. To alter the excitability of brainstem neurons and glial cells, the authors superfused brainstem with conditioning solutions in which chloride ions were replaced by acetate and tetraethylammonium chloride (to block potassium channels) and a small amount of KCl was added. SD waves elicited were characterized by DC shifts propagating in the brainstem that were blocked by MK-801, an NMDA receptor antagonist. SD waves were associated with transient increase in regional blood flow and systemic arterial blood pressure. However, KCl stimulation neither elicited SD nor induced changes in regional blood flow or blood pressure in the non-conditioned brainstem. The authors concluded that proper conditioning renders the brainstem susceptible to SD.
COMMENTARY
CSD occurs during pathophysiological brain conditions such as migraine aura, ischaemic brain infarction, epilepsy, traumatic brain injury, transient global amnesia and others (1). According to the traditional view, CSD occurs only in the cerebral cortex, where a critical density of neurons and glia is reached such as in the occipital lobe and can not propagate to white matter. Its occurrence is also affected by maturation of the central nervous system (1). Nevertheless, the authors demonstrated for the first time that with the accurate priming stimuli the brainstem of the rat, which is resistant to SD waves, can be easily transformed to a structure vulnerable to SD. While the adult brainstem is resistant to KCl-triggered DC shifts in the native state, altering the extracellular ionic concentrations could overwhelm the resistance and reproducible SDs can be generated, based on the computational model that SD is ignited when persistent inward currents exceed outward currents. It is notable that such a switch to a vulnerable state was reversible and can be accomplished in the adult rat without any genetic interference. How extrinsic or intrinsic triggering factors modulate the vulnerability and interact with ionic gradient forces across the cell membrane to surmount histomorphological resistance to SD needs to be addressed.
Does SD in the brainstem (an inhomogeneous structure with large extracellular space between ascending and descending white matter tracts and neuroglial cells) represent another manifestation of SD with characteristics different from known properties of the cortical one?
The authors demonstrated that the following properties of SD waves generated in the brainstem are not akin to cortical ones. The propagation of SD waves was limited by 1–2 mm, suggesting that its effects are local, whereas the propagation of SD could traverse the cerebral cortex of whole hemisphere in rodent brain (lissencephalic brain). Invasion of the entire hemisphere by CSD probably constitutes adequate stimulus to activate perivascular trigeminal nerve endings in all meningeal layers (pia, arachnoid and dura mater) overlying cerebral cortex (2). It is clear that the more trigeminovascular nerve endings triggered by CSD the greater the likelihood of lateralized brainstem trigeminal nucleus caudalis activation. Extracellular potassium levels in brainstem SD were not as high as that of CSD. Finally, SD-triggered transient regional cerebral blood flow increase (approximately 30%) was not as remarkable as that in the cerebral cortex (100% and above) (1, 2) and spreading oligaemia was not noted in the brainstem SD in rat.
What is the relevance of brainstem SD in humans? It could plausibly be the underlying mechanisms of basilar type migraine, a rare type of migraine. However, different properties of brainstem SD as showed by the authors could suggest that lateralized trigeminovascular activation as induced by cortical SD in the rat brain (3) seems to be an unlikely consequence following brainstem SD. Instead, transient loss of function and altered excitability would be expected not only within the affected structure but also in the connected distant areas. In that sense, it is likely that SD could manifest in other subcortical structures such as basal ganglia, thalamus or periaqueductal grey and could play a role in pain transmission or chronification. Whether SD is related to posterior circulation infarctions in the brainstem watershed areas (3) or iron accumulation in the subcortical structures (4, 5) remains to be investigated.
H Bolay
Gazi University, Department of Neurology & Neuropsychiatry Centre, Ankara, Turkey
References
Hashemi P, Bhatia R, Nakamura H, Dreier JP, Graf R, Strong AJ, Boutelle MG. Persisting depletion of brain glucose following cortical spreading depression, despite apparent hyperaemia: evidence for risk of an adverse effect of Leão's spreading depression. J Cereb Blood Flow Metab 2009; 29:166–75.
CSD induces some metabolic compromise that varies according to the metabolic state and the vascular supply of the brain as well as the species studied. The authors investigated CSD-triggered glucose and lactate alterations in cat cerebral cortex by using rapid sampling microdialysis method that allows identification of a combined time-series signature for glucose and lactate. CSD was induced by KCl injection through a micropipette into the cortex and verified by the occurrence of a characteristic transient DC shift. Laser speckle imaging was used to observe propagation of blood flow changes near the microdialysis probe. Ten minutes after depolarization, dialysate glucose fell and lactate rose by 28 and 58%, respectively. There was no recovery of dialysate glucose 30 min after depolarization while regional blood flow was still high (hyperaemia also persisted > 30 min). The authors indicated a high risk of depletion of extracellular glucose in association with CSD that was previously thought to be largely a benign brain event and emphasized the potential risk of repetitive and clustered depolarization waves in conditions where vascular and metabolic impairment is seen such as acute brain injury or stroke.
COMMENTARY
The study underlines another potential risk of CSD that is generally accepted to be a benign brain event. Neurovascular uncoupling is defined as when neuronal activity-induced metabolic demand such as oxygen or glucose supply is unmet. Impaired neurovascular coupling associated with CSD has been reported previously by other authors. Takano and colleagues showed in mice cerebral cortex that CSD induces significant redox state in the cerebral tissue that was improved by increased O2 supply (1). The transient tissue hypoxia was due to the increased O2 demand exceeding the vascular O2 supply and occurred when the blood flow increase was observed during CSD. In the presented paper, the authors conducted experiments in a gyrencephalic cat brain and therefore the results are more pertinent to human.
Familial hemiplegic migraine (FHM) phenotypes affect glutamatergic synapse and lower the threshold for CSD in the cerebral cortex. FHM2 mutation affecting astrocytic Na+/K+-ATPase uncovers a potential pathway by which neuronal demand for glucose utility could be affected in migraine patients (2). Glutamate released from presynaptic neurons is taken by perisynaptic astrocytes along with a sodium ion that induces Na+/K+-ATPase by increasing intra-astrocytic Na+. In addition, Na+/K+ pump maintains low extracellular K+ by exchanging K+ with intracellular Na+. Since Na+/K+-ATPase also activates glycolysis and glucose uptake from vascular bed, it couples synaptic glutamate to glucose utilization (3). Therefore, in FHM2 mutation where Na+/K+-ATPase is dysfunctional, glucose supply would not be sufficient for neuronal demand and K+ ions in the perisynaptic area would be increased (3). The results of the presented paper suggest that CSD per se somehow induces insufficient glucose supply for as long as 30 min in gyrencephalic brain. Whether that impairment is associated with any dysfunction of the pump or glycolytic pathway or glucose transporters needs to be addressed in further studies.
The emerging of CSD probably induces a vicious circle where long-term enhanced synaptic efficacy and increased excitability renders the cortical tissue for the next CSD, particularly in the setting of metabolic compromise such as hypoxia and decreased glucose supply. Taking account of those mechanisms in the tissue exposed to CSD, prophylaxis rather than remedy medication for headache would be more beneficial for migraine with aura patients.
H Bolay
Gazi University, Department of Neurology & Neuropsychiatry Centre, Ankara, Turkey