Brain Oral Session: Cortical Spreading Depression

331. Increased susceptibility to cortical spreading depression in cadasil mutant mice

K. Eikermann-Haerter¹, Y. Wang¹, E. Dilekoz¹, J.F. Arboleda-Velasquez², S. Artavanis-Tsakonas², A. Joutel³, M.A. Moskowitz¹ and C. Ayata^1,4

¹Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown; ²Cell Biology, Harvard Medical School, Boston, Massachusetts, USA; ³Faculte de Medecine, Universite Paris 7-Denis Diderot, Paris, France; ⁴Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA

Background and aims: Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukencephalopathy (CADASIL) first manifests with migraine before the occurrence of recurrent strokes and vascular dementia. CADASIL has been linked to missense mutations (e.g., R90C) in the NOTCH3 gene exclusively expressed in vascular smooth muscle cells, which degenerate during the disease progress. Cortical spreading depression (CSD) is a transient neuroglial depolarization, which slowly propagates centrifugally once evoked when extracellular K⁺ concentrations exceed a critical threshold. CSD is believed to underlie migraine aura. We have recently shown that human familial hemiplegic migraine mutations in Ca_V2.1 channel (i.e., neuronal mutation) enhance CSD susceptibility. Here, we tested whether CADASIL mutant mice (i.e., vascular mutation) overexpressing NOTCH3 R90C, and mice lacking the functional NOTCH3 protein (NOTCH3-KO), show altered CSD-susceptibility.

Methods: We determined CSD susceptibility by analyzing:

• (a)

  the electrical threshold for inducing CSD (single square pulses with increasing intensity on the occipital cortex), and

• (b)

  the frequency of repetitive CSDs induced by topical KCl (300 mmol/L for 1 h on the occipital cortex), in mechanically ventilated, physiologically monitored mice under isoflurane anesthesia.

CSDs were recorded from parietal and frontal cortex using glass micropipettes. Aquaporin4 (AQP4) expression was investigated by immunohistochemistry and Western blots in naïve brains. R90C transgenic mice were compared to non-transgenic mice as well as transgenic mice overexpressing WT human NOTCH3. NOTCH3-KO was compared to WT littermates.

Purpose: To test whether CADASIL mutations enhance CSD susceptibility.

Results: The electrical CSD threshold was significantly lower, and frequency of KCl-induced CSDs higher in R90C transgenic mice compared to WT. CSD speed was increased as well. NOTCH3-KO mice also displayed increased CSD frequency and propagation speed upon topical KCl, and this phenotype was even stronger than in R90C. No difference was detected in CSD susceptibility between male and female R90C mice. Senescence (22±5 mo) partially reduced CSD susceptibility when tested in NOTCH3-KO. Subcortical recordings did not show striatal SD in any group. Interestingly, the water channel protein AQP4, predominantly expressed on astrocytic endfeet, ependyma and brain endothelium, was increased by two-fold in R90C brains suggesting a link between CSD susceptibility and brain water regulation.

Conclusion: In summary, our data show that the archetypal R90C CADASIL mutation enhances CSD susceptibility, providing evidence for the first time that a vascular mutation can produce a migraine phenotype by increasing CSD susceptibility. This is the second gene to date associated with a human migraine syndrome that is shown to enhance CSD susceptibility.

Strain	Genotype	Gender	Threshold, μC	N	Frequency, #/h	N	Speed, mm/min
Notch3 R90C	WT	M	485 [98–65]	12	8±1	5	3.0±0.3
Notch3 R90C	WT	F	386 [161–114]	13	9±2	9	3.2±0.3
Notch3 R90C	TG	M	55 [2–55]*	13	12±1*	6	3.6±0.3*
Notch3 R90C	TG	F	25 [4–55]*	13	13±1*	7	4.3±0.5*
Notch3 KO	WT	M	—	—	9±2	8	—
Notch3 KO	KO	M	—	—	15±3*	8	—
Aged Notch3 KO	WT	M	—	—	9±2	8	3.0±0.3
Aged Notch3 KO	KO	M	—	—	13±1*	8	3.7±0.2*

Electrical threshold: median (interquartile range), others: mean±s.d.,

*

P<0.05.

1014. Acceleration of ischemic tissue damage by moderately decreased blood pressure: role of tissue hypoxia but not of cortical spreading depression

K. Waehner, N. Weinzierl and L. Schilling

Division of Neurosurgical Research, Medical Faculty Manheim, University of Heidelberg, Mannheim, Germany

Background: The pathophysiology of focal brain ischemia is complex and still poorly understood. The size of brain damage may be influenced by local and systemic changes of a variety of parameters such as arterial blood pressure (BP), occurrence of cortical spreading depression (CSD) or CSD-like episodes, and changes of blood sugar content occurring early, i.e. within hours after vessel occlusion. We addressed the question whether BP at a level around the lower range of autoregulation, i.e. around 80 mm Hg will affect development of ischemic damage by increasing the rate of CSD episodes.

Methods: Male Sprague-Dawley rats were intubated in isoflurane anesthesia the and artificially ventilated to maintain blood gases within physiological ranges. The femoral artery and vein were catheterized for BP monitoring and injection of pimonidazole (6 mL/100 g body weight), respectively. The right common carotid artery was exposed and a 4–0 nylon filament with the tip covered with a layer of silicone introduced. A burr hole was created in the right dorsal skull bone in the middle cerebral artery (MCA) territory to position:

• (i)

  calomel electrode for monitoring DC potential changes in the epidural space and

• (ii)

  a laser Doppler flowmeter (LDF) probe to measure changes of regional cerebral blood flow (rCBF).

Additional burr holes to measure rCBF changes were created in the right anterior cerebral artery and left MCA territory. Two experimental groups were studied: a moderate BP level group in which the animals were kept at a blood pressure above 95 mm Hg and a the second one in which BP was maintained around 80 mm Hg for at least two hours by applying negative pressure to the abdomen and the hindlimbs (low level BP group). At the end of the 4 h observation period the brain was removed and serial coronal sections were cut. Adjacent sections were taken every 500 μm and used for either silver nitrate staining to detect brain damage or pimonidazole adducts staining to detect tissue hypoxia. The sections were scanned, areas where measured and processed to yield the volumes of structural damage and of hypoxia.

Results: Decreasing BP after establishing focal ischemia did not result in fall of LDF signal in the left MCA territory. The numbers of negative DC potential shifts occurring during 4 h after MCAO ranged from 1 to 10 in the moderate and 1 to 8 in the low level BP group. Volume of ischemic damage was 92.3±25.8 cm³ (mean±s.e.m.; n = 11) in the moderate level PB group and 157.1±24.6 cm³ (n = 5) in the low level BP group (P = 0.1). Tissue hypoxia amounted to 168.8+20.6 cm³ (moderate level BP group) and 195+31.6 cm³ (low level BP group).

Conclusion: The results indicate that BP levels in the lower range of autoregulation occurring early after induction of focal cerebral ischemia accelerates the development of ischemic damage, probably by increasing the hypoxic burden to the tissue. Facilitation of CSD episodes, however, does not appear to play a major role.

760. Combined laser speckle and multispectral reflectance imaging of cortical spreading depression in wild type and familial hemiplegic migraine 1 mice

I. Yuzawa¹, K. Eikermann-Haerter¹, S. Sakadzic², D.A. Boas², M.A. Moskowitz¹ and C. Ayata^1,3

¹Stroke and Neurovascular Regulation Laboratory, Neuroscience Center, Massachusetts General Hospital & Harvard Medical School; ²MGH/MIT/HMS Athinuola A. Martinos Center for Biomedical Imaging, Department of Radiology; ³Stroke Service and Neuroscience Intensive Care Unit, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Charlestown, Massachusetts, USA

Objectives: Unlike in most other species, cortical spreading depression (CSD) is associated with a triphasic cerebral blood flow (CBF) response in mice, with a profound initial hypoperfusion followed by transient normalization of flow and a long-lasting post-CSD oligemia. We aimed to investigate the hemodynamic and metabolic impact of CSD non-invasively with high spatiotemporal resolution in mice using multimodal optical imaging. Using the model, we tested whether familial hemiplegic migraine type 1 (FHM1) R192Q mutation in Ca_v2.1 (P/Q-type) voltage-gated calcium channels augment the hemodynamic and metabolic impact of CSD.

Methods: Changes in CBF, oxyhemoglobin (oxyHb), deoxyhemoglobin (deoxyHb), total hemoglobin (as a surrogate measure for cerebral blood volume, CBV), mixed arteriovenous O₂ saturation (S_AVO₂), and cerebral metabolic rate of oxygen (CMRO₂) were simultaneously measured through intact skull using combined laser speckle flowmetry and multispectral reflectance imaging during CSD in isoflurane-anesthetized (70%N₂O/30%O₂) male wild-type (C57/BL6) and female Cacna1a R192Q knockin mice (n = 5 each). The region of interest was within middle cerebral artery territory away from large surface vessels. Two consecutive CSDs were induced in frontal cortex 15 mins apart by topical KCl (1 mol/L).

Results: The first CSD in cortex was associated with a profound hypoperfusion reaching ∼30% of baseline. CBV was also reduced by 30% confirming vasoconstriction as a mechanism. OxyHb and deoxyHb showed reciprocal changes simultaneously with and dominated by the CBF and CBV changes. Consequently, S_AVO₂ decreased by ∼50%. Although CMRO₂ showed a small early rise at the onset of CSD (<10%), it rapidly dropped to ∼50% of baseline, presumably due to O₂ supply limitation. After a characteristic transient normalization 1 to 2 mins after CSD onset, post-CSD oligemia ensued (CBF ∼30%, CBV ∼70%, CMRO₂ ∼50%, S_AVO₂ ∼60% of pre-CSD baseline). The hemodynamic and metabolic response to the second CSD was dramatically different: superimposed on the post-CSD oligemia, all measured parameters showed a monophasic increase lasting 3 to 4 mins. No significant difference was observed between wild type and R192Q knockin mice in any measured parameter (Figure).

Figure

Data are average of all mice shown as % of baseline (except S_AVO₂ expressed as % Hb saturation, resting state empirically taken as 65%). Horizontal axis shows time in min.

Conclusions: Our data demonstrate the unique vasoconstrictive vascular response to SD in mouse cortex, and show that O₂ metabolism becomes supply-limited. Post-CSD oligemia is associated with tissue hypoxia as evidenced by hemoglobin desaturation. Caution must be exercised to monitor for CSD occurrence during experimental preparation, and to distinguish the first CSD occurring in naïve mouse cortex and subsequent ones, as the vascular response to subsequent CSDs, and the baseline that they are superimposed upon, are very different.

256. Clusters of cortical spreading depolarizations after subarachoid hemorrhage may advance delayed cortical ischemia via reduced O₂-supply and increased O₂-consumption

B. Bosche^1,2, R. Graf¹, R.-I. Ernestus², C. Dohmen³, T. Reithmeier², G. Brinker², A.J. Strong⁴, J.P. Dreier⁵ and J. Woitzik^6,7

¹Max Planck Institut for Neurological Research with Klaus-Joachim-Zuelch-Laboratories of the Max Planck Society and the Faculty of Medicine of the University of Cologne; ²Neurosurgery; ³Neurology, University of Cologne, Cologne, Germany; ⁴Department of Clinical Neuroscience, Institute of Psychiatry, King's College London, London, UK; ⁵Departments of Neurology and Experimental Neurology, Charité Campus Mitte, University Medicine Berlin, Berlin; ⁶Department of Neurosurgery, University Medicine Mannheim, Mannheim; ⁷Department of Neurosurgery, Charité Campus Benjamin Franklin, University Medicine Berlin, Berlin, Germany

Objectives: The pathophysiology of delayed cortical ischemia (DCI) after subarachnoid hemorrhage (SAH) is poorly understood. It has recently been questioned whether vasospasm of large proximal arteries is the only cause for this life-threatening complication after SAH,^1,2 and DCI has been associated with clusters of cortical spreading depolarization (CSD). ³ In animal models, CSD is a well studied self-propagating wave of neuronal and astrocytic depolarization that can be associated with a biphasic response of cortical blood flow^4,5 in penumbra and/or hypoxic conditions. In this study, we examined how clusters of CSD may promote DCI in human cerebral cortex after SAH.

Methods: Nine patients were prospectively recruited from two different neurosurgery centers (Cologne and Mannheim/Heidelberg, Germany). We used subdural electrocorticographic (ECoG) ³ and tissue oxygen pressure (p_tiO₂) spatiotemporal co-recordings. P_tiO₂ is predominantly correlated to local cerebral blood flow but also affected by tissue metabolism. We differentiate between ‘single’ CSD (20%) and repetitive waves of CSD occurring within clusters (80%). We analyzed alterations of CSD-associated p_tiO₂ curve patterns (biphasic alteration, monophasic decrease or increase), the key features (baseline, minimum and maximum) as well as the specific phase durations and integrals.

Results: In a total recording time of 850 h, 120 waves of CSD were found in eight of nine patients (∼89%). CSD emerged predominantly between the fifth and seventh days after SAH. 53.2% of CSDs were associated with clear p_tiO₂ alterations. We detected CSD within clusters occurring simultaneously with mainly biphasic p_tiO₂ responses comprising a primary hypoxic and secondary hyperoxic phase. Monophasic p_tiO₂ decreases were minor in number, however ∼70% of them were detected in patients with MRI- or CT-proven DCI. The absolute number and the number/day of CSDs were positively correlated with the duration of the hypoxic phase of biphasic p_tiO₂ responses (P = 0.005; P = 0.003, respectively; Spearman's correlation) and the absolute number of CSDs was negatively correlated with the duration of the secondary hyperoxic phase (P = 0.008). Eight of nine patients showed clusters of CSD. In five of them, four or more repetitive CSD with p_tiO₂ responses were detected and characterized by their specific order as 1st, 2nd, 3rd, and 4th or higher rank within clusters. Analysis revealed that the primary hypoxic phase of the p_tiO₂ responses significantly increased over time within clusters (P = 0.011, Friedman-test). We attribute these findings mainly to changes in local cerebral blood flow in the cortical microcirculation but also to augmented metabolism, both associated to CSD.

Conclusions: Our results indicate that clusters of CSD are involved in the reduction of O₂ supply and the increase of O₂ consumption, thereby promoting DCI, which may reveal a novel pathophysiological mechanism with clinical relevance after SAH.

650. Local dynamic metabolic response to spontaneous spreading depolarisations in the penumbra of the acutely injured human brain

D. Feuerstein¹, A. Manning², P. Hashemi³, M. Fabricius⁴, A.J. Strong² and M.G. Boutelle³

¹Bioengineering, Imperial College London; ²King's College Hospital; ³Imperial College London, London, UK; ⁴Glostrup University Hospital, Copenhagen, Denmark

Objectives: It has become clear that spreading depolarisations (SD) occur spontaneously in patients with acute traumatic or ischaemic brain injury. ¹ They are waves of transient electrical depolarisations that propagate from the core of the injury to the surrounding compromised tissue. Although there is extensive evidence that SDs severely disrupt local cerebral blood flow ² and metabolism ³ in animal models of stroke, little is known about the metabolic response to SD in the human brain.

Here we have used online rapid sampling microdialysis (rsMD) in human penumbral tissue to quantify the dynamic metabolic response to SDs detected by electrocorticography (ECoG).

Methods: In 10 patients selected for craniotomy, a clinical microdialysis (MD) probe, along with a 6-electrode subdural strip for the ECoG signal, was inserted under direct vision into the immediate vicinity of the injured cortical tissue. It was perfused with artificial cerebrospinal fluid at 2 μL/min. The dialysate was assayed electrochemically using rsMD. ⁴ Glucose and lactate levels were thereby continuously measured at 1-minute intervals for up to five days following craniotomy. The rsMD data were analyzed and subsequently correlated to the SD events separately identified according to their ECoG signature. ⁵

Results: Metabolic changes were resolved in 90 identified SD events. The dialysis changes typically consisted of a rise in lactate peaking at +173.5±29.6 μmol/L and a drop in glucose down to −120.79±13.2 μmol/L relative to baseline values. The dialysis levels remained different from baseline values at 20 mins after the SD event being 68.7±25.8 μmol/L above baseline for lactate and 69.5±11.1 μmol/L below baseline for glucose. In all patients the glucose concentrations did not recover their initial levels before the onset of the following s.d., thus resulting in a progressive drop in extracellular glucose concentrations as measured by microdialysis.

Conclusions: These results agree well with experimental findings using rsMD. ⁶ The progressive fall in dialysate glucose suggests that the frequent occurrence of SD events leads to a failure of glucose supply to meet the energetic demand for the repolarisation of the cells. This could compromise the viability of the tissue and hence lead to further expansion of the lesion.

753. Decreased CA²⁺ spark and associated BK channel activity in parenchymal arteriolar myocytes following subarachnoid hemorrhage

G. Wellman, M. Koide and K. O'Connor

Pharmacology, University of Vermont, Burlington, Vermont, USA

Objectives: Aneurysmal subarachnoid hemorrhage (SAH) is associated with high rates of morbidity and mortality. It has been a long-held notion that delayed blood-induced vasospasm of large diameter conduit arteries on the brain's surface is a major cause of this death and disability. However, compelling data are emerging to suggest additional phenomena, including enhanced constriction of the microcirculation, may also contribute to the development of neurological deficits in SAH patients. The objective of the present study was to determine whether impairment of a key vasodilator pathway, Ca²⁺ sparks and associated activity of large-conductance Ca²⁺ activated K⁺ (BK) channels, contribute to enhanced constriction of parenchymal arterioles following SAH.

Methods: Using a rabbit SAH model, parenchymal arterioles were isolated from the temporal lobe of the cerebral cortex and cannulated for in vitro diameter measurements. Individual myocytes were also enzymatically isolated from parenchymal arterioles for Ca²⁺ spark, single channel and whole-cell BK current measurements. To evaluate Ca²⁺ sparks, fluorescence intensity changes were monitored using laser scanning confocal microscopy in freshly isolated parenchymal arteriolar myocytes loaded with the Ca²⁺ indicator, fluo-4. Expression of BK channel alpha and beta-1 subunits and ryanodine receptor-2 (RyR2) was examined by RT-PCR.

Results: Ca²⁺ spark frequency was decreased by greater than 50% in parenchymal arteriolar myocytes following SAH (0.31±0.07 Hz, n = 16) compared to healthy controls (0.78±0.09 Hz, n = 11). Although Ca²⁺ spark frequency was dramatically reduced, mean Ca²⁺ spark amplitude, expressed as a fractional change in fluorescence (F/F₀), was similar in myocytes isolated from control and SAH animals. Additional spatio-temporal characteristics such as rise-time, duration, size, and decay were similar for Ca²⁺ sparks recorded from myocytes of control and SAH animals. This decrease in Ca²⁺ spark frequency corresponded to decreased RyR2 expression in myocytes from SAH animals. Using perforated patch whole-cell electrophysiology, transient BK currents were observed in parenchymal myocytes isolated from both control and SAH rabbits. As with Ca²⁺ sparks, the frequency of transient BK currents was decreased by approximately 60% following SAH, with no change in current amplitude. Single channel recordings obtained from excised ‘inside-out’ membrane patches demonstrated no change in the Ca²⁺ or voltage sensitivity of BK channels following SAH. Further, the density of functional BK channels detected in excised membrane patches was similar between groups, and BK channel alpha and beta-1 subunit expression was unaltered following SAH. Consistent with decreased Ca²⁺ spark/BK channel activity, parenchymal arterioles isolated from SAH rabbits exhibited enhanced constriction compared to arterioles isolated from healthy control animals.

Conclusions: In summary, we report the frequency of Ca²⁺ sparks and associated transient BK currents are significantly decreased in smooth muscle of parenchymal arterioles following SAH, contributing to enhanced vasoconstriction. This study may help to explain the impact of SAH leading to enhanced constriction of arterioles within the brain parenchyma following cerebral aneurysm rupture.