- Email: [email protected]
Effects of experimental traumatic brain injury and impaired glutamate transport on cortical spreading depression
Accepted Manuscript Effects of experimental traumatic brain injury and impaired glutamate transport on cortical spreading depression
Mahshid Sadat Hosseini-Zare, Feng Gu, Ahmad Abdulla, Simon Powell, Jokūbas Žiburkus PII: DOI: Reference:
S0014-4886(17)30119-X doi: 10.1016/j.expneurol.2017.05.002 YEXNR 12530
To appear in:
Experimental Neurology
Received date: Revised date: Accepted date:
18 July 2016 14 April 2017 3 May 2017
Please cite this article as: Mahshid Sadat Hosseini-Zare, Feng Gu, Ahmad Abdulla, Simon Powell, Jokūbas Žiburkus , Effects of experimental traumatic brain injury and impaired glutamate transport on cortical spreading depression, Experimental Neurology (2017), doi: 10.1016/j.expneurol.2017.05.002
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Effects of experimental traumatic brain injury and impaired glutamate transport on cortical spreading depression.
Mahshid Sadat Hosseini-Zare1,2, Feng Gu1Ɣ, Ahmad Abdulla1,2, Simon Powell1, and Jokūbas Žiburkus*1,2 University of Houston, Department of Biology and Biochemistry; 2Texas Institute for
T
1
IP
Measurement and Evaluation Statistics (TIMES), Houston, Texas 77024, Houston, Texas 77204,
US
CR
U.S.A.
AN
* Corresponding author: Jokūbas Žiburkus
M
Associate Professor
University of Houston, Department of Biology and Biochemistry and Texas Institute for
Ɣ
PT
Email: [email protected]
ED
Measurement, Evaluation, and Statistics (TIMES) Houston, TX 77204, U.S.A.
– Current address: Feng Gu, University of Stanford Medical School, Department of
AC
CE
Neurology and Neurobiology Sciences, Stanford, CA 94305, U.S.A.
Acknowledgment: This research was sponsored by the Dravet Syndrome Foundation and University of Houston’s Grants to Advance and Enhance Research (GEAR) funds.
ACCEPTED MANUSCRIPT ABSTRACT
Cortical spreading depression (CSD) is associated with traumatic brain injury (TBI), stroke, migraines, and seizures. Typically, following TBIs and other insults, neuronal excitability in and around the area of the injury is affected, with reported increases in local glutamate signaling. Astrocytic glutamate transporters are critical for precise regulation of the extracellular glutamate
T
availability. However, it remains unclear how impaired astrocytic glutamate transport or an acute
IP
TBI affect characteristics of the CSD. We quantified the properties of CSD using whole-cell and
CR
extracellular electrophysiological recordings, and voltage-sensitive dye imaging (VSDI) in rat visual cortex in vitro. To model impaired astrocytic glutamate transport, we used astrocytic
US
glutamate transporter blocker (2S, 3S)-3-[3-[4-(trifluoromethyl) benzoylamino] benzyloxy] aspartate (TFB-TBOA). In addition, an acute incision through the superficial cortical layers was
AN
used to model the effects of acute traumatic brain injury (TBI) on CSD characteristics. Both manipulations; impaired glutamate cycling and acute cut profoundly affected the physiological
M
properties of cell firing, latency to CSD formation, and its frequency of occurrence. VSD imaging analysis revealed significant changes in spatiotemporal dynamics and propagation of the CSD,
ED
suggesting that that the cut itself may not initiate CSD depolarizing waves, but rather attract them. Blockade of GLT-1 caused significant reduction in whole-cell sodium currents and changes in
PT
CSD wave spatiotemporal characteristics as well, slowing it or even ‘trapping’ its propagation. Our results reveal new information about CSD properties in these pathological conditions and
AC
CE
demonstrate an important role of GLT-1 in regulation of CSD.
Key words: Cortical spreading depression, traumatic brain injury, astrocytic glutamate transporter
ACCEPTED MANUSCRIPT INTRODUCTION CSD is associated with several neurological disorders including stroke, migraine aura (Fabricius et al., 2006; Hossmann, 1996), epilepsy (Gorji. 2001), cerebral ischemia (Obrenovitch, 1995), hypoxia (Somjen, 2001), and TBI (Rogatsky, 1996; Strong et al., 2002). CSD can be observed in damaged brain regions and during the course of epileptic seizures (Lauritzen et al., 2011), but it can also be induced experimentally by raising glutamate or K+, manipulating Na+/K+
T
pump inhibitors, by using electrical stimulation or hyperthermia (Malkov et al., 2014).
IP
Electrically, CSD is characterized by profound depolarization of neuronal networks, followed by
CR
slow propagation (2-5 mm/min) of this depolarizing wave (Lauritzen et al., 2011; Larrosa et al., 2006). During the initial depolarizing portion of CSD, the extracellular glutamate and aspartate
US
concentrations increase (Van and Fifkova, 1970; Fabricius et al., 1993). Activation of N-methylD-aspartate (NMDA)-type glutamate receptors (NMDA-Rs) and increase in glutamate release
AN
further contribute to CSD progression (Obrenovitch and Zilkha, 1996; Zhou et al., 2013). Although mechanisms of CSD initiation and propagation are controversial, studies show that
M
both neurons and astrocytes are involved in this process (Eiselt et al., 2004). Extracellular glutamate levels are tightly regulated by glutamate transporter proteins located primarily on
ED
astrocytes (Rothstein et al., 1996). GLT-1 and GLAST are the two main glial glutamate transporters chiefly responsible for maintaining extracellular levels of glutamate. Despite the
PT
importance of glutamate clearance through the astrocytic transporters, to date it remains unknown how acute impairments in glutamate transport affect CSD.
CE
Furthermore, CSD is observed following TBI and seizures associated with it. Severe injuries that cause physical damage to the cortical tissue could become initiation zones for seizures and
AC
CSD (Fabricius et al., 2008). Following TBI, increased glutamate levels could also trigger seizures and/or CSDs (Goodrich et al., 2013). However, it remains unknown how an acute cut in the cortex can affect initiation site of CSDs in the cortex and its propagation in relation to the site of the injury. To address the role of glutamate and acute injury in CSD incidence and propagation, we employed electrophysiological, pharmacological, and high speed VSD imaging techniques. We quantified changes in CSD characteristics in two conditions: 1) in the presence of astrocytic glutamate transporter GLT-1 blocker TFB-TBOA, and 2) following acute cut through the superficial neocortical layers in vitro. Our findings suggest acute injury or TFB-TBOA
ACCEPTED MANUSCRIPT application significantly affected the initiation, duration, and spatiotemporal electrical properties of the CSD; however, the degree to which these effects could be attributed to dysregulation of astrocytic glutamate transport alone was confounded by the observation that TFB-TBOA reduced sodium conductance in cortical pyramidal cells. Thus, acute injury or concurrent blockade of astrocytic glutamate and neuronal sodium currents were responsible for the significant modulation
T
of CSD characteristics.
CR
IP
MATERIAL AND METHODS
Animals
US
Experiments were approved by University of Houston Institutional Animal Care and Use Committee. Studies were performed on juvenile, 15-24 days old, male wild-type Sprague Dawley
AN
rats. Potassium-induced spreading depression is well documented to form in the second postnatal week and is well developed by P15 (Bures, 1957; de Luca and Bures, 1977; Chuquet et al., 2007).
M
We used 5 animals for Na+ current recording, 5 animals for the TBI group, 7 animals for the TFBTBOA group and for each group of imaging, and 9 animals as controls. All numbers (n) refer to
ED
number of slices used in the experiments.
PT
Solutions and drugs
For tissue dissection and slice preparation, oxygenated (95% O2, 5% CO2) high sucrose
CE
dissection buffer (in mM): 248.3 sucrose, 2.5 KCl, 1.25 NaH2PO4, 3 MgSO4, 10 MgCl2, 0.5 CaCl2, 26 NaHCO3, and 10 dextrose) was used. Slices were pre-incubated in aerated (95% O2-
AC
5% CO2) standard artificial cerebrospinal fluid (ACSF) containing (in mM) 130 NaCl, 1.2 MgSO4, 3.5 KCl, 1.2 CaCl2, 10 glucose, 2.5 NaH2PO4, and 24 NaHCO3 (pH 7.3). The bath solution for voltage-clamp whole-cell sodium current recordings used ACSF containing (in mM): 95 NaCl, 5.6 KCl, 0.1 CaCl2, 5 MgCl2, 11 glucose, 1 NaH2PO4, 25 NaHCO3 and 20 TEA-Cl (pH 7.4 when bubbled with 95% O2–5% CO2) (Zhang et al., 2014). For induction of CSD, high K+ solution was made by an equimolar replacement of NaCl with 26 mM KCl. 2% Dimethyl sulfoxide (DMSO) and 98% deionized water solution was used to dissolve TFB-TBOA powder (Tocris Bioscience). TFB-TBOA was used with the final concentration of 50 nM: the IC50 value for GLT-1 and GLAST are 17 nM and 22 nM , respectively (Shimamoto et al., 2004).
ACCEPTED MANUSCRIPT
Visual cortex slices preparation Standard in vitro electrophysiology slice preparation techniques were used. In brief, the rats were deeply anesthetized with ether and decapitated using a guillotine. The brain was rapidly removed and placed in ice-cold high sucrose dissection solution. After isolation of visual cortex, coronal slices (350 µM) were prepared using a vibratome (Technical Products International),
IP
T
transferred to a room temperature incubation chamber and warmed to 30°C (Gu et al., 2014).
CR
Acute injury model
To induce acute injury, visual cortical slices were bisected through layers 1-4 using a 22
US
gauge scalpel blade and allowed to recover for one hour in the incubation buffer.
AN
Electrophysiology
Extracellular (EC) recording pipettes were made from borosilicate glass capillaries using Flaming/Brown model P-97 micropipette puller (Sutter Instruments CO.). The EC recording
M
electrodes were filled with 0.9% NaCl (1-2 MΩ). Electrodes were placed in layers 2-3 of visual
ED
cortical slices (Fig.1A). Multi-Clamp Commander (MCC) 700 amplifiers (Axon Instruments) were used for all electrical recordings. Data were low-pass filtered and digitized at 1 KHz for EC
PT
and 10 KHz for whole-cell recordings (Digidata; pCLAMP, Molecular Devices). CSD’s in individual pyramidal neurons of layers 2-3 of visual cortex were recorded using
CE
whole-cell patch clamp technique. Micropipettes (4-9MΩ) contained intracellular solution (in mM): 6 KCl, 0.5 EGTA, 20 HEPES, 10 phosphocreatine, 0.3 NaGTP, 2 NaCl, 4 MgATP;
AC
adjusted to pH 7.25 and osmolarity of 295 mOsm. Control ASCF was replaced with high KCl (26mM) ACSF and warmed to 36ºC (Fig.1B). After the initiation of CSD, high KCl solution was replaced with control ACSF. The pipette solution for voltage clamp Na+ current recording contained (mM) 5.8 NaCl, 134 CsCl, 1 MgCl2, 3 EGTA, and 10 HEPES (pH 7.3 adjusted with 9.2 mM NaOH) (Zhang et al., 2014). Slow ramp depolarizations (10-40 mV/1000ms) (Fig. 1C) from cell resting membrane potential were used to determine action potential threshold and cell spiking properties. Incremental hyperpolarizing and depolarizing current injections were used to study the passive
ACCEPTED MANUSCRIPT and active neuronal membrane properties (10 pA increments for 500-1000 ms) (Gu et al. 2014) (Fig.1D). Sodium currents were recorded using square-wave pulses in voltage clamp. To elicit fast voltage-gated sodium currents, cells were clamped at -80mV and depolarized to +30mV in 10mV increments, each 30ms in duration (Fig. 1E). Peak Na+ currents were measured before and in five minute intervals after TFB-TBOA exposure. We accounted for the effects of cell size on Na+
T
currents by calculating the Na+ current density (peak sodium current/ cell capacitance). To
IP
confirm that TFB-TBOA was affecting voltage gated Na+ channels, TTX was applied after 20
CR
minutes of TFB-TBOA incubation and Na+ currents were measured 5 minutes after TTX
US
exposure.
Imaging were
stained
with
voltage-sensitive
AN
Slices
naphthalenyl)ethenyl)-1-(3-sulfopropyl)pyridiniumhydroxide
dye
4-(2-(6-(Dibutylamino)-2-
(di-4-ANNEPS;
Sigma-Aldrich)
M
with the final concentration of 0.03 mg/ml. Dye was applied directly to the surface of slices and were incubated in no-light conditions at 30°C. Following the incubation period, individual slices 2-
l
in at
ED
were transferred to a recording chamber with oxygenated ACSF perfusing constantly at a rate of C. For a detailed description of the staining method, please see Hazra et al.,
PT
2013 and Gu et al., 2014. Upright wide-field epifluorescence microscope (BX51WI; Olympus) equipped with a fast CCD camera was used for recording and VSD signals (128×198 pixels,
CE
MiCam 02, SciMedia, USA). Imaging was done in conjunction with either whole-cell or EC recording of layer II-III of visual cortical slices. We determined locations of layers 2-3 based on
AC
the high density of pyramidal cells immediately below myelin dominant layer I. 150W halogen light source (Moritex Corp.) attached to a fiber optic cable was used to illuminate stained slices. Light was passed through a filter cube (excitation λ= 5
±1
n , dichroic λ=565 n
and
absorption λ>59 n , U-MWIG2, Olympus). Evoked optical signals were sampled at 250 Hz in a total of 90 second exposure time. We used brain vision analyzer (BV-Ana) software in analyzing the imaging data. We set the low pass filtration of DF/Fmax at 25% to increase the resolution of imaging for all samples. Moreover, we consistently set the signal threshold low, in order to detect small signals. To reduce the background noise, median filtering (3x3 pixels) was applied. Imaging and electrophysiological recording acquisitions were synchronized.
ACCEPTED MANUSCRIPT The CSD propagation velocity was measured by plotting a stripe from the start point to the end point. The software provides us the stripe length, its coordinates and the time duration of the propagation (Hazra et al, 2014). We refer to CSD wave propagation as laminar when the CSD propagated along the superficial layers (II-III) without expanding in to the deeper layers (IV-VI). When the signal clearly propagated between superficial and deep layers, we referred to it as
T
columnar.
IP
Data and Statistical analysis
CR
BrainVision (SciMedia), pClamp (v.10.1, Molecular Devices) and Graph pad Prism 5 software were used for analysis of optical and electrical data. Optical data were analyzed by using
US
the same setting for threshold and gain. All data are expressed in mean ± SEM. Unpaired student's t-tests with Welch's correction were used to determine significant differences between
AN
the two groups. For more than two group comparisons, one-way ANOVA followed by Newman-
AC
CE
PT
ED
M
Keuls were used. The value of p<0.05 was regarded as statistically significant.
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
Figure 1: Electrophysiological characteristics of CSD and firing properties of visual cortical
ED
neurons. (A) CSD in individual pyramidal neurons and neuronal network in layers 2-3 of visual cortex were recorded using whole-cell (WC) patch clamp and extracellular (EC) recording
PT
techniques. CSD in network was recorded by placing the borosilicate micropipettes 500 micrometers away from the whole-cell electrode in the layers 2-3. CSD in network and individual
CE
neurons was recorded simultaneously. (B) CSD was evoked by high KCl and identified based on a significant depolarization in both network and single cell. Typically, CSD in individual cells
AC
started roughly 2-10 seconds before CSD detected by the EC electrodes. CSD usually accompanied with spikes in the whole-cell recording. The black line indicates the duration of high K+ solution application. (C) Evoked APs in a representative neuron induced by depolarizing current ramp injections from -40pA to +60pA in 10pA increments into the soma for one second. AP threshold is measured as the beginning of the rapid depolarization of membrane potential following depolarizing current ramp injections. (D) AP discharge evoked in the rat neuron by square wave current pulses into the soma for one second. Square wave current injections ranging from -20pA to +80pA in 10pA increments were used. AP amplitude was measured as the voltage difference between the threshold and the peak value of the membrane potential during the AP. (E)
ACCEPTED MANUSCRIPT Sodium currents were recorded using square-wave pulses in voltage clamp. To elicit fast voltagegated sodium currents, cells were clamped at -80mV, then depolarized to +30mV in 10mV increments, each 30ms in duration.
RESULTS
T
Effect of TFB-TBOA on cell properties
IP
To assess effects of glutamate transporter blockade with TFB-TBOA on neuronal properties,
CR
we first quantified passive and active membrane properties in the cortical layer II-III pyramidal
AC
CE
PT
ED
M
AN
US
cells.
Figure 2: Effect of TFB-TBOA on neuronal membrane properties and excitability. AP’s were evoked by current pulses and studied under the current-clamp conditions in 9 neocortical slices. (A) TFB-TBOA (50nM) increased the cell membrane resistance after 10 minutes (p<0.05). (B) Current pulses evoked fewer AP’s in slices pretreated with 50nM TFB-TBOA (p<0.05, 0100pA current injection). (C) After 20min treatment with 50nM TFB-TBOA, AP threshold increased (p<0.05). (D) Pretreatment of visual cortical slices with TFB-TBOA decreased AP
ACCEPTED MANUSCRIPT amplitude (p<0.001). (E) TFB-TBOA significantly reduced fast voltage-gated sodium currents within 15 minutes (p<0.01). TTX co pletely blocked the channels’ activity within 5
inutes of
incubation. CSD in visual cortical slices To study CSD in visual cortex slices, we used a combination of electrophysiology and VSDI. Emergent CSDs were induced with high concentration of KCl in bath (Fig. 1B). EC and
T
intracellular (IC) electrodes were placed in the layer 2-3 of visual cortex about 500 micrometers
IP
apart. In high K+ ACSF, CSD typically occurred 30-45 second after high KCl application. CSD in
CR
the single cells started with a rapid depolarization followed by several spikes and a slow return to the resting membrane potential. The majority of cell spiking occurred before maximum
US
depolarization was reached. CSD in the EC recording was typically noticeable a few seconds
AC
CE
PT
ED
M
AN
later. CSD in single cells typically lasted for 30-180 seconds.
Figure 3: Electrophysiological properties of CSD in TBI and TFB-TBOA treated groups. (A) The network and single cell properties of CSD in control, TFB-TBOA, and TBI treated groups. (B) TFB-TBOA increased CSD latency in single neurons (n=9, p<0.001). (C) Both TBI and TFB-TBOA (n=9) prolonged CSD duration compare with the control group (p<0.001). (D&
ACCEPTED MANUSCRIPT E) Both, depolarizing and hyperpolarizing slope of CSD decreased in TBI or TFB-TBOA treated group in the individual neurons (p<0.05, n =9). (F) Network CSD amplitude in the presence of TFB-TBOA or TBI (n=9). Both, TFB-TBOA and TBI decreased the mean of amplitude compared to the control (p<0.001). (G) In network recordings, TFB-TBOA prolonged CSD in comparison to the control group (p<0.01), but the CSD duration differences between the TBI and control were not statistically significant. (H) The network frequency of CSD incidence in the
T
presence of TFB-TBOA or TBI. TFB-TBOA decreases the frequency of CSD incidence
IP
compared to both TBI (p<0.05) and control (p<0.01) groups. The differences between the TBI
CR
and control group was not statistically significant
Effects of TBI model and TFB-TBOA on CSD in individual neurons and neuronal network
US
Both TBI and TFB-TBOA altered CSD characteristics in both network and single cell patchclamp recording (Fig. 3A). We quantified spontaneous action potentials before, during, and
AN
immediately after the initial CSD event. TBI reduced the number of action potentials to 24.50 ± 6.98 (control 54.71± 10.91) (p=0.046), while TFB-TBOA completely blocked the action
M
potentials. Additionally, resting membrane potential was unaffected under either condition. CSD latency was measured based on the time that it took to form the first CSD after the application of
ED
high KCl solution. Surprisingly, our results showed that TFB-TBOA increased the latency to the first CSD (p<0.0001) (Fig. 3B). CSD duration was defined and measured as the time from the
PT
initiation of rapid depolarization in individual neurons and network to the time when the membrane potential repolarized to its pre-CSD potential value. Either incubation of coronal slices
CE
with TFB-TBOA for 20 minutes or acute cut caused prolonged CSD (Fig. 3C). In both conditions, CSD depolarization and repolarization slopes were shallower (Fig. 3D & E).
CSD wave
AC
depolarized with the rate of 5.12± 0.81 (mV/ms) in control slices, while it decreased to 2.23± 0.41 (mV/ms) and 2.06± 0.42 (mV/ms) in the TBI and TFB-TBOA, respectively. The same trend was observed in the repolarizing slope. The rate at which the CSD wave typically repolarized was 1.71± 0.12 (mV/ms) in control slices, 0.81± 0.08 (mV/ms) in cut slices and 0.82± 0.11 (mV/ms) in TFB-TBOA groups. We next measured the CSD amplitude, duration, and inter-CSD interval obtained from EC recordings of the local neuronal network. CSD amplitude significantly decreased in the presence of TFB-TBOA or acute injury (Fig. 3F). The duration of network CSD was significantly increased in response to TFB-TBOA; whereas, acute injury increased duration, but not
ACCEPTED MANUSCRIPT significantly (Fig. 3G). In order to quantify how TFB-TBOA or acute injury affects network CSD frequency, we measured the duration of the time interval between subsequent CSDs and found that TFB-TBOA prolonged inter-CSD interval in both, the network and single cell recordings (Fig. 3H). Our data showed that although both TFB-TBOA and TBI increased the CSD occurrence rate; the inter-CSD interval was significantly shorter in the presence of TFB-TBOA (Fig. 3H).
ED
M
AN
US
CR
IP
T
Spatiotemporal patterns of CSD and their modulation by TBI and TFB-TBOA:
PT
Figure 4: Spatiotemporal properties of CSD in control, TBI model and TFB-TBOA treated rat visual cortical slices. (A) Initiation and propagation of CSD in the control slices. CSD
CE
initiated in layers 2-3 of visual cortex, and spread in a laminar pattern. In addition, CSD also had columnar pattern with propagation to the adjacent columns (n=7). (B) The spatiotemporal
AC
properties of CSD in TBI model. CSD does not initiate at the site of the injury. CSD initiated close to the injured site in layers 2-3 and propagated toward the deep layers and along the actual lesion (n=7). (C) CSD spatiotemporal properties in slices pretreated for 20 minutes with 50nM TFB-TBOA. CSD initiated in the deeper layers of cortical slices. CSD formed predominant columnar patterns. It did not have clear propagation and, instead, while remaining stationary, gradually increased in signal strength (n=7). (D) CSD propagation velocity in the control and TBI model. CSD propagated significantly (p<0.05) slower in the TBI model than the control slices. (E) The initiation point of CSD in control, TBI model and TFB-TBOA pretreated slices. Although
ACCEPTED MANUSCRIPT all CSDs in control and TBI model initiated in the superficial layers, TFB-TBOA changed the initiation point of CSD to layer 4-5 of cortical slices. To observe whether TFB-TBOA or the acute cut changes spatiotemporal patterns of CSD, fast voltage-sensitive dye imaging techniques were used simultaneously with electrophysiological recordings. Concurrent imaging with EC and IC recordings in cortical layers II-III allowed for the clear identification of specific CSD depolarizing wave pattern characteristics. When the CSD
T
origin was quantified using threshold analysis, it was observed that CSD in the intact cortical
IP
slices originated from confined spatial areas in layers II-III. The waves then propagated mostly in
CR
a laminar manner (7 out of 10) and sometimes in a curve-like shape that extended into the columnar cortical structure, spanning from layer II to V-VI (3 out of 10) (Fig. 4A).
US
Based on previous reports, we hypothesized that the acute cut would likely be the source of CSD in the injured coronal slices; however, we found that CSD initiated nearby the area of the
AN
injury in layers II-III, but then propagated toward the cut in the slice in either curved columnar (5 out of 7) or laminar (2 out of 7) waves (Fig. 4B). On the other hand, in all TFB-TBOA treated
M
slices (intact and cut) CSD initiated in layers II-V, then formed a semicircular shape and a columnar pattern including both layers II-III and IV-V of the visual cortex (Fig. 4C).
ED
Finally, acute cut reduced the CSD propagation rate significantly (Fig. 4D). TFB-TBOA seemed to have trapped CSD in the layers where it typically initiates, causing CSD depolarization
PT
to increase in strength locally and expanded from its origin without showing clear layer or column-specific propagation across the tissue (Fig. 4D). CSD initiated from layers II-III in all
CE
slices of the control and TBI groups. In TFB-TBOA, however, CSD initiated in layers II-III and IV-V (Fig. 4E). Thus, global glutamate transport blockade has shifted the site of CSD initiation
AC
and the localized cut acted as an attractor of the depolarizing wave. This shift of the active signal origins into the deep layers suggests a layer-specific regulation of glutamate and spatiotemporal properties of CSD in distinct traumatic conditions.
DISCUSSION Slowly propagating (2-5 mm/min) CSD is thought to initiate in layers II-III of the neocortex, accompanied by spreading depolarization and neuronal and glial swelling (Basarsky et al., 1999; Kimelberg et al., 1990), and depression of activity (Lauritzen et al., 2011; Larrosa et al., 2006). During CSD, inward cationic Ca+2 and Na+ currents exceed the outward K+ current. This
ACCEPTED MANUSCRIPT pathological cation influx causes significant cell depolarization and glutamate release (Van and Fifkova, 1970; Fabricius et al., 1993; Somjen, 2001). Increased glutamate leads to increased activation of glutamate receptors, especially NMDARs, and further CSD propagation (Kramer, 2016; Lauritzen, 2011; Torrente et al., 2014 (a); Eikermann-Haerter and Ayata , 2010). Astrocytic glutamate transporters play the decisive role in clearance of glutamate and thus, CSD properties. Glutamate transporters, GLT-1 and GLAST, are prominently expressed in
T
astrocytes and are responsible for more than 90% of extracellular glutamate re-uptake (Sheldon
IP
and Robinson., 2007). GLT-1 also has a significant effect on the extracellular glutamate buffering
CR
during CSD (Torrente et al., 2014 (a)). We showed that TFB-TBOA, a selective glutamate transporter blocker, reduced cell excitability and CSD amplitude. Moreover, the blocker delayed
US
the initiation of CSD, shifted its spatial formation to deeper cortical layers, and reduced its propagation velocity. These findings suggest that pharmacological manipulation of glutamate
AN
transporters could be considered as a potential strategy for modulating CSDs associated with
M
neurological conditions, such as migraines, TBI, and stroke.
Effects of TFB-TBOA on single cell and network activity
ED
TFB-TBOA had significant effects on single cell and cortical network excitability. TFBTBOA increased cell membrane resistance, and reduced sodium currents. It also decreased
PT
neuronal excitability by increasing the threshold and reducing amplitude of evoked action potentials. In addition, TFB-TBOA prolonged CSD while decreasing CSD amplitude and cell
CE
spiking activity. Prolonged CSDs were also reported using 0.5 and 1mM of DL-TBOA, which was suggested to be due to activation of pre-synaptic NMDA receptors (Hinzman et al., 2015).
AC
Interestingly, Bozzo et al. showed that TFB-TBOA inactivates Na+ channels in IC50 43nM (Bozzo et al., 2010). In our study, TFB-TBOA had significant effects on neuronal excitability and reduced Na+ influx, suggesting that the effect of TFB-TBOA on neuronal activity is probably due to its direct effects on voltage-gated sodium channels, in addition to the blockade of glutamate transport in astrocytes. This dysregulation of sodium influx and action potentials in neurons by TFB-TBOA is likely contributing to the spread of CSD through mechanisms other than the chemical neurotransmission. Although glutamate transporters are thought to facilitate glutamate uptake into the astrocytes, recent studies show that high extracellular K+ or the presence of high levels of intracellular
ACCEPTED MANUSCRIPT glutamate causes these transporters to work in reverse (Zhang, 2007; Szatkowski et al., 1990). Thus, in the presence of high K+ and hyperexcitability, like CSD, reversal of glutamate transport in astrocytes could become an additional unwanted source of extracellular glutamate build-up, causing large and prolonged depolarizations. The blockade of GLT transporters in our experimental conditions may have prevented glutamate transport reversal and caused glutamate to be trapped inside the glia. In this way, blockade of glutamate transport may counter-intuitively
T
decrease pathological glutamate reversal and extracellular glutamate levels, partly accounting for
IP
the reduced depolarization in the presence of TFB-TBOA, as well as decreased cell excitability.
CR
Under different experimental conditions, however, it was also observed that a different glutamate transporter blocker (D,L-threo-beta-hydroxyaspartate (THA)) increased the amount of
US
depolarization in the presence of high K+ (Larrosa et al., 2006).
AN
Effects of TFB-TBOA on CSD spatial formation and propagation VSDI experiments show that in compromised glutamate transport conditions, CSD dominates
M
the deep cortical layers. In most instances, CSD was confined and, when it spread, it did so slowly, indicating that TFB-TBOA compromises CSD propagation. One explanation regarding
ED
the effects of TFB-TBOA on CSD propagation could be attributed to glutamate transport. Glutamate, in addition to control of CSD amplitude, is also known to modulate CSD initiation and
PT
propagation. For example, when NMDA receptors are blocked, CSD is reduced in size and area (Marrannes et al. 1988; Tozzi et al. 2012). Blocking AMPA, NMDA, and kainate glutamate
CE
receptors in the rat visual cortex slices using kyuneric acid traps CSD in the deeper coronal layers and prevents its propagation (Világi et al. 2001). Kyuneric acid in this case also compromises
AC
synaptic transmission. Not surprisingly, multiple reports indicate that during CSD, synaptic transmission is impaired (Somjen, 2004). Another explanation for the effect of TFB-TBOA on CSD propagation would be via its effect on the TTX-sensitive sodium channels. Tozzi et al. showed that blockade of the TTX-sensitive sodium channels in the neocortical slices reduced the CSD area (Tozzi et al., 2012). The CSD propagation velocity in the current study is 3±0.62 mm/min in control slices, which is similar to the reported velocity by Torrente (Torrente et al. 2014 (b)). CSD seemed to have been trapped by TFB-TBOA, and it is likely that this effect on propagation was via direct control of sodium channels and compromised chemical neurotransmission.
ACCEPTED MANUSCRIPT
Changes of CSD properties in the TBI model TBI is often accompanied by CSD and increases in extracellular glutamate (Lauritzen et al., 2011; Rao, 1998). In addition, formation of reactive astrocytes increases following TBI (Myer et al. 2006) while GLT-1 expression is reduced following TBI and stroke (Rao et al. 1998). Here, following acute cut in the cortex, CSD was prolonged, and spiking activity and velocity of CSD
T
propagation were decreased (1.33± 0.3 mm/min). In addition, our imaging data showed that the
IP
acute injury changed CSD spatiotemporal characteristics. Surprisingly, CSD did not initiate at the
CR
site of the actual injury, but rather in the surrounding peri-injury zone. It seemed that the networks at the cut itself were incapable of initiating the CSD. This is likely due to the physical trauma and
US
compromised neural connections.
Using a freeze lesion model, Dulla et al. (2013) found that TBI alters the input-output
AN
response and decreases the evoked network responses. In addition, the freeze lesion caused glutamate signal to shift to the deeper cortical layers (V), away from the lesion zone. This also
M
suggests that the networks in the injured area itself, similarly to our results, may not be capable of initiating CSD. Instead, the injury zone seems to attract and have the ability to sustain CSDs
ED
initiated by nearby, less compromised networks.
PT
CONCLUSION
CE
The results demonstrate that both TBI and astrocytic glutamate transporters modulate neuronal excitability in addition to the electrophysiological and spatiotemporal properties of CSD.
AC
The cortical injury site which is not capable of initiating CSD, instead, acts as an attractor to the depolarizing CSD waves. These results provide new insights about the mechanisms of CSD, and support the idea that glutamate transporters may become viable drug targets for control and/or modulation of the spreading depression.
ACCEPTED MANUSCRIPT References Basarsky, T. A., Feighan, D., & MacVicar, B. A. (1999). Glutamate release through volumeactivated channels during spreading depression. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 19(15), 6439-6445.
T
Bozzo, L., & Chatton, J. Y. (2010). Inhibitory effects of (2S, 3S)-3-[3-[4-
IP
(trifluoromethyl)benzoylamino]benzyloxy]aspartate (TFB-TBOA) on the astrocytic sodium responses to glutamate. Brain Research, 1316, 27-34. doi:10.1016/j.brainres.2009.12.028
CR
[doi]
US
Bures, J. (1957). The ontogenetic development of steady potential differences in the cerebral cortex in animals. Electroencephalography and Clinical Neurophysiology, 9(1), 121-130.
AN
Chuquet, J., Hollender, L., & Nimchinsky, E. A. (2007). High-resolution in vivo imaging of the neurovascular unit during spreading depression. The Journal of Neuroscience : The Official
M
Journal of the Society for Neuroscience, 27(15), 4036-4044. doi:27/15/4036 [pii]
ED
de Luca, B., & Bures, J. (1977). Development of cortical spreading depression and of its transition to the caudate nucleus in rats. Developmental Psychobiology, 10(4), 289-297.
PT
doi:10.1002/dev.420100402 [doi]
CE
Dulla, C. G., Tani, H., Brill, J., Reimer, R. J., & Huguenard, J. R. (2013). Glutamate biosensor imaging reveals dysregulation of glutamatergic pathways in a model of developmental
[doi]
AC
cortical malformation. Neurobiology of Disease, 49, 232-246. doi:10.1016/j.nbd.2012.09.001
Eikermann-Haerter, K., & Ayata, C. (2010). Cortical spreading depression and migraine. Current Neurology and Neuroscience Reports, 10(3), 167-173. doi:10.1007/s11910-010-0099-1 [doi] Eiselt, M., Giessler, F., Platzek, D., Haueisen, J., Zwiener, U., & Rother, J. (2004). Inhomogeneous propagation of cortical spreading depression-detection by electro- and magnetoencephalography in rats. Brain Research, 1028(1), 83-91. doi:S00068993(04)01482-9 [pii]
ACCEPTED MANUSCRIPT Fabricius, M., Fuhr, S., Bhatia, R., Boutelle, M., Hashemi, P., Strong, A. J., & Lauritzen, M. (2006). Cortical spreading depression and peri-infarct depolarization in acutely injured human cerebral cortex. Brain : A Journal of Neurology, 129(Pt 3), 778-790. doi:awh716 [pii] Fabricius, M., Fuhr, S., Willumsen, L., Dreier, J. P., Bhatia, R., Boutelle, M. G., . . . Lauritzen, M. (2008). Association of seizures with cortical spreading depression and peri-infarct
T
depolarisations in the acutely injured human brain. Clinical Neurophysiology : Official
IP
Journal of the International Federation of Clinical Neurophysiology, 119(9), 1973-1984.
CR
doi:10.1016/j.clinph.2008.05.025 [doi]
Fabricius, M., Jensen, L. H., & Lauritzen, M. (1993). Microdialysis of interstitial amino acids
US
during spreading depression and anoxic depolarization in rat neocortex. Brain
AN
Research, 612(1-2), 61-69. doi:0006-8993(93)91644-8 [pii] Goodrich, G. S., Kabakov, A. Y., Hameed, M. Q., Dhamne, S. C., Rosenberg, P. A., &
M
Rotenberg, A. (2013). Ceftriaxone treatment after traumatic brain injury restores expression of the glutamate transporter, GLT-1, reduces regional gliosis, and reduces post-traumatic
ED
seizures in the rat. Journal of Neurotrauma, 30(16), 1434-1441. doi:10.1089/neu.2012.2712 [doi]
PT
Gorji, A. (2001). Spreading depression: A review of the clinical relevance. Brain Research.Brain
CE
Research Reviews, 38(1-2), 33-60. doi:S0165017301000819 [pii] Gu, F., Hazra, A., Aulakh, A., & Ziburkus, J. (2014). Purinergic control of hippocampal circuit
AC
hyperexcitability in dravet syndrome. Epilepsia, 55(2), 245-255. doi:10.1111/epi.12487 [doi] Hazra, A., Gu, F., Aulakh, A., Berridge, C., Eriksen, J. L., & Ziburkus, J. (2013). Inhibitory neuron and hippocampal circuit dysfunction in an aged mouse model of alzheimer's disease. PloS One, 8(5), e64318. doi:10.1371/journal.pone.0064318 [doi] Hazra, A., Macolino, C., Elliott, M. B., & Chin, J. (2014). Delayed thalamic astrocytosis and disrupted sleep-wake patterns in a preclinical model of traumatic brain injury. Journal of Neuroscience Research, 92(11), 1434-1445. doi:10.1002/jnr.23430 [doi]
ACCEPTED MANUSCRIPT Hinzman, J. M., DiNapoli, V. A., Mahoney, E. J., Gerhardt, G. A., & Hartings, J. A. (2015). Spreading depolarizations mediate excitotoxicity in the development of acute cortical lesions. Experimental Neurology, 267, 243-253. doi:10.1016/j.expneurol.2015.03.014 [doi] Hossmann, K. A. (1996). Periinfarct depolarizations. Cerebrovascular and Brain Metabolism
T
Reviews, 8(3), 195-208.
IP
Kimelberg, H. K., Goderie, S. K., Higman, S., Pang, S., & Waniewski, R. A. (1990). Swellinginduced release of glutamate, aspartate, and taurine from astrocyte cultures. The Journal of
CR
Neuroscience : The Official Journal of the Society for Neuroscience, 10(5), 1583-1591.
US
Kramer, D. R., Fujii, T., Ohiorhenuan, I., & Liu, C. Y. (2016). Cortical spreading depolarization: Pathophysiology, implications, and future directions. Journal of Clinical Neuroscience :
doi:10.1016/j.jocn.2015.08.004 [doi]
AN
Official Journal of the Neurosurgical Society of Australasia, 24, 22-27.
M
Larrosa, B., Pastor, J., Lopez-Aguado, L., & Herreras, O. (2006). A role for glutamate and glia in
ED
the fast network oscillations preceding spreading depression. Neuroscience, 141(2), 10571068. doi:S0306-4522(06)00488-X [pii]
PT
Lauritzen, M., Dreier, J. P., Fabricius, M., Hartings, J. A., Graf, R., & Strong, A. J. (2011). Clinical relevance of cortical spreading depression in neurological disorders: Migraine,
CE
malignant stroke, subarachnoid and intracranial hemorrhage, and traumatic brain injury. Journal of Cerebral Blood Flow and Metabolism : Official Journal of the
AC
International Society of Cerebral Blood Flow and Metabolism, 31(1), 17-35. doi:10.1038/jcbfm.2010.191 [doi] Malkov, A., Ivanov, A. I., Popova, I., Mukhtarov, M., Gubkina, O., Waseem, T., . . . Zilberter, Y. (2014). Reactive oxygen species initiate a metabolic collapse in hippocampal slices: Potential trigger of cortical spreading depression. Journal of Cerebral Blood Flow and Metabolism : Official Journal of the International Society of Cerebral Blood Flow and Metabolism, 34(9), 1540-1549. doi:10.1038/jcbfm.2014.121 [doi]
ACCEPTED MANUSCRIPT Marrannes, R., Willems, R., De Prins, E., & Wauquier, A. (1988). Evidence for a role of the Nmethyl-D-aspartate (NMDA) receptor in cortical spreading depression in the rat. Brain Research, 457(2), 226-240. doi:0006-8993(88)90690-7 [pii] Myer, D. J., Gurkoff, G. G., Lee, S. M., Hovda, D. A., & Sofroniew, M. V. (2006). Essential protective roles of reactive astrocytes in traumatic brain injury. Brain : A Journal of
IP
T
Neurology, 129(Pt 10), 2761-2772. doi:awl165 [pii]
Obrenovitch, T. P. (1995). The ischaemic penumbra: Twenty years on. Cerebrovascular and
CR
Brain Metabolism Reviews, 7(4), 297-323.
US
Obrenovitch, T. P., & Zilkha, E. (1996). Inhibition of cortical spreading depression by L-701,324, a novel antagonist at the glycine site of the N-methyl-D-aspartate receptor complex. British
AN
Journal of Pharmacology, 117(5), 931-937.
Rao, V. L., Baskaya, M. K., Dogan, A., Rothstein, J. D., & Dempsey, R. J. (1998). Traumatic
M
brain injury down-regulates glial glutamate transporter (GLT-1 and GLAST) proteins in rat
ED
brain. Journal of Neurochemistry, 70(5), 2020-2027. Rogatsky, G., Mayevsky, A., Zarchin, N., & Doron, A. (1996). Continuous multiparametric
PT
monitoring of brain activities following fluid-percussion injury in rats: Preliminary
CE
results. Journal of Basic and Clinical Physiology and Pharmacology, 7(1), 23-43. Rothstein, J. D., Dykes-Hoberg, M., Pardo, C. A., Bristol, L. A., Jin, L., Kuncl, R. W., Welty, D.
AC
F. (1996). Knockout of glutamate transporters reveals a major role for astroglial transport in excitotoxicity and clearance of glutamate. Neuron, 16(3), 675-686. doi:S08966273(00)80086-0 [pii] Sheldon, A. L., & Robinson, M. B. (2007). The role of glutamate transporters in neurodegenerative diseases and potential opportunities for intervention. Neurochemistry International, 51(6-7), 333-355. doi:S0197-0186(07)00077-0 [pii] Shimamoto, K., Sakai, R., Takaoka, K., Yumoto, N., Nakajima, T., Amara, S. G., & Shigeri, Y. (2004). Characterization of novel L-threo-beta-benzyloxyaspartate derivatives, potent
ACCEPTED MANUSCRIPT blockers of the glutamate transporters. Molecular Pharmacology, 65(4), 1008-1015. doi:10.1124/mol.65.4.1008 [doi] Somjen, G. G. (2001). Mechanisms of spreading depression and hypoxic spreading depressionlike depolarization. Physiological Reviews, 81(3), 1065-1096.
T
Somjen, G. G. (2004). Ions in the brain: Normal function, seizures, and stroke. Oxford ; New
IP
York: Oxford University Press.
CR
Strong, A. J., Fabricius, M., Boutelle, M. G., Hibbins, S. J., Hopwood, S. E., Jones, R., . . . Lauritzen, M. (2002). Spreading and synchronous depressions of cortical activity in acutely
US
injured human brain. Stroke, 33(12), 2738-2743.
Szatkowski, M., Barbour, B., & Attwell, D. (1990). Non-vesicular release of glutamate from glial
AN
cells by reversed electrogenic glutamate uptake. Nature, 348(6300), 443-446.
M
doi:10.1038/348443a0 [doi]
Torrente, D., Cabezas, R., Avila, M. F., Garcia-Segura, L. M., Barreto, G. E., & Guedes, R. C.
ED
(2014a). Cortical spreading depression in traumatic brain injuries: Is there a role for
PT
astrocytes? Neuroscience Letters, 565, 2-6. doi:10.1016/j.neulet.2013.12.058 [doi] Torrente, D., Mendes-da-Silva, R. F., Lopes, A. A., Gonzalez, J., Barreto, G. E., & Guedes, R. C.
CE
(2014b). Increased calcium influx triggers and accelerates cortical spreading depression in vivo in male adult rats. Neuroscience Letters, 558, 87-90. doi:10.1016/j.neulet.2013.11.004
AC
[doi]
Tozzi, A., de Iure, A., Di Filippo, M., Costa, C., Caproni, S., Pisani, A., . . . Calabresi, P. (2012). Critical role of calcitonin gene-related peptide receptors in cortical spreading depression. Proceedings of the National Academy of Sciences of the United States of America, 109(46), 18985-18990. doi:10.1073/pnas.1215435109 [doi] Van Harreveld, A., & Fifkova, E. (1970). Glutamate release from the retina during spreading depression. Journal of Neurobiology, 2(1), 13-29. doi:10.1002/neu.480020103 [doi]
ACCEPTED MANUSCRIPT Vilagi, I., Klapka, N., & Luhmann, H. J. (2001). Optical recording of spreading depression in rat neocortical slices. Brain Research, 898(2), 288-296. doi:S0006-8993(01)02196-5 [pii] Zhang, Y., He, J. C., Liu, X. K., Zhang, Y., Wang, Y., & Yu, T. (2014). Assessment of the effect of etomidate on voltage-gated sodium channels and action potentials in rat primary sensory cortex pyramidal neurons. European Journal of Pharmacology, 736, 55-62.
IP
T
doi:10.1016/j.ejphar.2014.04.036 [doi]
Zhang, Z., Tao, Z., Gameiro, A., Barcelona, S., Braams, S., Rauen, T., & Grewer, C. (2007).
CR
Transport direction determines the kinetics of substrate transport by the glutamate transporter EAAC1. Proceedings of the National Academy of Sciences of the United States of
US
America, 104(46), 18025-18030. doi:0704570104 [pii]
AN
Zhou, N., Rungta, R. L., Malik, A., Han, H., Wu, D. C., & MacVicar, B. A. (2013). Regenerative glutamate release by presynaptic NMDA receptors contributes to spreading
M
depression. Journal of Cerebral Blood Flow and Metabolism : Official Journal of the International Society of Cerebral Blood Flow and Metabolism, 33(10), 1582-1594.
AC
CE
PT
ED
doi:10.1038/jcbfm.2013.113 [doi]
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
Graphical abstract
ACCEPTED MANUSCRIPT
Highlights
IP
sodium currents, network hyperexcitability, and CSD characteristics.
T
Glutamate transporter blocker TFB-TBOA has significant effects on neuronal voltage-gated
CR
Acute cortical cut and compromised glutamate transporter modulate physiological and
US
spatiotemporal properties of CSD.
AN
The area of the actual cortical cut does not initiate CSDs, but in the surrounding areas. The
M
actual injury site attracts propagating CSD depolarizing waves.
AC
CE
PT
ED
TFB-TBO significantly affects sodium conductances in neurons.
Mahshid Sadat Hosseini-Zare, Feng Gu, Ahmad Abdulla, Simon Powell, Jokūbas Žiburkus PII: DOI: Reference:
S0014-4886(17)30119-X doi: 10.1016/j.expneurol.2017.05.002 YEXNR 12530
To appear in:
Experimental Neurology
Received date: Revised date: Accepted date:
18 July 2016 14 April 2017 3 May 2017
Please cite this article as: Mahshid Sadat Hosseini-Zare, Feng Gu, Ahmad Abdulla, Simon Powell, Jokūbas Žiburkus , Effects of experimental traumatic brain injury and impaired glutamate transport on cortical spreading depression, Experimental Neurology (2017), doi: 10.1016/j.expneurol.2017.05.002
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Effects of experimental traumatic brain injury and impaired glutamate transport on cortical spreading depression.
Mahshid Sadat Hosseini-Zare1,2, Feng Gu1Ɣ, Ahmad Abdulla1,2, Simon Powell1, and Jokūbas Žiburkus*1,2 University of Houston, Department of Biology and Biochemistry; 2Texas Institute for
T
1
IP
Measurement and Evaluation Statistics (TIMES), Houston, Texas 77024, Houston, Texas 77204,
US
CR
U.S.A.
AN
* Corresponding author: Jokūbas Žiburkus
M
Associate Professor
University of Houston, Department of Biology and Biochemistry and Texas Institute for
Ɣ
PT
Email: [email protected]
ED
Measurement, Evaluation, and Statistics (TIMES) Houston, TX 77204, U.S.A.
– Current address: Feng Gu, University of Stanford Medical School, Department of
AC
CE
Neurology and Neurobiology Sciences, Stanford, CA 94305, U.S.A.
Acknowledgment: This research was sponsored by the Dravet Syndrome Foundation and University of Houston’s Grants to Advance and Enhance Research (GEAR) funds.
ACCEPTED MANUSCRIPT ABSTRACT
Cortical spreading depression (CSD) is associated with traumatic brain injury (TBI), stroke, migraines, and seizures. Typically, following TBIs and other insults, neuronal excitability in and around the area of the injury is affected, with reported increases in local glutamate signaling. Astrocytic glutamate transporters are critical for precise regulation of the extracellular glutamate
T
availability. However, it remains unclear how impaired astrocytic glutamate transport or an acute
IP
TBI affect characteristics of the CSD. We quantified the properties of CSD using whole-cell and
CR
extracellular electrophysiological recordings, and voltage-sensitive dye imaging (VSDI) in rat visual cortex in vitro. To model impaired astrocytic glutamate transport, we used astrocytic
US
glutamate transporter blocker (2S, 3S)-3-[3-[4-(trifluoromethyl) benzoylamino] benzyloxy] aspartate (TFB-TBOA). In addition, an acute incision through the superficial cortical layers was
AN
used to model the effects of acute traumatic brain injury (TBI) on CSD characteristics. Both manipulations; impaired glutamate cycling and acute cut profoundly affected the physiological
M
properties of cell firing, latency to CSD formation, and its frequency of occurrence. VSD imaging analysis revealed significant changes in spatiotemporal dynamics and propagation of the CSD,
ED
suggesting that that the cut itself may not initiate CSD depolarizing waves, but rather attract them. Blockade of GLT-1 caused significant reduction in whole-cell sodium currents and changes in
PT
CSD wave spatiotemporal characteristics as well, slowing it or even ‘trapping’ its propagation. Our results reveal new information about CSD properties in these pathological conditions and
AC
CE
demonstrate an important role of GLT-1 in regulation of CSD.
Key words: Cortical spreading depression, traumatic brain injury, astrocytic glutamate transporter
ACCEPTED MANUSCRIPT INTRODUCTION CSD is associated with several neurological disorders including stroke, migraine aura (Fabricius et al., 2006; Hossmann, 1996), epilepsy (Gorji. 2001), cerebral ischemia (Obrenovitch, 1995), hypoxia (Somjen, 2001), and TBI (Rogatsky, 1996; Strong et al., 2002). CSD can be observed in damaged brain regions and during the course of epileptic seizures (Lauritzen et al., 2011), but it can also be induced experimentally by raising glutamate or K+, manipulating Na+/K+
T
pump inhibitors, by using electrical stimulation or hyperthermia (Malkov et al., 2014).
IP
Electrically, CSD is characterized by profound depolarization of neuronal networks, followed by
CR
slow propagation (2-5 mm/min) of this depolarizing wave (Lauritzen et al., 2011; Larrosa et al., 2006). During the initial depolarizing portion of CSD, the extracellular glutamate and aspartate
US
concentrations increase (Van and Fifkova, 1970; Fabricius et al., 1993). Activation of N-methylD-aspartate (NMDA)-type glutamate receptors (NMDA-Rs) and increase in glutamate release
AN
further contribute to CSD progression (Obrenovitch and Zilkha, 1996; Zhou et al., 2013). Although mechanisms of CSD initiation and propagation are controversial, studies show that
M
both neurons and astrocytes are involved in this process (Eiselt et al., 2004). Extracellular glutamate levels are tightly regulated by glutamate transporter proteins located primarily on
ED
astrocytes (Rothstein et al., 1996). GLT-1 and GLAST are the two main glial glutamate transporters chiefly responsible for maintaining extracellular levels of glutamate. Despite the
PT
importance of glutamate clearance through the astrocytic transporters, to date it remains unknown how acute impairments in glutamate transport affect CSD.
CE
Furthermore, CSD is observed following TBI and seizures associated with it. Severe injuries that cause physical damage to the cortical tissue could become initiation zones for seizures and
AC
CSD (Fabricius et al., 2008). Following TBI, increased glutamate levels could also trigger seizures and/or CSDs (Goodrich et al., 2013). However, it remains unknown how an acute cut in the cortex can affect initiation site of CSDs in the cortex and its propagation in relation to the site of the injury. To address the role of glutamate and acute injury in CSD incidence and propagation, we employed electrophysiological, pharmacological, and high speed VSD imaging techniques. We quantified changes in CSD characteristics in two conditions: 1) in the presence of astrocytic glutamate transporter GLT-1 blocker TFB-TBOA, and 2) following acute cut through the superficial neocortical layers in vitro. Our findings suggest acute injury or TFB-TBOA
ACCEPTED MANUSCRIPT application significantly affected the initiation, duration, and spatiotemporal electrical properties of the CSD; however, the degree to which these effects could be attributed to dysregulation of astrocytic glutamate transport alone was confounded by the observation that TFB-TBOA reduced sodium conductance in cortical pyramidal cells. Thus, acute injury or concurrent blockade of astrocytic glutamate and neuronal sodium currents were responsible for the significant modulation
T
of CSD characteristics.
CR
IP
MATERIAL AND METHODS
Animals
US
Experiments were approved by University of Houston Institutional Animal Care and Use Committee. Studies were performed on juvenile, 15-24 days old, male wild-type Sprague Dawley
AN
rats. Potassium-induced spreading depression is well documented to form in the second postnatal week and is well developed by P15 (Bures, 1957; de Luca and Bures, 1977; Chuquet et al., 2007).
M
We used 5 animals for Na+ current recording, 5 animals for the TBI group, 7 animals for the TFBTBOA group and for each group of imaging, and 9 animals as controls. All numbers (n) refer to
ED
number of slices used in the experiments.
PT
Solutions and drugs
For tissue dissection and slice preparation, oxygenated (95% O2, 5% CO2) high sucrose
CE
dissection buffer (in mM): 248.3 sucrose, 2.5 KCl, 1.25 NaH2PO4, 3 MgSO4, 10 MgCl2, 0.5 CaCl2, 26 NaHCO3, and 10 dextrose) was used. Slices were pre-incubated in aerated (95% O2-
AC
5% CO2) standard artificial cerebrospinal fluid (ACSF) containing (in mM) 130 NaCl, 1.2 MgSO4, 3.5 KCl, 1.2 CaCl2, 10 glucose, 2.5 NaH2PO4, and 24 NaHCO3 (pH 7.3). The bath solution for voltage-clamp whole-cell sodium current recordings used ACSF containing (in mM): 95 NaCl, 5.6 KCl, 0.1 CaCl2, 5 MgCl2, 11 glucose, 1 NaH2PO4, 25 NaHCO3 and 20 TEA-Cl (pH 7.4 when bubbled with 95% O2–5% CO2) (Zhang et al., 2014). For induction of CSD, high K+ solution was made by an equimolar replacement of NaCl with 26 mM KCl. 2% Dimethyl sulfoxide (DMSO) and 98% deionized water solution was used to dissolve TFB-TBOA powder (Tocris Bioscience). TFB-TBOA was used with the final concentration of 50 nM: the IC50 value for GLT-1 and GLAST are 17 nM and 22 nM , respectively (Shimamoto et al., 2004).
ACCEPTED MANUSCRIPT
Visual cortex slices preparation Standard in vitro electrophysiology slice preparation techniques were used. In brief, the rats were deeply anesthetized with ether and decapitated using a guillotine. The brain was rapidly removed and placed in ice-cold high sucrose dissection solution. After isolation of visual cortex, coronal slices (350 µM) were prepared using a vibratome (Technical Products International),
IP
T
transferred to a room temperature incubation chamber and warmed to 30°C (Gu et al., 2014).
CR
Acute injury model
To induce acute injury, visual cortical slices were bisected through layers 1-4 using a 22
US
gauge scalpel blade and allowed to recover for one hour in the incubation buffer.
AN
Electrophysiology
Extracellular (EC) recording pipettes were made from borosilicate glass capillaries using Flaming/Brown model P-97 micropipette puller (Sutter Instruments CO.). The EC recording
M
electrodes were filled with 0.9% NaCl (1-2 MΩ). Electrodes were placed in layers 2-3 of visual
ED
cortical slices (Fig.1A). Multi-Clamp Commander (MCC) 700 amplifiers (Axon Instruments) were used for all electrical recordings. Data were low-pass filtered and digitized at 1 KHz for EC
PT
and 10 KHz for whole-cell recordings (Digidata; pCLAMP, Molecular Devices). CSD’s in individual pyramidal neurons of layers 2-3 of visual cortex were recorded using
CE
whole-cell patch clamp technique. Micropipettes (4-9MΩ) contained intracellular solution (in mM): 6 KCl, 0.5 EGTA, 20 HEPES, 10 phosphocreatine, 0.3 NaGTP, 2 NaCl, 4 MgATP;
AC
adjusted to pH 7.25 and osmolarity of 295 mOsm. Control ASCF was replaced with high KCl (26mM) ACSF and warmed to 36ºC (Fig.1B). After the initiation of CSD, high KCl solution was replaced with control ACSF. The pipette solution for voltage clamp Na+ current recording contained (mM) 5.8 NaCl, 134 CsCl, 1 MgCl2, 3 EGTA, and 10 HEPES (pH 7.3 adjusted with 9.2 mM NaOH) (Zhang et al., 2014). Slow ramp depolarizations (10-40 mV/1000ms) (Fig. 1C) from cell resting membrane potential were used to determine action potential threshold and cell spiking properties. Incremental hyperpolarizing and depolarizing current injections were used to study the passive
ACCEPTED MANUSCRIPT and active neuronal membrane properties (10 pA increments for 500-1000 ms) (Gu et al. 2014) (Fig.1D). Sodium currents were recorded using square-wave pulses in voltage clamp. To elicit fast voltage-gated sodium currents, cells were clamped at -80mV and depolarized to +30mV in 10mV increments, each 30ms in duration (Fig. 1E). Peak Na+ currents were measured before and in five minute intervals after TFB-TBOA exposure. We accounted for the effects of cell size on Na+
T
currents by calculating the Na+ current density (peak sodium current/ cell capacitance). To
IP
confirm that TFB-TBOA was affecting voltage gated Na+ channels, TTX was applied after 20
CR
minutes of TFB-TBOA incubation and Na+ currents were measured 5 minutes after TTX
US
exposure.
Imaging were
stained
with
voltage-sensitive
AN
Slices
naphthalenyl)ethenyl)-1-(3-sulfopropyl)pyridiniumhydroxide
dye
4-(2-(6-(Dibutylamino)-2-
(di-4-ANNEPS;
Sigma-Aldrich)
M
with the final concentration of 0.03 mg/ml. Dye was applied directly to the surface of slices and were incubated in no-light conditions at 30°C. Following the incubation period, individual slices 2-
l
in at
ED
were transferred to a recording chamber with oxygenated ACSF perfusing constantly at a rate of C. For a detailed description of the staining method, please see Hazra et al.,
PT
2013 and Gu et al., 2014. Upright wide-field epifluorescence microscope (BX51WI; Olympus) equipped with a fast CCD camera was used for recording and VSD signals (128×198 pixels,
CE
MiCam 02, SciMedia, USA). Imaging was done in conjunction with either whole-cell or EC recording of layer II-III of visual cortical slices. We determined locations of layers 2-3 based on
AC
the high density of pyramidal cells immediately below myelin dominant layer I. 150W halogen light source (Moritex Corp.) attached to a fiber optic cable was used to illuminate stained slices. Light was passed through a filter cube (excitation λ= 5
±1
n , dichroic λ=565 n
and
absorption λ>59 n , U-MWIG2, Olympus). Evoked optical signals were sampled at 250 Hz in a total of 90 second exposure time. We used brain vision analyzer (BV-Ana) software in analyzing the imaging data. We set the low pass filtration of DF/Fmax at 25% to increase the resolution of imaging for all samples. Moreover, we consistently set the signal threshold low, in order to detect small signals. To reduce the background noise, median filtering (3x3 pixels) was applied. Imaging and electrophysiological recording acquisitions were synchronized.
ACCEPTED MANUSCRIPT The CSD propagation velocity was measured by plotting a stripe from the start point to the end point. The software provides us the stripe length, its coordinates and the time duration of the propagation (Hazra et al, 2014). We refer to CSD wave propagation as laminar when the CSD propagated along the superficial layers (II-III) without expanding in to the deeper layers (IV-VI). When the signal clearly propagated between superficial and deep layers, we referred to it as
T
columnar.
IP
Data and Statistical analysis
CR
BrainVision (SciMedia), pClamp (v.10.1, Molecular Devices) and Graph pad Prism 5 software were used for analysis of optical and electrical data. Optical data were analyzed by using
US
the same setting for threshold and gain. All data are expressed in mean ± SEM. Unpaired student's t-tests with Welch's correction were used to determine significant differences between
AN
the two groups. For more than two group comparisons, one-way ANOVA followed by Newman-
AC
CE
PT
ED
M
Keuls were used. The value of p<0.05 was regarded as statistically significant.
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
Figure 1: Electrophysiological characteristics of CSD and firing properties of visual cortical
ED
neurons. (A) CSD in individual pyramidal neurons and neuronal network in layers 2-3 of visual cortex were recorded using whole-cell (WC) patch clamp and extracellular (EC) recording
PT
techniques. CSD in network was recorded by placing the borosilicate micropipettes 500 micrometers away from the whole-cell electrode in the layers 2-3. CSD in network and individual
CE
neurons was recorded simultaneously. (B) CSD was evoked by high KCl and identified based on a significant depolarization in both network and single cell. Typically, CSD in individual cells
AC
started roughly 2-10 seconds before CSD detected by the EC electrodes. CSD usually accompanied with spikes in the whole-cell recording. The black line indicates the duration of high K+ solution application. (C) Evoked APs in a representative neuron induced by depolarizing current ramp injections from -40pA to +60pA in 10pA increments into the soma for one second. AP threshold is measured as the beginning of the rapid depolarization of membrane potential following depolarizing current ramp injections. (D) AP discharge evoked in the rat neuron by square wave current pulses into the soma for one second. Square wave current injections ranging from -20pA to +80pA in 10pA increments were used. AP amplitude was measured as the voltage difference between the threshold and the peak value of the membrane potential during the AP. (E)
ACCEPTED MANUSCRIPT Sodium currents were recorded using square-wave pulses in voltage clamp. To elicit fast voltagegated sodium currents, cells were clamped at -80mV, then depolarized to +30mV in 10mV increments, each 30ms in duration.
RESULTS
T
Effect of TFB-TBOA on cell properties
IP
To assess effects of glutamate transporter blockade with TFB-TBOA on neuronal properties,
CR
we first quantified passive and active membrane properties in the cortical layer II-III pyramidal
AC
CE
PT
ED
M
AN
US
cells.
Figure 2: Effect of TFB-TBOA on neuronal membrane properties and excitability. AP’s were evoked by current pulses and studied under the current-clamp conditions in 9 neocortical slices. (A) TFB-TBOA (50nM) increased the cell membrane resistance after 10 minutes (p<0.05). (B) Current pulses evoked fewer AP’s in slices pretreated with 50nM TFB-TBOA (p<0.05, 0100pA current injection). (C) After 20min treatment with 50nM TFB-TBOA, AP threshold increased (p<0.05). (D) Pretreatment of visual cortical slices with TFB-TBOA decreased AP
ACCEPTED MANUSCRIPT amplitude (p<0.001). (E) TFB-TBOA significantly reduced fast voltage-gated sodium currents within 15 minutes (p<0.01). TTX co pletely blocked the channels’ activity within 5
inutes of
incubation. CSD in visual cortical slices To study CSD in visual cortex slices, we used a combination of electrophysiology and VSDI. Emergent CSDs were induced with high concentration of KCl in bath (Fig. 1B). EC and
T
intracellular (IC) electrodes were placed in the layer 2-3 of visual cortex about 500 micrometers
IP
apart. In high K+ ACSF, CSD typically occurred 30-45 second after high KCl application. CSD in
CR
the single cells started with a rapid depolarization followed by several spikes and a slow return to the resting membrane potential. The majority of cell spiking occurred before maximum
US
depolarization was reached. CSD in the EC recording was typically noticeable a few seconds
AC
CE
PT
ED
M
AN
later. CSD in single cells typically lasted for 30-180 seconds.
Figure 3: Electrophysiological properties of CSD in TBI and TFB-TBOA treated groups. (A) The network and single cell properties of CSD in control, TFB-TBOA, and TBI treated groups. (B) TFB-TBOA increased CSD latency in single neurons (n=9, p<0.001). (C) Both TBI and TFB-TBOA (n=9) prolonged CSD duration compare with the control group (p<0.001). (D&
ACCEPTED MANUSCRIPT E) Both, depolarizing and hyperpolarizing slope of CSD decreased in TBI or TFB-TBOA treated group in the individual neurons (p<0.05, n =9). (F) Network CSD amplitude in the presence of TFB-TBOA or TBI (n=9). Both, TFB-TBOA and TBI decreased the mean of amplitude compared to the control (p<0.001). (G) In network recordings, TFB-TBOA prolonged CSD in comparison to the control group (p<0.01), but the CSD duration differences between the TBI and control were not statistically significant. (H) The network frequency of CSD incidence in the
T
presence of TFB-TBOA or TBI. TFB-TBOA decreases the frequency of CSD incidence
IP
compared to both TBI (p<0.05) and control (p<0.01) groups. The differences between the TBI
CR
and control group was not statistically significant
Effects of TBI model and TFB-TBOA on CSD in individual neurons and neuronal network
US
Both TBI and TFB-TBOA altered CSD characteristics in both network and single cell patchclamp recording (Fig. 3A). We quantified spontaneous action potentials before, during, and
AN
immediately after the initial CSD event. TBI reduced the number of action potentials to 24.50 ± 6.98 (control 54.71± 10.91) (p=0.046), while TFB-TBOA completely blocked the action
M
potentials. Additionally, resting membrane potential was unaffected under either condition. CSD latency was measured based on the time that it took to form the first CSD after the application of
ED
high KCl solution. Surprisingly, our results showed that TFB-TBOA increased the latency to the first CSD (p<0.0001) (Fig. 3B). CSD duration was defined and measured as the time from the
PT
initiation of rapid depolarization in individual neurons and network to the time when the membrane potential repolarized to its pre-CSD potential value. Either incubation of coronal slices
CE
with TFB-TBOA for 20 minutes or acute cut caused prolonged CSD (Fig. 3C). In both conditions, CSD depolarization and repolarization slopes were shallower (Fig. 3D & E).
CSD wave
AC
depolarized with the rate of 5.12± 0.81 (mV/ms) in control slices, while it decreased to 2.23± 0.41 (mV/ms) and 2.06± 0.42 (mV/ms) in the TBI and TFB-TBOA, respectively. The same trend was observed in the repolarizing slope. The rate at which the CSD wave typically repolarized was 1.71± 0.12 (mV/ms) in control slices, 0.81± 0.08 (mV/ms) in cut slices and 0.82± 0.11 (mV/ms) in TFB-TBOA groups. We next measured the CSD amplitude, duration, and inter-CSD interval obtained from EC recordings of the local neuronal network. CSD amplitude significantly decreased in the presence of TFB-TBOA or acute injury (Fig. 3F). The duration of network CSD was significantly increased in response to TFB-TBOA; whereas, acute injury increased duration, but not
ACCEPTED MANUSCRIPT significantly (Fig. 3G). In order to quantify how TFB-TBOA or acute injury affects network CSD frequency, we measured the duration of the time interval between subsequent CSDs and found that TFB-TBOA prolonged inter-CSD interval in both, the network and single cell recordings (Fig. 3H). Our data showed that although both TFB-TBOA and TBI increased the CSD occurrence rate; the inter-CSD interval was significantly shorter in the presence of TFB-TBOA (Fig. 3H).
ED
M
AN
US
CR
IP
T
Spatiotemporal patterns of CSD and their modulation by TBI and TFB-TBOA:
PT
Figure 4: Spatiotemporal properties of CSD in control, TBI model and TFB-TBOA treated rat visual cortical slices. (A) Initiation and propagation of CSD in the control slices. CSD
CE
initiated in layers 2-3 of visual cortex, and spread in a laminar pattern. In addition, CSD also had columnar pattern with propagation to the adjacent columns (n=7). (B) The spatiotemporal
AC
properties of CSD in TBI model. CSD does not initiate at the site of the injury. CSD initiated close to the injured site in layers 2-3 and propagated toward the deep layers and along the actual lesion (n=7). (C) CSD spatiotemporal properties in slices pretreated for 20 minutes with 50nM TFB-TBOA. CSD initiated in the deeper layers of cortical slices. CSD formed predominant columnar patterns. It did not have clear propagation and, instead, while remaining stationary, gradually increased in signal strength (n=7). (D) CSD propagation velocity in the control and TBI model. CSD propagated significantly (p<0.05) slower in the TBI model than the control slices. (E) The initiation point of CSD in control, TBI model and TFB-TBOA pretreated slices. Although
ACCEPTED MANUSCRIPT all CSDs in control and TBI model initiated in the superficial layers, TFB-TBOA changed the initiation point of CSD to layer 4-5 of cortical slices. To observe whether TFB-TBOA or the acute cut changes spatiotemporal patterns of CSD, fast voltage-sensitive dye imaging techniques were used simultaneously with electrophysiological recordings. Concurrent imaging with EC and IC recordings in cortical layers II-III allowed for the clear identification of specific CSD depolarizing wave pattern characteristics. When the CSD
T
origin was quantified using threshold analysis, it was observed that CSD in the intact cortical
IP
slices originated from confined spatial areas in layers II-III. The waves then propagated mostly in
CR
a laminar manner (7 out of 10) and sometimes in a curve-like shape that extended into the columnar cortical structure, spanning from layer II to V-VI (3 out of 10) (Fig. 4A).
US
Based on previous reports, we hypothesized that the acute cut would likely be the source of CSD in the injured coronal slices; however, we found that CSD initiated nearby the area of the
AN
injury in layers II-III, but then propagated toward the cut in the slice in either curved columnar (5 out of 7) or laminar (2 out of 7) waves (Fig. 4B). On the other hand, in all TFB-TBOA treated
M
slices (intact and cut) CSD initiated in layers II-V, then formed a semicircular shape and a columnar pattern including both layers II-III and IV-V of the visual cortex (Fig. 4C).
ED
Finally, acute cut reduced the CSD propagation rate significantly (Fig. 4D). TFB-TBOA seemed to have trapped CSD in the layers where it typically initiates, causing CSD depolarization
PT
to increase in strength locally and expanded from its origin without showing clear layer or column-specific propagation across the tissue (Fig. 4D). CSD initiated from layers II-III in all
CE
slices of the control and TBI groups. In TFB-TBOA, however, CSD initiated in layers II-III and IV-V (Fig. 4E). Thus, global glutamate transport blockade has shifted the site of CSD initiation
AC
and the localized cut acted as an attractor of the depolarizing wave. This shift of the active signal origins into the deep layers suggests a layer-specific regulation of glutamate and spatiotemporal properties of CSD in distinct traumatic conditions.
DISCUSSION Slowly propagating (2-5 mm/min) CSD is thought to initiate in layers II-III of the neocortex, accompanied by spreading depolarization and neuronal and glial swelling (Basarsky et al., 1999; Kimelberg et al., 1990), and depression of activity (Lauritzen et al., 2011; Larrosa et al., 2006). During CSD, inward cationic Ca+2 and Na+ currents exceed the outward K+ current. This
ACCEPTED MANUSCRIPT pathological cation influx causes significant cell depolarization and glutamate release (Van and Fifkova, 1970; Fabricius et al., 1993; Somjen, 2001). Increased glutamate leads to increased activation of glutamate receptors, especially NMDARs, and further CSD propagation (Kramer, 2016; Lauritzen, 2011; Torrente et al., 2014 (a); Eikermann-Haerter and Ayata , 2010). Astrocytic glutamate transporters play the decisive role in clearance of glutamate and thus, CSD properties. Glutamate transporters, GLT-1 and GLAST, are prominently expressed in
T
astrocytes and are responsible for more than 90% of extracellular glutamate re-uptake (Sheldon
IP
and Robinson., 2007). GLT-1 also has a significant effect on the extracellular glutamate buffering
CR
during CSD (Torrente et al., 2014 (a)). We showed that TFB-TBOA, a selective glutamate transporter blocker, reduced cell excitability and CSD amplitude. Moreover, the blocker delayed
US
the initiation of CSD, shifted its spatial formation to deeper cortical layers, and reduced its propagation velocity. These findings suggest that pharmacological manipulation of glutamate
AN
transporters could be considered as a potential strategy for modulating CSDs associated with
M
neurological conditions, such as migraines, TBI, and stroke.
Effects of TFB-TBOA on single cell and network activity
ED
TFB-TBOA had significant effects on single cell and cortical network excitability. TFBTBOA increased cell membrane resistance, and reduced sodium currents. It also decreased
PT
neuronal excitability by increasing the threshold and reducing amplitude of evoked action potentials. In addition, TFB-TBOA prolonged CSD while decreasing CSD amplitude and cell
CE
spiking activity. Prolonged CSDs were also reported using 0.5 and 1mM of DL-TBOA, which was suggested to be due to activation of pre-synaptic NMDA receptors (Hinzman et al., 2015).
AC
Interestingly, Bozzo et al. showed that TFB-TBOA inactivates Na+ channels in IC50 43nM (Bozzo et al., 2010). In our study, TFB-TBOA had significant effects on neuronal excitability and reduced Na+ influx, suggesting that the effect of TFB-TBOA on neuronal activity is probably due to its direct effects on voltage-gated sodium channels, in addition to the blockade of glutamate transport in astrocytes. This dysregulation of sodium influx and action potentials in neurons by TFB-TBOA is likely contributing to the spread of CSD through mechanisms other than the chemical neurotransmission. Although glutamate transporters are thought to facilitate glutamate uptake into the astrocytes, recent studies show that high extracellular K+ or the presence of high levels of intracellular
ACCEPTED MANUSCRIPT glutamate causes these transporters to work in reverse (Zhang, 2007; Szatkowski et al., 1990). Thus, in the presence of high K+ and hyperexcitability, like CSD, reversal of glutamate transport in astrocytes could become an additional unwanted source of extracellular glutamate build-up, causing large and prolonged depolarizations. The blockade of GLT transporters in our experimental conditions may have prevented glutamate transport reversal and caused glutamate to be trapped inside the glia. In this way, blockade of glutamate transport may counter-intuitively
T
decrease pathological glutamate reversal and extracellular glutamate levels, partly accounting for
IP
the reduced depolarization in the presence of TFB-TBOA, as well as decreased cell excitability.
CR
Under different experimental conditions, however, it was also observed that a different glutamate transporter blocker (D,L-threo-beta-hydroxyaspartate (THA)) increased the amount of
US
depolarization in the presence of high K+ (Larrosa et al., 2006).
AN
Effects of TFB-TBOA on CSD spatial formation and propagation VSDI experiments show that in compromised glutamate transport conditions, CSD dominates
M
the deep cortical layers. In most instances, CSD was confined and, when it spread, it did so slowly, indicating that TFB-TBOA compromises CSD propagation. One explanation regarding
ED
the effects of TFB-TBOA on CSD propagation could be attributed to glutamate transport. Glutamate, in addition to control of CSD amplitude, is also known to modulate CSD initiation and
PT
propagation. For example, when NMDA receptors are blocked, CSD is reduced in size and area (Marrannes et al. 1988; Tozzi et al. 2012). Blocking AMPA, NMDA, and kainate glutamate
CE
receptors in the rat visual cortex slices using kyuneric acid traps CSD in the deeper coronal layers and prevents its propagation (Világi et al. 2001). Kyuneric acid in this case also compromises
AC
synaptic transmission. Not surprisingly, multiple reports indicate that during CSD, synaptic transmission is impaired (Somjen, 2004). Another explanation for the effect of TFB-TBOA on CSD propagation would be via its effect on the TTX-sensitive sodium channels. Tozzi et al. showed that blockade of the TTX-sensitive sodium channels in the neocortical slices reduced the CSD area (Tozzi et al., 2012). The CSD propagation velocity in the current study is 3±0.62 mm/min in control slices, which is similar to the reported velocity by Torrente (Torrente et al. 2014 (b)). CSD seemed to have been trapped by TFB-TBOA, and it is likely that this effect on propagation was via direct control of sodium channels and compromised chemical neurotransmission.
ACCEPTED MANUSCRIPT
Changes of CSD properties in the TBI model TBI is often accompanied by CSD and increases in extracellular glutamate (Lauritzen et al., 2011; Rao, 1998). In addition, formation of reactive astrocytes increases following TBI (Myer et al. 2006) while GLT-1 expression is reduced following TBI and stroke (Rao et al. 1998). Here, following acute cut in the cortex, CSD was prolonged, and spiking activity and velocity of CSD
T
propagation were decreased (1.33± 0.3 mm/min). In addition, our imaging data showed that the
IP
acute injury changed CSD spatiotemporal characteristics. Surprisingly, CSD did not initiate at the
CR
site of the actual injury, but rather in the surrounding peri-injury zone. It seemed that the networks at the cut itself were incapable of initiating the CSD. This is likely due to the physical trauma and
US
compromised neural connections.
Using a freeze lesion model, Dulla et al. (2013) found that TBI alters the input-output
AN
response and decreases the evoked network responses. In addition, the freeze lesion caused glutamate signal to shift to the deeper cortical layers (V), away from the lesion zone. This also
M
suggests that the networks in the injured area itself, similarly to our results, may not be capable of initiating CSD. Instead, the injury zone seems to attract and have the ability to sustain CSDs
ED
initiated by nearby, less compromised networks.
PT
CONCLUSION
CE
The results demonstrate that both TBI and astrocytic glutamate transporters modulate neuronal excitability in addition to the electrophysiological and spatiotemporal properties of CSD.
AC
The cortical injury site which is not capable of initiating CSD, instead, acts as an attractor to the depolarizing CSD waves. These results provide new insights about the mechanisms of CSD, and support the idea that glutamate transporters may become viable drug targets for control and/or modulation of the spreading depression.
ACCEPTED MANUSCRIPT References Basarsky, T. A., Feighan, D., & MacVicar, B. A. (1999). Glutamate release through volumeactivated channels during spreading depression. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 19(15), 6439-6445.
T
Bozzo, L., & Chatton, J. Y. (2010). Inhibitory effects of (2S, 3S)-3-[3-[4-
IP
(trifluoromethyl)benzoylamino]benzyloxy]aspartate (TFB-TBOA) on the astrocytic sodium responses to glutamate. Brain Research, 1316, 27-34. doi:10.1016/j.brainres.2009.12.028
CR
[doi]
US
Bures, J. (1957). The ontogenetic development of steady potential differences in the cerebral cortex in animals. Electroencephalography and Clinical Neurophysiology, 9(1), 121-130.
AN
Chuquet, J., Hollender, L., & Nimchinsky, E. A. (2007). High-resolution in vivo imaging of the neurovascular unit during spreading depression. The Journal of Neuroscience : The Official
M
Journal of the Society for Neuroscience, 27(15), 4036-4044. doi:27/15/4036 [pii]
ED
de Luca, B., & Bures, J. (1977). Development of cortical spreading depression and of its transition to the caudate nucleus in rats. Developmental Psychobiology, 10(4), 289-297.
PT
doi:10.1002/dev.420100402 [doi]
CE
Dulla, C. G., Tani, H., Brill, J., Reimer, R. J., & Huguenard, J. R. (2013). Glutamate biosensor imaging reveals dysregulation of glutamatergic pathways in a model of developmental
[doi]
AC
cortical malformation. Neurobiology of Disease, 49, 232-246. doi:10.1016/j.nbd.2012.09.001
Eikermann-Haerter, K., & Ayata, C. (2010). Cortical spreading depression and migraine. Current Neurology and Neuroscience Reports, 10(3), 167-173. doi:10.1007/s11910-010-0099-1 [doi] Eiselt, M., Giessler, F., Platzek, D., Haueisen, J., Zwiener, U., & Rother, J. (2004). Inhomogeneous propagation of cortical spreading depression-detection by electro- and magnetoencephalography in rats. Brain Research, 1028(1), 83-91. doi:S00068993(04)01482-9 [pii]
ACCEPTED MANUSCRIPT Fabricius, M., Fuhr, S., Bhatia, R., Boutelle, M., Hashemi, P., Strong, A. J., & Lauritzen, M. (2006). Cortical spreading depression and peri-infarct depolarization in acutely injured human cerebral cortex. Brain : A Journal of Neurology, 129(Pt 3), 778-790. doi:awh716 [pii] Fabricius, M., Fuhr, S., Willumsen, L., Dreier, J. P., Bhatia, R., Boutelle, M. G., . . . Lauritzen, M. (2008). Association of seizures with cortical spreading depression and peri-infarct
T
depolarisations in the acutely injured human brain. Clinical Neurophysiology : Official
IP
Journal of the International Federation of Clinical Neurophysiology, 119(9), 1973-1984.
CR
doi:10.1016/j.clinph.2008.05.025 [doi]
Fabricius, M., Jensen, L. H., & Lauritzen, M. (1993). Microdialysis of interstitial amino acids
US
during spreading depression and anoxic depolarization in rat neocortex. Brain
AN
Research, 612(1-2), 61-69. doi:0006-8993(93)91644-8 [pii] Goodrich, G. S., Kabakov, A. Y., Hameed, M. Q., Dhamne, S. C., Rosenberg, P. A., &
M
Rotenberg, A. (2013). Ceftriaxone treatment after traumatic brain injury restores expression of the glutamate transporter, GLT-1, reduces regional gliosis, and reduces post-traumatic
ED
seizures in the rat. Journal of Neurotrauma, 30(16), 1434-1441. doi:10.1089/neu.2012.2712 [doi]
PT
Gorji, A. (2001). Spreading depression: A review of the clinical relevance. Brain Research.Brain
CE
Research Reviews, 38(1-2), 33-60. doi:S0165017301000819 [pii] Gu, F., Hazra, A., Aulakh, A., & Ziburkus, J. (2014). Purinergic control of hippocampal circuit
AC
hyperexcitability in dravet syndrome. Epilepsia, 55(2), 245-255. doi:10.1111/epi.12487 [doi] Hazra, A., Gu, F., Aulakh, A., Berridge, C., Eriksen, J. L., & Ziburkus, J. (2013). Inhibitory neuron and hippocampal circuit dysfunction in an aged mouse model of alzheimer's disease. PloS One, 8(5), e64318. doi:10.1371/journal.pone.0064318 [doi] Hazra, A., Macolino, C., Elliott, M. B., & Chin, J. (2014). Delayed thalamic astrocytosis and disrupted sleep-wake patterns in a preclinical model of traumatic brain injury. Journal of Neuroscience Research, 92(11), 1434-1445. doi:10.1002/jnr.23430 [doi]
ACCEPTED MANUSCRIPT Hinzman, J. M., DiNapoli, V. A., Mahoney, E. J., Gerhardt, G. A., & Hartings, J. A. (2015). Spreading depolarizations mediate excitotoxicity in the development of acute cortical lesions. Experimental Neurology, 267, 243-253. doi:10.1016/j.expneurol.2015.03.014 [doi] Hossmann, K. A. (1996). Periinfarct depolarizations. Cerebrovascular and Brain Metabolism
T
Reviews, 8(3), 195-208.
IP
Kimelberg, H. K., Goderie, S. K., Higman, S., Pang, S., & Waniewski, R. A. (1990). Swellinginduced release of glutamate, aspartate, and taurine from astrocyte cultures. The Journal of
CR
Neuroscience : The Official Journal of the Society for Neuroscience, 10(5), 1583-1591.
US
Kramer, D. R., Fujii, T., Ohiorhenuan, I., & Liu, C. Y. (2016). Cortical spreading depolarization: Pathophysiology, implications, and future directions. Journal of Clinical Neuroscience :
doi:10.1016/j.jocn.2015.08.004 [doi]
AN
Official Journal of the Neurosurgical Society of Australasia, 24, 22-27.
M
Larrosa, B., Pastor, J., Lopez-Aguado, L., & Herreras, O. (2006). A role for glutamate and glia in
ED
the fast network oscillations preceding spreading depression. Neuroscience, 141(2), 10571068. doi:S0306-4522(06)00488-X [pii]
PT
Lauritzen, M., Dreier, J. P., Fabricius, M., Hartings, J. A., Graf, R., & Strong, A. J. (2011). Clinical relevance of cortical spreading depression in neurological disorders: Migraine,
CE
malignant stroke, subarachnoid and intracranial hemorrhage, and traumatic brain injury. Journal of Cerebral Blood Flow and Metabolism : Official Journal of the
AC
International Society of Cerebral Blood Flow and Metabolism, 31(1), 17-35. doi:10.1038/jcbfm.2010.191 [doi] Malkov, A., Ivanov, A. I., Popova, I., Mukhtarov, M., Gubkina, O., Waseem, T., . . . Zilberter, Y. (2014). Reactive oxygen species initiate a metabolic collapse in hippocampal slices: Potential trigger of cortical spreading depression. Journal of Cerebral Blood Flow and Metabolism : Official Journal of the International Society of Cerebral Blood Flow and Metabolism, 34(9), 1540-1549. doi:10.1038/jcbfm.2014.121 [doi]
ACCEPTED MANUSCRIPT Marrannes, R., Willems, R., De Prins, E., & Wauquier, A. (1988). Evidence for a role of the Nmethyl-D-aspartate (NMDA) receptor in cortical spreading depression in the rat. Brain Research, 457(2), 226-240. doi:0006-8993(88)90690-7 [pii] Myer, D. J., Gurkoff, G. G., Lee, S. M., Hovda, D. A., & Sofroniew, M. V. (2006). Essential protective roles of reactive astrocytes in traumatic brain injury. Brain : A Journal of
IP
T
Neurology, 129(Pt 10), 2761-2772. doi:awl165 [pii]
Obrenovitch, T. P. (1995). The ischaemic penumbra: Twenty years on. Cerebrovascular and
CR
Brain Metabolism Reviews, 7(4), 297-323.
US
Obrenovitch, T. P., & Zilkha, E. (1996). Inhibition of cortical spreading depression by L-701,324, a novel antagonist at the glycine site of the N-methyl-D-aspartate receptor complex. British
AN
Journal of Pharmacology, 117(5), 931-937.
Rao, V. L., Baskaya, M. K., Dogan, A., Rothstein, J. D., & Dempsey, R. J. (1998). Traumatic
M
brain injury down-regulates glial glutamate transporter (GLT-1 and GLAST) proteins in rat
ED
brain. Journal of Neurochemistry, 70(5), 2020-2027. Rogatsky, G., Mayevsky, A., Zarchin, N., & Doron, A. (1996). Continuous multiparametric
PT
monitoring of brain activities following fluid-percussion injury in rats: Preliminary
CE
results. Journal of Basic and Clinical Physiology and Pharmacology, 7(1), 23-43. Rothstein, J. D., Dykes-Hoberg, M., Pardo, C. A., Bristol, L. A., Jin, L., Kuncl, R. W., Welty, D.
AC
F. (1996). Knockout of glutamate transporters reveals a major role for astroglial transport in excitotoxicity and clearance of glutamate. Neuron, 16(3), 675-686. doi:S08966273(00)80086-0 [pii] Sheldon, A. L., & Robinson, M. B. (2007). The role of glutamate transporters in neurodegenerative diseases and potential opportunities for intervention. Neurochemistry International, 51(6-7), 333-355. doi:S0197-0186(07)00077-0 [pii] Shimamoto, K., Sakai, R., Takaoka, K., Yumoto, N., Nakajima, T., Amara, S. G., & Shigeri, Y. (2004). Characterization of novel L-threo-beta-benzyloxyaspartate derivatives, potent
ACCEPTED MANUSCRIPT blockers of the glutamate transporters. Molecular Pharmacology, 65(4), 1008-1015. doi:10.1124/mol.65.4.1008 [doi] Somjen, G. G. (2001). Mechanisms of spreading depression and hypoxic spreading depressionlike depolarization. Physiological Reviews, 81(3), 1065-1096.
T
Somjen, G. G. (2004). Ions in the brain: Normal function, seizures, and stroke. Oxford ; New
IP
York: Oxford University Press.
CR
Strong, A. J., Fabricius, M., Boutelle, M. G., Hibbins, S. J., Hopwood, S. E., Jones, R., . . . Lauritzen, M. (2002). Spreading and synchronous depressions of cortical activity in acutely
US
injured human brain. Stroke, 33(12), 2738-2743.
Szatkowski, M., Barbour, B., & Attwell, D. (1990). Non-vesicular release of glutamate from glial
AN
cells by reversed electrogenic glutamate uptake. Nature, 348(6300), 443-446.
M
doi:10.1038/348443a0 [doi]
Torrente, D., Cabezas, R., Avila, M. F., Garcia-Segura, L. M., Barreto, G. E., & Guedes, R. C.
ED
(2014a). Cortical spreading depression in traumatic brain injuries: Is there a role for
PT
astrocytes? Neuroscience Letters, 565, 2-6. doi:10.1016/j.neulet.2013.12.058 [doi] Torrente, D., Mendes-da-Silva, R. F., Lopes, A. A., Gonzalez, J., Barreto, G. E., & Guedes, R. C.
CE
(2014b). Increased calcium influx triggers and accelerates cortical spreading depression in vivo in male adult rats. Neuroscience Letters, 558, 87-90. doi:10.1016/j.neulet.2013.11.004
AC
[doi]
Tozzi, A., de Iure, A., Di Filippo, M., Costa, C., Caproni, S., Pisani, A., . . . Calabresi, P. (2012). Critical role of calcitonin gene-related peptide receptors in cortical spreading depression. Proceedings of the National Academy of Sciences of the United States of America, 109(46), 18985-18990. doi:10.1073/pnas.1215435109 [doi] Van Harreveld, A., & Fifkova, E. (1970). Glutamate release from the retina during spreading depression. Journal of Neurobiology, 2(1), 13-29. doi:10.1002/neu.480020103 [doi]
ACCEPTED MANUSCRIPT Vilagi, I., Klapka, N., & Luhmann, H. J. (2001). Optical recording of spreading depression in rat neocortical slices. Brain Research, 898(2), 288-296. doi:S0006-8993(01)02196-5 [pii] Zhang, Y., He, J. C., Liu, X. K., Zhang, Y., Wang, Y., & Yu, T. (2014). Assessment of the effect of etomidate on voltage-gated sodium channels and action potentials in rat primary sensory cortex pyramidal neurons. European Journal of Pharmacology, 736, 55-62.
IP
T
doi:10.1016/j.ejphar.2014.04.036 [doi]
Zhang, Z., Tao, Z., Gameiro, A., Barcelona, S., Braams, S., Rauen, T., & Grewer, C. (2007).
CR
Transport direction determines the kinetics of substrate transport by the glutamate transporter EAAC1. Proceedings of the National Academy of Sciences of the United States of
US
America, 104(46), 18025-18030. doi:0704570104 [pii]
AN
Zhou, N., Rungta, R. L., Malik, A., Han, H., Wu, D. C., & MacVicar, B. A. (2013). Regenerative glutamate release by presynaptic NMDA receptors contributes to spreading
M
depression. Journal of Cerebral Blood Flow and Metabolism : Official Journal of the International Society of Cerebral Blood Flow and Metabolism, 33(10), 1582-1594.
AC
CE
PT
ED
doi:10.1038/jcbfm.2013.113 [doi]
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
Graphical abstract
ACCEPTED MANUSCRIPT
Highlights
IP
sodium currents, network hyperexcitability, and CSD characteristics.
T
Glutamate transporter blocker TFB-TBOA has significant effects on neuronal voltage-gated
CR
Acute cortical cut and compromised glutamate transporter modulate physiological and
US
spatiotemporal properties of CSD.
AN
The area of the actual cortical cut does not initiate CSDs, but in the surrounding areas. The
M
actual injury site attracts propagating CSD depolarizing waves.
AC
CE
PT
ED
TFB-TBO significantly affects sodium conductances in neurons.