Characterization of optogenetically-induced cortical spreading depression in awake mice using graphene micro-transistor arrays

Animal experiments were conducted in accordance with the United Kingdom Animal (Scientific Procedures) Act 1986, and approved by the Home Office (license PPL70-13691). C57Bl/mice (38 males and 6 females) were bred and the majority (n = 33) were used at the ages of ∼3–6 months (140.6 ± 4.65 d), while a subset (n = 11) were used at an age of ∼7–9 months (250.0 ± 7.259 d), due to the closure of the lab resulting from the Coronavirus pandemic (p = <0.001, ***; unpaired T-Test). Animals were housed on a 12 h/12 h dark/light cycle, and food and water were given ad libitum. Animals were group housed until their headbar attachment surgery and were then subsequently individually housed. For all procedures, animals were weighed before being anaesthetized in a chamber using isoflurane (100% w/w; 988-3245, Henry Schein, USA) and placed in a stereotaxic frame (David Kopf Instruments Ltd, USA). Viscotears (Bausch + Lomb, USA) were applied to each eye and both pain relief, consisting of Buprenorphine (0.5 mg kg⁻¹; subcutaneous; Ceva, France) and Metacam (15 mg kg⁻¹; subcutaneous; Boehringer Ingelheim, Germany) and ∼0.5 ml of saline, and dexamethasone (1 mg kg⁻¹; subcutaneous; MSD Animal Health, USA) were given. All surgical procedures were performed using sterilized tools and aseptic techniques. At the end of the surgical procedure, warmed saline (0.15 ml/10 g) was administered subcutaneously. Mice were checked daily and recovered for at least a week before experiments commenced, with the exception of the craniotomy surgery, where recording occurred on the same day, ∼2–4 h post-surgery.

The skin on top of the skull was cut away. The skull was cleaned using a delicate bone scraper (10075-16; Fine Science Tools, Germany) and brief topical application of hydrogen peroxide (6%). Small bur-holes were drilled in the following two positions from bregma: (a) antero-posterior: +2.2 mm; lateral: +1.0 mm and (b) antero-posterior: +0.75 mm; lateral: +1.6 mm. To express channelrodopsin 2 (ChR2), high titer (1 × 10¹³ vg ml⁻¹) AAV9-hSyn-ChR2-EYFP virus (AAV9_hSyn_hCHR2(H134R)-EYFP; 26973, Addgene, USA) was diluted 1:1 in 0.9% NaCl and delivered using a 5 µl syringe (Model 95 Hamilton Company, Switzerland) with a 35-gauge needle. At both sites, the needle tip was lowered from pia mater until 500 µm deep. Using a microinjection pump (WPI Ltd, USA), 500 nl was injected at 100 nl min⁻¹ to achieve expression in both layer 5 and layer 2/3 pyramidal neurons. To apply the headbars for the Neurotar system a small hole in the skull over the left visual cortex was drilled for a metal support screw (00-96X3-32, Plastics One, USA). Using vetbond (1469SB, 3M, USA), the headplate (Model 9, Neurotar, Finland) was firmly attached and strengthened using dental cement before Kwik-cast (W.P.I., UK) covered the exposed skull. Mice were checked daily to ensure recovery. Habituation was performed by placing the mouse in the Neurotar mobile-home cage for increasing periods of time (15–60 min) over several days the week prior to craniotomy surgery.

Due to the lab closure as a result of Covid19 the amount of time cortical neurons expressed ChR2 prior to optogenetic experiments varied. Pre-Covid, the average length of time from viral injection to experiment was 59.61 ± 3.142 d, n = 33 animals, and post-Covid this was extended to 175.9 ± 2.959 d, n = 11 animals (p = <0.001, ***; unpaired T-Test). There was no difference in the intensity of light required to induce a CSD (threshold) between the two groups, 34.26 ± 7.971 mJ, n = 33 animals pre-Covid verses 33.15 ± 15.76 mJ, n = 11 animals post-Covid, (p = 0.255; Mann–Whitney unpaired T-Test).

On the day of recording, craniotomies were performed. Three areas were exposed: (a) a large (2 × 2 mm) craniotomy over somatosensory and visual cortex on the right hand side, (b) the skull was thinned or removed over the MC ipsilateral to the main craniotomy, and (c) a small drill hole over the MC on the left-hand side (contralateral to main craniotomy) (figure 1(B)). After completion, exposed dura was covered with cortex buffered saline, sterilized sylgard 184 (∼200 μm thickness), and a kwik-cast layer. After ∼2–4 h post-recovery, the animal was moved to the Neurotar frame and the craniotomies were exposed by removal of the kwickcast and sylgard.

Zoom In Zoom Out Reset image size

Electrophysiological recordings were performed using flexible gSGFETs arrays. The transistor arrays were carefully connected to a PCB and lowered onto the dura or skull using a micromanipulator. Whereas most currently available electrodes are passive, gSGFETs are active devices that transduce local voltage changes to current and permit a wide recording bandwidth (Masvidal-codina et al 2019). A custom g.HIamp biosignal amplifier, (g.RAPHENE, g.tec medical engineering GmbH Austria) was used for signal acquisition at 9600 kHz and 24 Bit. The system enables simultaneous recording in two frequency bands with different gains preventing amplifier saturation. Prior to recording, I_ds − V_gs curves were obtained at the start and end of each experiment to allow determination of the performance of the gSGFET and custom code was used to calculate the optimal bias point (average across the arrays of the maximum of the mean absolute transconductance left from charge neutrality point). Subsequent interpolation of acquired current signals into the transfer curve results in DC-coupled voltage signals.

A reference wire (Ag/AgCl₂) was placed in the contralateral MC. An LED cannula (400 µm diameter, CFM14L20, Thor Labs, NJ, USA) was lowered via a manipulator over right-hand side MC. For some experiments (n = 8), two transistor arrays were used (section 2.9, figures 8 and 9) and larger craniotomies were performed as required for the protocol.

Prior to experimentation the LED current-to-power relationship was calibrated using a power meter (PM16-130, Thor Labs, NJ, USA), and mW values were multiplied by the duration (seconds) of stimulation to obtain a measurement of light power in mJ. After 10 min of baseline recording, the light-threshold required to induce a CSD was determined using continuous 5 or 10 s stimulations of blue light (470 nm; M470F3, Thors Labs, NJ, USA). Controls were performed using the same power intensity and duration with a green LED (595 nm; M595F2, Thors Labs, NJ, USA) or by using pulse blue light (20 Hz, 50% duty cycle) with a combined light-on duration identical to the square pulsed light duration (5 or 10 s), (figure 1(G)).

63 mice were prepared for this study. 44 mice (38 males, 6 females) showed reproducible induction of CSD from light stimulation and were included in this manuscript. The remaining 19 mice were excluded for the following reasons: eight animals because the initial light-evoked CSD was followed by at least one spontaneous CSD. Post-mortem analysis of these animals revealed lesions through subcortical (motor) white matter tracts, indicative of accidental damage during surgery (data not shown). Four animals were excluded as it was not possible to evoke a light induced CSD and three animals were excluded due to an inability to trigger a repeatable CSD every trial. Post hoc analysis of these brains revealed a marked reduction in transduction volume ⩽1 mm³ (figure 2). No data was obtained from four mice due to hardware or software problems during experimentation.

Zoom In Zoom Out Reset image size

CSD was induced in the MC by the following three methods: (a) optogenetic (b) pinprick of a 35 gauge Nanofil injection needle (NV33BV-2, World Precision Instruments, USA) attached to a Nanofil Injection system (UMP3T-1, SMARTtouch Microinjection System, World Precision Instruments, USA) or (c) injection of 1000 nL, at a rate of 100 nl s⁻¹, of 1M KCl into the superficial layers of the cortex (∼500 μm deep from pia; using aforementioned injection system). For this protocol, the order above was used each time, due to the trauma caused from the pinprick, which was required for the local injection of KCL.

The lowest LED current found to induce a CSD twice was multiplied by 1.5 to ensure that optogenetic induction was supra-threshold throughout the experiment. The inter-induction interval was set to 10 min to allow full recovery of neuronal activity (figure 3(D)). Two baseline light-induced CSDs were followed by a randomized I.P. administration of a substance, either: vehicle (Saline; 0.5 ml), MK801 (3 mg kg⁻¹; 10009019, Cayman Chemical, Michigan, USA), or ketamine (15 mg kg⁻¹, Dechra, UK). A minimum of six post-injection light stimuli were delivered. Analysis of neural activity was conducted post injection of NMDA antagonists. One animal injected with ketamine showed no change post-injection in power (theta), similar to saline injected animals, and in contrast to all other animals injected with either MK801 (n = 5) and ketamine (n = 3), and was excluded from the study due to concerns that the i.p injection had failed, and the drug did not reach the brain. To check for a difference in the stimulation thresholds between treatment groups we compared the OptoThreshold and this was not significantly different between treatment groups (Vehicle, 13.18 ± 4.99 mJ; MK-801, 18.00 ± 4.99 mJ; ketamine, 4.55 ± 2.17 mJ) (p = 0.0591, Kruskal–Wallis ANOVA).

Zoom In Zoom Out Reset image size

To explore the neurophysiology at the site of induction of the CSD, experiments were performed with arrays over both the MC and somatosensory/visual cortex (n = 3; figure 8). Here, a larger craniotomy was completed over the MC to facilitate the placement of the gSGFET. Some experiments were conducted to compare recording CSD through the skull with subdural recordings (n = 5; figure 9). Here, the craniotomy for on dura placement was narrower and over the precise neuroanatomical area of interest.

The experiment was terminated using a sodium pentobarbital overdose (i.p. injection). Brains were removed and placed into 4% PFA (J19943.K2; Alfa Aesar, UK) chilled at 4° overnight. Coronal sections of 50 µm thickness were cut using a vibrating microtome (Leica, Wetzlar, Germany). Sections were mounted using Vectashied (2B Scientific Ltd). Tiling images were acquired using a Zeiss Axio Observer inverted widefield microscope and Zen microscope software (ZEISS), with 10× objective at 1024 × 1024 pixels resolution. All images were acquired using the same microscope parameters for analysis purposes. Cortical fluorescence distribution in coronal slices was quantified using ImageJ. The cortex was manually selected as the region of interest and an intensity threshold, consistent across animals applied. Pixels above this value were counted with total area in mm² calculated based on pixel dimensions and magnification. Using the Allen Mouse Brain Atlas, each slice was binned based on distance from bregma. The transduced area within each slice was taken as a representation of the surrounding posterior 250 µm of tissue. The total transduction volume (mm³) was calculated as the summation of each 250 µm volume.

Flexible epidural neural probes (10 μm in thickness) containing an array of 16 graphene micro-transistors (4 × 4 array, 400 μm separation) with active areas of either 150 × 100 μm^2, 100 × 100 μm² or 50 × 50 μm² were used in this study. gSGFET arrays were fabricated at the clean room facilities of IMB-CNM as reported in (Masvidal-codina et al 2019). In brief, single-layer graphene was grown by chemical vapor deposition and transferred to a silicon wafer previously coated with a polyimide layer and patterned metal traces. After defining the graphene channels and before evaporating a second metal layer, UVO treatment was applied to improve the graphene-metal interface and reduce its contact resistance (Schaefer et al 2020). Finally, SU-8 was used as passivation layer and the polyimide layer was etched to define the geometry of the neural probes. Devices were gently peeled off from the wafer and inserted to zero insertion force connectors for electronic interfacing.

Recorded signals were analyzed using Python 3.7 packages (Matplotlib 3.2.0, Numpy 1.17.4, Pandas 0.25.3, Seaborn 0.9.0, Neo 0.8.0) and the custom library PhyREC (PhyREC4, https://github.com/aguimera/PhyREC). Transistor recordings were calibrated by interpolating the current signals into the corresponding branch of the in vivo measured transfer curve as reported by Masvidal-codina et al (2019). For two experiments the signal interpolation to the transfer curve was not possible due to insufficient measured voltage range or presence of artifacts, for these experiments the gain at the V_gs bias point of the transfer curve for each transistor was used. No high pass filtering is applied to the recorded extracellular potentials for the analysis of CSD waveforms, extraction of CSD parameters or plotting to avoid any signal distortion. Error bars in all plots represent standard error.

For each CSD extracellular voltage shift the following parameters were extracted: latency, peak amplitude, duration and area under the curve (AUC). Raw signals were down sampled to 3 Hz and zero voltage was set to the average value of the 5 s before the light stimulus. We defined the onset of the CSD as the onset of the negative shift with a threshold at −4 mV. Latency was calculated as the difference between the starting stimulation time and CSD onset. Peak amplitude was defined as the minimum value of the wave and reported as an absolute value. CSD duration was defined as the time that the extracellular voltage remained below −4 mV; it was determined using below and above thresholds. AUC was defined as the area below zero voltage during the duration of the depolarization determined. CSD propagation speed was estimated as the distance between the LED induction site and the middle of the transistor array (4 mm) divided by the mean latency CSD onset. Code available upon request.

To determine the direction of propagation, a vector pointing from the lowest to highest latency values in the transistor array matrix was used. A threshold of 2.5 standard deviations from the mean of the latencies was set to remove outliers that could distort the direction of propagation. The angles between the propagation vector of the first induced CSD and the subsequent ones was determined by $\angle \left( {x,y} \right) = \arccos \frac{{x,y}}{{x \cdot y}}.$, where x and y are the two vectors.

For each CSD extracellular voltage shift, the calibrated raw signals were either not filtered for full-band or bandpass filtered for the different bands (delta: 1–4 Hz, theta: 4–8 Hz, alpha: 8–12 Hz, beta: 12–30 Hz and gamma: 30–100 Hz). Then, the root mean squared (RMS) was calculated using a sliding window (window_size: 0.5 s, steptime = window_size/10) and resulting traces were passed through a median filter (scipy.signal.medfilt) to remove artefacts. RMS values for each band were averaged at given epochs with respect to the LED stimulation (referred to as 0 s), the depolarization onset (t_onset, defined as the time extracellular voltage goes below −4 mV of the raw signal) and the depolarization end (t_end, defined as the time the extracellular voltage crossed again the −4 mV). The epochs are: before the LED stimulus (bf.: −100 s to −10 s), a few seconds after (aft.: 0 s to t_onset − 10 s), depolarization onset (onset: t_onset − 10 s to t_onset + 10 s), depolarization (dep.: t_onset + 10 s to t_end) and recovery 1–7 being every minute after the end of the depolarization (r1: t_end to t_end + 60 s; r2: t_end + 60 s to t_end + 120 s, etc). To evaluate the neural effects of CSD, RMS power values where normalized with respect to the bf. timepoint. Values for each transistor in the array over multiple CSD events were averaged to yield a single value for each experiment.

To classify the different shapes of the depolarization waves, raw signals were resampled at 10 Hz. Signals from −30 to 400 s around each light-stimulus were extracted and zero voltage was set to the average value of the 30 s before the light stimulus. A threshold of −4 mV was used for CSD detection. Then, all detected waves were time aligned to the minimum of the (first) depolarization peak (0 s) and epochs from −50 to 140 s extracted. Depolarization voltage is defined as the mean voltage value between −1 and 1 s. Then, three features where selected to perform the clustering: integral of the prehyperpolarization (area >0 mV in the −15 s to 0 s), integral of recovery hyperpolarization (area >0 mV in the 0s to 75 s) and the second to first peak amplitude ratio. The second peak was detected during the 100 s after the first peak by finding the valley (with the minimum of the first derivative) and then searching from the minimum of voltage after the valley. Based on these three features waveforms were clustered by defining a threshold for each group: first they were classified as DP (second to first peak amplitude ratio >0) or SP. Then DPh (PreHyper >1); DP (PreHyper <1); SPh (RecHyper >1) and others were classified as SP, LV if amplitude was lower than −6 mV or ND if a depolarization was not detected at that channel.

In these studies, the variable of interest was isolated and statistically examined. In general, for comparisons between groups of two, unpaired or paired T-tests were used, while for three or more groups, ANOVA with post hoc corrections were utilized. To examine the frequency of evoked CSDs for different optogenetic stimulations (figure 1), since the same animals were used in a non-randomized fashion for each modality, Fisher's exact test was used to compare between groups. For comparing the parameters of CSDs to different modalities (figure 4), each animal, either male or female, was given the same sequential treatment (due to the nature of induction methods), with at least a single CSD being evoked to each modality. With this data collected, a one-way ANOVA was run to compare between (a) genders, and (b) groups. As genders showed no differences between induction using either optogenetics, pin-prick from a needle, or injection of KCl, the data from each gender was combined to allow for a more informed comparison between modality types. Examining the change in normalized frequency bands between the different CSD shapes (figure 5) utilized the two-way repeated measures ANOVA with Sidak's post-hoc comparisons for both (a) sub-types and (b) combined sub-types. For the pharmacological study (figure 7) parameters (Amplitude), a two-way repeated measures ANOVA was used with Sidak post-hoc comparisons. Examining the time response profile for recording at the site of induction also utilized a two-way repeated measures ANOVA with Sidak's post-hoc test to examine the differences between responses over the 10 s of optogenetic stimulation and up until recovery of the voltage. To examine electrophysiological recordings between epidural and skull positioning of the gSGFET, the two arrays were placed and optogenetic stimulation was applied in the MC (figure 9). As the CSD propagated across the brain, it was recorded at both sites. SNR was calculated as the ratio of the peak CSD amplitude to the standard deviation of the 30 s prior to CSD induction. To compare the SNR, an artificial SNR level was set to 3, with slight variation around this, and then a one-way ANOVA with Tukey's post-hoc test was used to compare this regularly used artificial SNR level to the SNR from epidural and skull gGFET recordings. Next, for other parameters such as amplitude, AUC, and propagation velocity (pro. vel.), Welch's, unpaired, T tests were used to examine whether there were differences between recording sites. Finally, two-way ANOVA was used to compare normalized frequency band activity at the two recording sites. Throughout the manuscript standard significance classifications were used (* = p < 0.05; ** = p < 0.01; *** = p < 0.001).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The experimental paradigm for triggering light-induced CSDs in awake mice is illustrated in figure 1(A). Surgery was performed to inject viral vector to transduce neurons in the MC and express light-sensitive channelrhodopsin (ChR2) (see section 2, figure 2). Head bars were attached and in the week prior to experimentation mice were habituated to the head-fixation apparatus over several days (figure 1(B)). To detect CSD, a 16-channel gSGFET was placed on the dura over the somatosensory and visual cortex posterior to the ChR2-injected area (illustrated in figures 1(C) and (D)). Before recording, a transfer curve is obtained that is used to determine the optimal recording settings and also for post-recording current-to-voltage conversion (figure 1(E) and section 2).

Continuous 488 nm blue light illumination (5–10 s) reliably induced CSD (figure 1(Fi)). Note the correlated activity suppression shown in the high-pass filtered signal and in the corresponding z-scored spectrogram. To examine the specificity of light wavelength and stimulation parameters we used a green LED 594 nm (figure 1(Fii)) with the same light duration and power intensity that successfully evoked a CSD with blue light, or used pulsed blue light (20 Hz) with a total light-on duration identical to the square pulsed light duration (figure 1(Fiii)). Whereas constant blue light illumination always evoked a CSD, pulsed blue light or continuous green light rarely evoked a CSD (figure 1(G)). This indicates that a sustained neuronal depolarization, caused by a single, seconds-long square pulse stimulation with blue light in ChR2-expressing areas is an effective and robust mechanism of CSD induction.

Examination of viral transduction volume post hoc revealed that expression of cortical tissue, greater than 1 mm³, was required to express ChR2 to reliably induce CSD optogenetically (figure 2). Viral injections were variable in the total volume of tissue transduced (5.46 ± 1.7 mm³ n = five animals) but consistently transduced the area of MC from +0.5 to +2.5 mm (anteroposterior coordinates relative to bregma) where the optic fiber light was positioned during experiments. The light intensity required to trigger a CSD ranged 5.89–147.6 mJ, with mean of 33.98 ± 7.065 mJ (n = 44 animals).

CSD parameters were extracted from individual micro-transistors (figure 3(A)). High amplitude (14 ± 0.3 mV) negative shifts in the recorded extracellular voltage lasting tens of seconds (46.0 ± 1.2 s) in duration were recorded. Data corresponding to 633 CSD waveforms recorded from a total of 54 CSDs (2 CSDs evoked in each of 27 animals). However, as the histograms indicate (figure 3(B)), there was a large degree of variability in these parameters, the 25%–75% range for amplitude and duration were 9.8–16.4 mV, and 28.9–50.2 s respectively. This variability reflects the heterogeneity in CSD waveforms recorded (see figure 5). In contrast, propagation speed was more consistent, 3.0 ± 0.1 mm min⁻¹ (2.8–3.3 mm min⁻¹, 25%–75% range). Light-evoked CSDs in the MC also induced transient changes in LFP activity. For these experiments we recorded CSDs in areas of propagation, somatosensory and visual cortex (figures 1(C) and (D)). Within seconds of light stimulation, a gradual decrease in 1–30 Hz RMS power could be detected, minutes before the actual CSD invaded the area of the brain underneath the transistors. In contrast, gamma RMS power (30–100 Hz) increased just prior to CSD invasion, indicative of intense neuronal firing preceding the massive depolarization and subsequent neuronal silencing (figures 3(C) and (D)) similar to an increase in multi-unit activity previously reported (Houben et al 2017). The rate of recovery of neuronal activity post CSD were similar for frequencies 1–30 Hz with full recovery evident 4 min after the end of the DC-shift. In some recordings there was a prominent increase in gamma RMS power 30–100 Hz upon recovery, indicative of rebound excitation (figures 3(D) and 5(D)).

The electrophysiological features of optogenetically-induced CSDs in virally-injected mice were compared to standard induction methodologies including direct application of KCl to the brain and pinprick in the same animal (figures 4(A)–(C)). CSD waveform parameters (i.e. peak amplitude, duration and AUC values were not statistically different between CSD induction types, nor between male and female mice. These values (figure 4(D)) are within previously reported ranges for KCl-induced CSDs in non-virally-injected, wild-type mice (Eikermann-Haerter et al 2015). However, although there is no difference in the duration or AUC of CSDs between groups, we did observe a difference in the time taken for activity suppression in frequencies above 1 Hz to fully recover (figure 4(E)). Compared to optogenetic- and pinprick-induced CSDs, recovery from KCl-mediated CSD was significantly slower (figure 4(E)).

Due to the mapping capabilities of the gSGFET arrays, we observed variability in CSD amplitudes and duration (figure 3(B)), resulting from distinct waveform patterns. Figure 5(A) displays the CSD waveform captured by 14 individual transistors as a CSD propagated over the array. Some waveforms have a single peak, whereas others present a double peak. An additional feature is the presence of either a pre- or post-CSD hyperpolarization. We obtained a database of 2323 CSDs waveforms recorded from 224 optogenetically-induced CSD events from 26 mice. We analyzed three waveform features pre-hyperpolarization (F1), recovery post-hyperpolarization (F2) and second-to-first peak amplitude ratio (F3) to classify the recorded waveforms into 4 different clusters: DP (double-peak), DPh (double-peak hyperpolarization), SPh (single-peak), SPh (single-peak hyperpolarization) and a 5th category where the CSD was of low amplitude (<6 mV peak), (figures 5(B)–(E) and section 2). The high channel number and low-invasive induction methodology of our preparation allowed us to study the occurrence and spatial distribution of each waveform. Under our recording conditions, using 'normal', non-disease model mice, single peak accounted for 84% of waveforms split similarly between SP and SPh (42% each). DP and DPh were less common (10%), 5% each. Double peak waveforms (DP and DPh) had a higher duration and amplitude than single peak waveforms (SP, SPh) (figure 5(E)) and interestingly, also a faster rate of depolarization (data not shown). DPh was the cluster with the highest values in all parameters tested. We performed frequency band analysis on the waveform types and observed that double peak waveforms caused a higher percentage of activity suppression compared to single peak waveforms (figures 5(F)–(G)). The arrays were predominately placed in areas of propagation. In these locations no spatial distributional trends across the grid for specific waveforms were identified.

We examined the reproducibility of repeated optogenetic stimulation (figure 6(A)). Separation between light stimulations was set to 10 min to allow recovery from transient silencing of neural activity (figure 3(D)). Figure 6(A) displays 14 transistor recordings from one animal in response to eight optogenetic stimulations. In contrast to AC-coupled, or DC-coupled passive electrodes which encounter baseline drift and require either high-pass filtering or electronic off-set readjustments (Nelson et al 2008, Nasretdinov et al 2017), gSGFETs are able to record repeated rounds of CSD over several hours with minimal baseline drift (0.16 ± 0.9 mV/10 min, n = six animals), highlighting the superiority of this methodology for long (hours) continuous recordings (figure 6(A)). We analyzed the propagation patterns of CSD across the transistor array (figure 6(B)), and categorized the CSD waveform recorded from each transistor for repeated trials. Overall CSD waveform remained stable with roughly the same percentage of events classified as single peak and double peak, although SPh waveforms decreased and SP waveforms increased during repeated stimulations (figure 6(D)), and there was a trend towards more double peaks by stimulation 7 and 8. Figure 6(C) shows an angular histogram of the difference between the propagation direction of any CSD and the direction of the first induced CSD of that experiment. Results indicate very small changes in the propagation angle over repeated optogenetic stimulations.

To test the effect of drugs capable of suppressing or modulating CSDs we increased the light stimulus to 1.5 times the threshold, OptoThreshold, to ensure supra-threshold stimulation across the experiment duration. The protocol consists of two CSD inductions prior to intraperitoneal systemic injection of vehicle or drug followed by an additional six supra-threshold stimulations post-drug (figure 7(A)). In this study, we evaluated the efficacy of two NMDA receptor antagonists. A representative trace of the recorded signal corresponding to one transistor of the 16-channel array is shown for an animal in the saline, MK-801 and ketamine groups (figure 7(A)). To validate the methodology for testing drug effects we used MK-801, a well reported CSD-inhibitor (Chung et al 2019). Intraperitoneal (i.p.) injection of 3 mg kg⁻¹ MK-801 produced rapid (<10 min) suppression of CSDs that did not appear in any of the subsequent light stimulations (n = 5). MK-801 cannot be used therapeutically due to a number of serious adverse side-effects, so we next turned our attention to a potentially more clinically relevant NMDA antagonist, ketamine. Rodent studies indicate that ketamine can reduce the amplitude of CSDs (Krüger et al 1999), or block CSD induction completely (Hernándéz-cáceres et al 1987, Rashidy-pour et al 1995). The effects are however transient, lasting 20–45 min after a single injection. An effect we also observed in our studies (figure 7); complete inhibition only lasted 15–25 min post-injection, although the CSD amplitude and AUC was smaller (figures 7(B) and (D)) and propagation was slower (figure 7(E)) for the remaining trials compared to pre-drug values.

To examine whether MK-801 and ketamine prevented induction, as well as propagation of CSDs, we modified the experimental protocol and used two epidural arrays; one at the site of LED illumination (gFET#1), motor cortex, and the other, (gFET#2), over somatosensory/visual cortex (figure 8(A)). LED illumination of the ChR2 expressing area resulted in an immediate depolarization, most prominently on the top-left corner of the transistor array (gFET#1). This immediate depolarization is not seen in gFET#2, ∼3 mm away from the illumination site (figures 8(B) and (C)). CSDs induced in the MC propagated and were detected later by the second transistor array gFET#2. After administration of MK-801 (3 mg kg⁻¹, n = 2), or ketamine (15 mg kg⁻¹, n = 1), no CSD was detected at gFET#2 in accordance with the single-transistor array experiments (figure 7); while at the induction site (MC) a neuronal depolarization can still be observed (figure 8(D)). Light stimulation results in a time-locked depolarization of ∼5 mV, which then transitions into a larger amplitude depolarization and full CSD (figure 8(D)). In contrast, after administration of MK-801 or ketamine the initial light-induced depolarization returns quickly post-light stimulus to baseline values without induction of a CSD (figure 8(E)).

Zoom In Zoom Out Reset image size

Finally, we investigated the possibility of recording CSDs electrographically through an intact skull. To do this and to quantify our results with epidural placement of arrays, we utilized two gSGFET arrays in a single experiment, one placed over the motor/somatosensory cortex, and the other over the ipsilateral somatosensory/visual cortex and optogenetically-induced CSDs (figure 9(A)). One of the arrays was placed epidurally and the other on top of the skull. There is a reduction in spatiotemporal properties of recorded CSDs across the array (figure 9(Eii)). The resolution in propagation and CSD waveform across and between individual transistors achieved by subdural placement of the gSGFET is lost. There is also a reduction in amplitude of the detected CSDs when recording through bone (epidural was 17.14 ± 2.66 mV, through skull was 9.75 ± 1.57 mV; n = six animals; p = 0.0436, Welch's unpaired T-Test) (figure 9(Cii)). However, CSDs can be detected through the skull travelling at a propagation velocity of 3.198 ± 0.265 mm min⁻¹ (figure 9(Civ)). Additionally, we could detect suppression of higher frequency (1–30 Hz) rms amplitude correlated with the CSD, although this suppression recovered faster compared to epidural recordings (figure 9(D)). In order to exclude potential electrical paths through burr holes and craniotomies, we performed non-invasive optogenetic induction and recording of CSD (supplementary figure 1 (available online at stacks.iop.org/JNE/18/055002/mmedia)). These experiments demonstrate that CSDs can be electrographically recorded through a completely intact mouse skull using gSGFETs.

Zoom In Zoom Out Reset image size