This study explores the novel use of enzyme-based microelectrode array (MEA) technology to monitor in vivo neurotransmitter activity in piglets. The hypothesis was that glutamate dysregulation contributes to the mechanism of anesthetic neurotoxicity. Here, we present a protocol to adapt MEA technology to study the mechanism of anesthesia-induced neurotoxicity.
Every year, millions of children undergo anesthesia for a multitude of procedures. However, studies in both animals and humans have called into question the safety of anesthesia in children, implicating anesthetics as potentially toxic to the brain in development. To date, no studies have successfully elucidated the mechanism(s) by which anesthesia may be neurotoxic. Animal studies allow investigation of such mechanisms, and neonatal piglets represent an excellent model to study these effects due to their striking developmental similarities to the human brain.
This protocol adapts the use of enzyme-based microelectrode array (MEA) technology as a novel way to study the mechanism(s) of anesthesia-induced neurotoxicity (AIN). MEAs enable real-time monitoring of in vivo neurotransmitter activity and offer exceptional temporal and spatial resolution. It is hypothesized that anesthetic neurotoxicity is caused in part by glutamate dysregulation and MEAs offer a method to measure glutamate. The novel implementation of MEA technology in a piglet model presents a unique opportunity for the study of AIN.
Every year, millions of children undergo anesthesia for both invasive and non-invasive procedures in the United States1. For years, anesthesia providers have reassured parents of the safety of anesthetics, even in small children and neonates. However, in 1999 it was found that transient blockage of the N-methyl-D-aspartate (NMDA) subtype of glutamate receptors during early life could cause widespread neuronal apoptosis in rats2. Recently, the FDA released a drug safety communication that will require the labels of anesthetic drugs to include a warning about general anesthetics and their potential negative effect on the developing brains of children younger than 3 years old (U.S. Food and Drug Administration, 2017). However, there is still a need to elucidate possible mechanisms and potential neuroprotective measures.
Normal activity of neurotransmitters such as glutamate and gamma-amino butyric acid (GABA) are critical for normal neurodevelopment to occur. Although most of the pathways involved in AIN are still elusive, neurotransmitter systems are very likely to be involved since anesthetics are known to modulate these pathways to produce unconsciousness. In particular, the excitatory neurotransmitter glutamate causes excitotoxicity when its activity is dysregulated. This neurotransmitter is normally involved in neurogenesis, neural plasticity, synaptic and neural growth, and a number of other critically important brain functions. However, prolonged activation of glutamate receptors can cause excitotoxicity and neuronal apoptosis, particularly under stress conditions such as surgery, oxygen deprivation, and reduced energy availability3. Binding of glutamate to the NMDA receptor has been shown to cause Na+ and Cl− influx. The subsequent depolarization is thought to lead to Ca2+ channel opening and release of intracellular Ca2+ stores4. This dysfunction likely leads to a cascade of metabolic derangements which eventually decrease neuronal proliferation, increase inflammation, and lead to neuronal death. Despite these hypotheses, the true mechanism(s) of AIN remain unclear5. Because of its role in apoptosis, glutamate dysregulation represents a novel pathway that may contribute to the mechanism of previously documented neuronal apoptosis, a feature of AIN.
One of the hindrances on the study of neuronal processes is their high complexity, especially in the setting of neuronal development. The first few months of life are the period of maximal vulnerability to injury, during which most of the important steps of neuronal development such as physiologic apoptosis (neuronal pruning), synaptogenesis, gliogenesis, and myelination take place6. Given the complex nature of neuronal communication and the difficulty of studying these processes without disrupting normal CNS function, new technologies have been developed which aim at the in vivo detection and quantification of important elements of neuronal communication.
Enzyme-linked MEA technology was used in this study as a novel way to study the mechanisms of AIN in a clinically-relevant piglet model. This technology can be used to study complex in vivo electrochemical brain processes, including glutamate dysregulation. The incorporated 4-channel platinum recording sites of the MEAs (2 glutamate-sensitive sites and 2 sentinel sites) allow self-referencing, which contributes to detection accuracy. In addition, an exclusion layer is applied to each electrode, conferring selectivity by preventing other interfering molecules from being detected7. Moreover, the low-profile design of the MEA allows for minimal tissue trauma compared to earlier technologies. This same feature confers to MEAs a higher spatial resolution, which facilitates the study of microscopic regions in the brain. For example, discrete regions of the hippocampus (dentate gyrus, CA1, CA2) can be specifically studied8. Specific details on the functionality of MEAs have been previously described9.
In comparison to MEA electrochemistry, microdialysis incorporates a membrane placed between the solution of interest and a solution of similar composition, allowing detection of extracellular fluid changes10. Although microdialysis is a mainstay of neurochemistry and has long been used for the detection of neurotransmitters, it has the disadvantage of low time resolution, delayed detection of glutamate change, and significant tissue trauma11.
MEAs can indirectly detect neurotransmitters such as glutamate, acetylcholine, and choline, by using appropriate oxidase enzymes that produce electroactive reporter molecules such as H2O2 or O212,13.
MEA technology has been widely used in rats and non-human primates for the study of neurotoxicity in the context of pathophysiological processes other than AIN7,14. Among some of these pathophysiological processes, MEA technology has been used for the study of Alzheimer's disease, epilepsy, traumatic brain injury, and the effect of pharmacological compounds on synaptic communication8,15,16,17. Although MEAs have been used to study these pathologies in rats and non-human primates, the high developmental similarity between humans and piglet brains makes the adaptation of the MEA technology in piglets a highly suitable technique for the study of AIN mechanism(s)18.
Piglets (Sus scrofa) are received through a local farm pre-approved by The Ohio State University (OSU) Institutional Animal Care and Use Committee (IACUC). Following approval of the protocol, animal experimentations are done in accordance with IACUC policy.
1. Piglets and Piglet Handling
2. Development and Customization of MEAs for AIN Studies in a Piglet Model
NOTE: This technology uses enzyme-based MEAs that are pre-coated with enzyme and electroplated with m-phenylenediamine dihydrochloride (mPD). The electrodes were custom designed with a 40-mm rigid shaft and manufactured for use in piglets (Figure 1).
3. Anesthesia and Use of Custom Stereotaxic Apparatus for the Piglet
4. Measurement of Extracellular Glutamate Under Sevoflurane Anesthesia
5. Perfusion and Sacrifice
This proof-of-concept study with enzyme-based ceramic MEA technology in a piglet model can provide exceptional insight into the glutamate dynamics underlying AIN. This study further demonstrates that enzyme-based MEA technology can be successfully adapted in the piglet model to measure physiologic and anesthesia-associated changes in neurotransmitter activity with high sensitivity, and high spatial and temporal resolution. Physiologic homeostasis was maintained throughout our experiments using clinically relevant methods and standards, and no piglet exhibited signs of physiologic perturbations.
The data obtained indicates the ability of MEAs to precisely and spatially resolve neurotransmitter measurements in cortical and sub-cortical brain structures. The use of a stereotaxic apparatus enables clear identification of a reference surface structure (bregma) in order to consistently locate the region of interest, regardless of individual differences in piglet size and anatomy. Clear visualization of the sutures facilitates consistent regional placement of the MEA with accuracy in the micrometer range (Figure 4). Obtaining access to the cortical surface of the brain is minimally traumatic with negligible bleeding, ensuring that any in vivo glutamate dynamics are not due to unintentional systemic or local insult (Figure 5). The custom, rigid MEA is then aligned perpendicular to the frontal plane of the piglet (Figure 6). Failure to properly align the MEA prior to insertion may prevent accurate spatial recording of the targeted region, especially for subcortical regions.
Real-time in vivo glutamate measurements were taken in the hippocampi of 3-4-day old piglets (n = 4) under 2.5-3% sevoflurane anesthesia (approximately 1 MAC). Amperometry measurements were recorded at 4 Hz and converted to concentration using a linear regression based on calibration parameters (Figure 8, Panel A). For each time point, the signals from the two glutamate-sensitive sites were averaged before subtracting the averaged sentinel signal to yield a corrected glutamate signal. These continuous measurements were smoothed by applying a moving average to better visualize the overall trend over time (Figure 8, Panel B). The mean basal glutamate concentration was calculated to be 4.61 ± 0.02 µM and remained relatively stable over the course of anesthetic exposure. Transient glutamatergic activity was identified in one animal by analyzing peaks in the signal that were not correlated with the sentinel signal (R2 < 0.5) and exceeded a signal-to-noise ratio of 3 (Figure 9, Panel A). A total of 116 transient peaks were detected over a period of 3.5 h (Figure 10, Panel A). The amplitude of the resulting transient peaks was generally observed to be within the 1 µM range (Figure 10, Panel B). In order to quantify the duration of each transient, the time (t80) required for each maximum peak value to decay 80% was obtained (Figure 9, Panel B). The mean t80 of all glutamate transients during the 3.5 h recording period was 4.68 ± 0.82 s (Figure 10, Panel C). These data demonstrate that it is possible to accurately measure both prolonged and transient neurotransmitter activity in a subcortical region of the anesthetized piglet brain.
Figure 1: Visual comparison of SG-2 microelectrode array types. SG-2 arrays contain two glutamate-sensitive sites and two glutamate-insensitive sentinel sites (150 µm x 20 µm per site).(A) A flexible-shaft microelectrode array is shown on the left. The rigid-shaft microelectrode array was custom-designed for use in piglets and permits deeper implantation in large animals. Please click here to view a larger version of this figure.
Figure 2: Overview of microelectrode array preparation and calibration process. The total MEA preparation and calibration last approximately one week. The coating enzyme, exclusion layer, and calibration analytes are specific to the neurotransmitter of interest. Please click here to view a larger version of this figure.
Figure 3: Placement of piglet in the stereotaxic apparatus. The piglet mouth is placed on mouth bar directly posterior to the canine teeth. The penetrating ear bars are inserted into the ear canals to secure the posterior end of the skull. Please click here to view a larger version of this figure.
Figure 4: Placement of piglet in the stereotaxic apparatus for craniotomy. (A) The piglet's head is tightly secured within the custom stereotaxic frame, ensuring consistent placement of the MEA. Equidistant placement of penetrating ear bars is visible. (B) Midline anterior-posterior incision along the scalp. Scoring of the skull was avoided to visualize coronal and sagittal sutures and optimize visualization of bregma. The scale bar is shown to indicate the relative size of the incision and the location of craniotomy window. Please click here to view a larger version of this figure.
Figure 5: Craniotomy for access to the hippocampus. (A) The scalp further reflected to expose the approximate location of the MEA insertion according to stereotaxic coordinates. The circled area is marked (black dot) to guide the craniotomy. (B) The craniotomy window (0.25 cm2) with skull flap removed to expose the underlying dura mater. (C) The meninges carefully removed to expose superficial cerebral cortex without tissue trauma. Please click here to view a larger version of this figure.
Figure 6: MEA positioning and insertion into the hippocampus. (A) Placement of MEA at the bregma to determine a relative stereotaxic location of the hippocampus. (B) Stereotaxic placement of the MEA at the brain surface to determine the hippocampus insertion depth. Silver pseudo-reference electrode securely placed under scalp (indicated by arrow). (C) the MEA inserted at the appropriate depth to obtain real-time, in vivo extracellular glutamate measurements in the hippocampus. Please click here to view a larger version of this figure.
Figure 7: MEA behavior during the 30-min baselining period. The initial rapid increase corresponds to the descent of the MEA into the hippocampus using the micromanipulator. The baseline period begins once the MEA has reached the appropriate depth (dotted line). Extracellular glutamate measurements will decrease over a period of 30 min and should not be interpreted as physiologic readings. Please click here to view a larger version of this figure.
Figure 8: Real-time extracellular glutamate measurements in the hippocampus of a neonatal pig under sevoflurane anesthesia. (A) The moving average of the glutamate concentration in the hippocampus of one neonatal pig under sevoflurane anesthesia (using 10 data points). Measurements were taken at 4Hz for 3 h after a brief 30 min baselining period. (B) Smoothing of glutamate measurements using a moving average of 100 points every 15 min to better visualize the trend. Please click here to view a larger version of this figure.
Figure 9: Identification of transient glutamate activity in the hippocampus of a neonatal pig under sevoflurane anesthesia. (A) Transient glutamate peaks (in red) are indicated on the real-time glutamate tracing. Peaks were considered significant when the signal to noise ratio exceeded 3 and their signal was not correlated with the sentinel signal (R2 < 0.5). (B) A representative transient peak identified in Figure 9, Panel A. Dotted lines indicate the total duration required for the peak to decay 80%. Please click here to view a larger version of this figure.
Figure 10: Characterization of glutamate transients in the hippocampus of a neonatal pig under sevoflurane anesthesia. (A) A number of transient glutamate peaks in 15 min bins. Peaks were considered significant when the signal to noise ratio exceeded 3 and their signal was not correlated with the sentinel signal (R2 < 0.5). (B) The amplitude of transient glutamate peaks. Error bars indicate the standard error of the mean. (C) Mean T80 of transient glutamate peaks. Error bars indicate the standard error of the mean. Please click here to view a larger version of this figure.
From the start of the experiment, the piglet's physiological homeostasis must be maintained as described in this lab's prior publication21. Minimal monitoring should include pulse oximetry, electrocardiography, capnography, non-invasive blood pressure, and temperature. Trained investigators are required so that physiologic perturbations (e.g., hypo/hyperthermia, hypoxia, hypotension, arrhythmia) can be appropriately corrected.
Prior to induction, in vitro MEA calibrations are performed to establish functionality and selectivity of the MEA under known conditions. The calibration and plating of MEAs is critical for effective use of the technology. There are many potential errors that can arise during calibration. Calibration can identify these issues as well as improper plating, which leads to incorrect interferent response. A more detailed tabular account of errors that may occur in MEA response has been compiled, along with notable causes and suggested solutions, which should prove a useful instrument for likely troubleshooting (Table 1). It is important to note that prior to both calibration and plating, the glass reference electrode should be checked for the presence of air bubbles or white discoloration, as either will negatively impact MEA function and recording accuracy.
Symptom | Cause | Corrective Action |
No Signal | Electrode not connected | Properly connect electrode to headstage and headstage to FAST amperometry system. |
No power to the FAST amperometry system | Turn on power switch on back of FAST system | |
Signal Noise | Electrode contaminated by blood | Continuously irrigate the brain surface during electrode insertion |
Rinse the electrode immediately in dH2O | ||
Enzyme coating is loose | Clean and recoat the electrode | |
Reference electrode was not inserted or coated | Coat and place the reference electrode farther under the scalp | |
Electrode is detecting movement of brain surface | Usually occurs in superficial structures. Insert the electrode deeper (1 mm at a time) if possible | |
Animal Movement | Animal is inadequately secured | Move the animal in the posterior direction to better secure earbars on the skull. If necessary, elevate the torso to allow for better body alignment. |
Animal is inadequately anesthetized | Verify the integrity of the anesthetic equipment. Titrate the anesthetic to an effective dose and administer an intramuscular rocuronium dose (5 mg/kg) | |
Inaccurate electrode placement | Electrode is not correctly aligned | Readjust the electrode while maintaining proper connection to the headstage. |
Stereotaxic coordinates are inaccurate | Ensure that the piglet atlas being referenced does not use another reference point or plane of alignment. | |
Be careful not to obscure the suture marks by scoring the skull. |
Table 1: Instructions for troubleshooting MEA use in piglets. Possible causes and corrective actions to assist with optimization and troubleshooting.
A stereotaxic atlas for the piglet is used to determine the stereotaxic coordinates of the area of interest with respect to a known point such as bregma18. Ear bars should be properly secured to ensure that the skull is level and fully immobilized. Care should be taken during the midline incision of the scalp to avoid scoring the skull as this may affect visualization of the suture lines. The craniotomy window should be large enough to accommodate the MEA.
This protocol presents a number of technical challenges that require a well-stocked operating suite and an investigator/team skilled in the surgical and anesthetic aspects of the protocol. The model additionally presents financial limitations in that the piglet model is more expensive than the rodent model; however, it is significantly less costly than the use of non-human primates, which can cost thousands of dollars. The use of MEA technology presents its own challenges, as the procedure of coating and plating the electrodes manually require a skilled investigator or assistant to ensure sufficient selectivity and reliable function. The microelectrodes themselves are fragile, as they are made of ceramic, and thus easily damaged if proper caution is not observed. Microelectrodes are subject to interference from other electrical devices, which can create noise in recordings, and from blood at the operative site, which can occlude the recording sites. The need for specialized equipment presents an additional burden as a stereotaxic surgical frame must be custom built to immobilize the piglet skull during implantation. The stereotaxic frame, glutamate oxidase, and the electrodes themselves are all costly. Additionally, the lack of a piglet stereotaxic atlas from within the last decade poses technical limitations that require particular expertise to determine the specific location of deep structures in the piglet brain. Development of a new stereotaxic atlas, perhaps using magnetic resonance imaging, would greatly enhance the ability to use this technology in piglets.
The piglet is a clinically relevant model for the study of AIN largely due to the parallels that exist between this species and the human neonate, as both possess similar brain structure and development. Unlike more commonly used models such as mice or rats, the piglet has a greater CNS similarity to humans, which lends to the translatability of the model's results. The piglet model is additionally cheaper and involves less complicated handling than a non-human primate model. The piglet model is intended to examine the process by which anesthesia might induce developmental neurotoxicity, measure its contribution to neurological damage, and combat the issue of damage caused by confounding variables. For instance, hypoxia may be misconstrued for damage caused by anesthetics as it has global effects on the brain. The piglet is utilized with the same surgical and anesthetic conditions as those used in human medicine to ensure fidelity of results.
The use of ceramic-based MEA technology eliminates several of the disadvantages associated with the contemporary technique of microdialysis. Microdialysis has limited temporal and spatial resolution in comparison to amperometric methods such as the MEA, which can continuously record glutamate events in multiple, microscopic regions at up to 10 Hz23. This rapid sampling rate eliminates the confounding factor of localized neurotransmitter diffusion that is inherent to slow-sampling methods like microdialysis24. Additionally, the MEA is a less invasive method than a microdialysis probe, which can cause significant gliosis during insertion and may alter neurotransmitter activity at the insertion site22.
Previous studies utilizing a range of mammalian models, measurement techniques, and regions of the brain, have demonstrated basal glutamate levels comparable to those found using this technique. This suggests that MEA technology, when adapted to the piglet model, provides valid recordings of in vivo glutamate concentration (Table 2).
Author (Year) | Recording Technique | Animal Model | Age | Brain Region(s) | Mean Basal Glutamate Concentration (µM) |
Hascup et al. (2008)23 | MEA (Enzyme-based) | Rodent | 20 – 24 weeks | Prefrontal Cortex, Striatum | 3.3 ± 1.0; 5.0 ± 1.2 |
Hascup et al. (2010)25 | MEA (Enzyme-based) | Rodent | 3 – 6 months | Hippocampus | 4.7 – 10.4 |
Rutherford et al. (2007)9 | MEA (Enzyme-based) | Rodent | 3 – 6 months | Prefrontal Cortex, Striatum | 44.9 ± 4.7; 7.3 ± 0.9 |
Miele et al. (1996)26 | Microdialysis (Enzyme-based) | Rodent | – | Striatum | 3.6 ± 0.5 |
Day et al. (2006)27 | MEA (Enzyme-based) | Rodent | 3 – 6 months | Frontal Cortex, Striatum | 1.6 ± 0.3 ;1.4 ± 0.2 |
Quintero et al. (2007)28 | MEA (Enzyme-based) | Non-Human Primate | 5.3 – 5.5 years | Premotor Cortex, Motor Cortex | 3.8 ± 1.7; 3.7 ± 0.9 |
Stephens et al. (2010)29 | MEA [Spencer-Gerhardt-2 (SG-2)] | Non-Human Primate | 11 – 21 years | Putamen | 8.53 |
Kodama et al. (2002)30 | Microdialysis (Enzyme-based) | Non-Human Primate | – | Prefrontal Cortex | 1.29 – 2.21 |
Galvan et al. (2003)31 | Microdialysis (Enzyme-based) | Non-Human Primate | Juvenile | Striatum | 28.74 ± 2.73 |
During and Spencer (1993)32 | Microdialysis (Enzyme-based) | Human | 18 – 35 years | Hippocampus | 20.3 ± 6.6 |
Reinstrup et al. (2000)33 | Microdialysis (Enzyme-based) | Human | – | Frontal Cortex | 16 ± 16 |
Cavus et al. (2005)34 | Microdialysis (Enzyme-based) | Human | 15 – 52 years | Neocortex | 2.6 ± 0.3 |
Table 2. Comparison of basal extracellular glutamate levelsacross various animal models. A selected review of studies establishing normal extracellular glutamate levels in healthy awake and anesthetized animals using microdialysis or microelectrodes.
The use of MEA technology to monitor in vivo glutamate concentrations in the piglet model can allow for the future evaluation of piglet neurological outcomes post-anesthesia. Survival experiments have been planned, which will further an understanding of the long-term impact of anesthesia on the neurocognitive well-being of human neonates. Survival experiments will allow for behavioral testing and monitoring of glutamate changes long after anesthesia exposure. It is also common for children to undergo anesthesia in conditions where they might experience physiological stress in the form of surgical intervention. Future studies addressing the influence of surgery in terms of neurological injury and increase in neurotoxicity would allow for more accurate modeling of a common clinical setting for children. The use of alternate animal models is also feasible, as is the study of these various models through chronic implantation, allowing us to track behavioral changes associated with neurotoxicity. MEA technology itself is versatile, so future study need not be limited to analysis of glutamate levels (e.g., GABA, choline, lysine, etc. could be analyzed).
The authors have nothing to disclose.
The authors would like to acknowledge the contributions of the University of Kentucky Center for Microelectrode Technology (CenMeT), and the Ohio State University Laboratory Animal Resource Center (ULAR).
Advance Liqui-Wean Pig Milk Replacer | PBS Animal Health | 292-13 | |
Piglet Anesthesia Face-Cone Mask | VetEquip | 921428 | |
Integra SL Anesthesia Workstation | DRE Veterinary | 2350 | This anesthesia workstation is chosen to best mimic the clinical monitoring experienced by pediatric patients in the operating room. Any anesthesia machine can be used as long as it allows for sufficient physiologic monitoring and intervention. |
Sevoflurane | Ultane | 0074-4456-04 | |
Rocuronium Bromide Injection | Hospira | 0409-9558-05 | |
Medfusion 4000 IV Infusion | Smiths Medical | ||
Model 1530 Heavy-Duty Research Model Stereotax | Kopf | custom made | |
Model 1541 Piglet Adaptor | Kopf | custom made | |
Infrared Spot Lamp | Amazon | B000HHQ94C | |
Bair Hugger Torso Blanket | 3M | 540 | |
Bair Hugger | 3M | 750 | |
Sterile Alcohol Prep Pad | Fisherbrand | 22-363-750 | |
Carbon Steel Rib-Back Surgical Blade | Bard-Parker | #10 | |
Scalpel Handel | Havel's | HAN-G4 | |
Surgical Scissors | World Precision Instruments | 504615 | |
Mosquito Forceps | Sklar Surgical Instruments | 17-1225 | |
Gauze Pads | Fisherbrand | 22-246-069 | |
Adson Tissue Forceps | Teleflex | 181223 | |
Dremel 111 Engraving Cutter | Amazon | Dremel 111 | |
Microelectrode Array | Center for Microelectrdoe Technology, University of Kentucky | S2 4Ch MEA; custom made | |
Headstage | Quanteon | 2pA/mV | |
Wire, silver, PFA, .008" Bare, .0110" coated | A-M Systems | 786500 | |
Fine Micromanipulator | Narishige Scientific Instrument Lab | MO-8 |