Here, it is demonstrated how an awake closed-head injury model can be used for examining the effects of repeated mild traumatic brain injury (r-mTBI) on synaptic plasticity in the hippocampus. The model replicates important features of r-mTBI in patients and is used in conjunction with in vitro electrophysiology.
Mild traumatic brain injuries (mTBIs) are a prevalent health issue in North America. There is increasing pressure to utilize ecologically valid models of closed-head mTBI in the preclinical setting to increase translatability to the clinical population. The awake closed-headed injury (ACHI) model uses a modified controlled cortical impactor to deliver closed-headed injury, inducing clinically relevant behavioral deficits without the need for a craniotomy or the use of an anesthetic.
This technique does not normally induce fatalities, skull fractures, or brain bleeds, and is more consistent with being a mild injury. Indeed, the mild nature of the ACHI procedure makes it ideal for studies investigating repetitive mTBI (r-mTBI). Growing evidence indicates that r-mTBI can result in a cumulative injury that produces behavioral symptoms, neuropathological changes, and neurodegeneration. r-mTBI is common in youths playing sports, and these injuries occur during a period of robust synaptic reorganization and myelination, making the younger population particularly vulnerable to the long-term influences of r-mTBI.
Further, r-mTBI occurs in cases of intimate partner violence, a condition for which there are few objective screening measures. In these experiments, synaptic function was assessed in the hippocampus in juvenile rats that had experienced r-mTBI using the ACHI model. Following the injuries, a tissue slicer was utilized to make hippocampal slices to evaluate bidirectional synaptic plasticity in the hippocampus at either 1 or 7 days following the r-mTBI. Overall, the ACHI model provides researchers with an ecologically valid model to study changes in synaptic plasticity following mTBI and r-mTBI.
Traumatic brain injury (TBI) is a significant health issue, with ~2 million cases in Canada and the United States every year1,2. TBI affects all age groups and genders and has an incidence rate greater than any other disease, notably including breast cancer, AIDS, Parkinson's disease, and multiple sclerosis3. Despite the prevalence of TBI, its pathophysiology remains poorly understood, and treatment options are limited. In part, this is because 85% of all TBIs are classified as mild (mTBI), and mTBI has previously been thought to produce only limited and transient behavioral changes with no long-term neuropsychiatric consequences4,5. It is now recognized that mTBI recovery can take weeks to years5,6, precipitate more serious neurological conditions4, and that even repeated "sub-concussive" impacts affect the brain7. This is alarming as athletes in sports such as hockey/football have >10 head sub-concussive impacts per game/practice session7,8,9,10.
Adolescents have the highest incidence of mTBI, and in Canada, roughly one in 10 teens will seek medical care for a sport-related concussion annually11,12. In reality, any sub-concussive head impact or mTBI can cause diffuse damage to the brain, and this could also create a more vulnerable state for subsequent injuries and/or more serious neurological conditions13,14,15,16,17. In Canada, it is recognized legally via Rowan's law that prior injury can increase the vulnerability of the brain to further injury18, but mechanistic understanding of r-mTBI remains woefully inadequate. It is clear, however, that single and r-mTBI can impact learning capacity during school years19,20, have sex-specific outcomes21,22,23,24, and impair cognitive capacity later in life16,25,26. Indeed, cohort analyses strongly associate r-mTBI early in life with dementia later on27,28. r-mTBI is also potentially associated with chronic traumatic encephalopathy (CTE), which is characterized by the accumulation of hyperphosphorylated tau protein and progressive cortical atrophy and precipitated by significant inflammation27,29,30,31. Although the links between r-mTBI and CTE are currently controversial32, this model will allow them to be explored in greater detail in a preclinical setting.
An mTBI is often described as an "unseen injury," as it occurs within a closed skull and is difficult to detect even with modern imaging techniques33,34. An accurate experimental model of mTBI should adhere to two tenets. First, it should recapitulate the biomechanical forces normally observed in the clinical population35. Second, the model should induce heterogeneous behavioral outcomes, something that is also highly prevalent in clinical populations36,37,38. Currently, the majority of preclinical models tend to be more severe, involving craniotomy, stereotaxic head restraint, anesthesia, and controlled cortical impacts (CCI) that produce significant structural damage and more extensive behavioral deficits than normally observed clinically33. Another concern with many preclinical models of concussion that involve craniotomies is that this procedure itself creates inflammation in the brain, and this can exacerbate mTBI symptoms and neuropathology from any subsequent injury39,40. Anesthesia also introduces several complex confounds, including reducing inflammation41,42,43, modulating microglial function44, glutamate release45, Ca2+ entry through NMDA receptors46, intracranial pressure, and cerebral metabolism47. Anesthesia further introduces confounds by increasing blood-brain barrier (BBB) permeability, tau hyperphosphorylation, and corticosteroid levels, while reducing cognitive function48,49,50,51. Additionally, diffuse, closed-headed injuries represent the vast majority of clinical mTBIs52. They also allow one to better study the multitude of factors that can influence behavioral outcomes, including sex21, age53, inter-injury-interval15, severity54, and the number of injuries23.
The direction of the accelerative/decelerative forces (vertical or horizontal) is also an important consideration for behavioral and molecular outcomes. Research from Mychasiuk and colleagues have compared two models of diffuse closed-headed mTBI: weight-drop (vertical forces) and lateral impact (horizontal forces)55. Both the behavioral and molecular analyses revealed heterogeneous model- and sex-dependent outcomes following mTBI. Thus, animal models that help avoid surgical procedures, while incorporating linear and rotational forces, are more representative of the physiological conditions under which these injuries normally occur33,56. The ACHI model was created in response to this need, allowing for the rapid and reproducible induction of mTBI in rats while avoiding procedures (i.e., anesthesia) that are known to bias sex differences57.
Approval for all animal procedures was provided by the University of Victoria Animal Care Committee in compliance with Canadian Council on Animal Care (CCAC) standards. All male Long-Evans rats were bred in-house or purchased (see the Table of Materials).
1. Housing and breeding conditions
2. Setup of awake closed-head injury procedure
3. Induction of mTBI
4. Induction of sham injury
5. Neurological assessment protocol
NOTE: The NAP can be used to measure the level of consciousness, as well as cognitive and sensorimotor functioning.
6. Slice preparation
NOTE: In the current study, synaptic plasticity was assessed in animals following r-mTBI at either 1 or 7 days after mTBI. On these days, the animals were brought individually into the laboratory in covered cages prior to sacrifice.
7. Field electrophysiology
NOTE: To acquire extracellular field recordings from the dentate gyrus (DG), perform the following steps. Following the 60 min recovery, individual hippocampal slices are ready for extracellular field recordings.
The awake closed-head injury model is a viable method of inducing r-mTBI in juvenile rats. Rats exposed to r-mTBI with the ACHI model did not show overt behavioral deficits. Subjects in these experiments did not exhibit latency to right or apnea at any point during the r-mTBI procedure, indicating that this was indeed a mild TBI procedure. Subtle behavioral differences did emerge in the NAP; as described above, the rats were scored on four sensorimotor tasks (startle response, limb extension, beam walk, and rotating beam) on a scale from 0 to 3, with 3 representing no impairment with the task. Thus, the lower the NAP score, the more impaired the animal was. At baseline, there were no differences in the NAP scores between sham and r-mTBI rats. Following all ACHI sessions, the r-mTBI rats showed significant impairments within the NAP tasks when compared to shams (Figure 4). However, as reported previously for impacts delivered over multiple days (i.e., 2 or 4 days), the subsequent addition of injuries over the course of the day did not compound or produce additional behavioral deficits. Thus, the ACHI model of r-mTBI produces subtle, yet significant, behavioral deficits during these acute post injury time points.
Following the injury protocol, evoked field responses and synaptic plasticity were examined in the MPP input to the DG of the hippocampus on post injury day 1 (PID1), and PID7. Slice health was examined using fEPSPs in response to an ascending series of pulse widths in each slice. As is shown in Figure 3C, there was no difference in the input-output curves generated in slices obtained from sham and r-mTBI rats. To examine presynaptic transmitter release, a series of paired pulses (50 ms interpulse interval) were administered, and the ratio of the size of the second fEPSP was calculated relative to the first fEPSP. The paired-pulse ratios did not differ between sham and r-mTBI rats (Figure 3D). Thus, these data indicate that r-mTBI did not alter basic synaptic physiology in the MPP input to the DG. To examine LTD, a 10 Hz LTD protocol was administered to induce an LTD dependent on endocannabinoids64. On PID1, there was a significant decrease in the capacity of the MPP input to the DG to sustain LTD (Figure 3E). This reduction in LTD was transient, however, and by PID7, slices from sham and r-mTBI animals displayed equivalent LTD (Figure 3F), although there was an indication of a slight trend for slices from r-mTBI animals to exhibit an increase in LTD.
Figure 1: The ACHI procedure setup used to model r-mTBI. (A) A modified controlled cortical impactor was used to rapidly displace the animal's head 10 mm at a velocity of 6.0 m/s. (B,C) Custom 3D printed helmet with a left parietal cortex target site. (D) Subjects were placed in a plastic restraint bag on a foam platform, with the helmet placed around the restraint cone and positioned so the target site is directly under the impactor tip. Abbreviations: ACHI = awake closed-head injury; r-mTBI = repeated mild traumatic brain injury. Please click here to view a larger version of this figure.
Figure 2: Materials and setup required for slice preparation. (A) Tools used for brain extraction, mounting, slicing, and incubation: (a) culture dish with filter paper; (b) various dissection tools, including standard scissors, dissecting scissors, forceps, a rongeur, and spatulas; (c) tissue adhesive; (d) Compresstome piston and specimen tube; (e) feather blade and blade holder; (f) chilling block; (g) slice incubation chamber. (B) Compresstome tissue slicer. (C) Slices incubating in a bath containing artificial cerebrospinal fluid that is continuously oxygenated with 95% O2/5% CO2. Please click here to view a larger version of this figure.
Figure 3: Acute impairments in synaptic plasticity in juvenile male rats due to r-mTBI using the ACHI model. (A) The major hippocampal pathways. The medial perforant pathway is comprised of the input from the entorhinal cortex into the dentate gyrus (blue). The medial perforant path inputs synapse onto granule cells in the dentate gyrus (purple). (B) Brightfield photomicrograph of a hippocampal brain slice (4x magnification), showing the actual placement of a bipolar stimulating electrode (left) and a glass recording electrode pipette (right) in the medial performant path of the dentate gyrus. (C) Input-output plot (fEPSP slope) for different simulation intensities (10-300 µs) on PID1 and PID7 for sham and r-mTBI rats. (D) Paired-pulse ratios for sham and r-mTBI rats (50 ms interpulse interval). (E) Time course of fEPSP changes prior to, and following, administration of an LTD induction paradigm in hippocampal slices obtained from sham and r-mTBI rats at PID1. (F) Time course of fEPSP changes prior to, and following, administration of an LTD induction paradigm in hippocampal slices obtained from sham and r-mTBI rats at PID7. Abbreviations: ACHI = awake closed-head injury; r-mTBI = repeated mild traumatic brain injury; PID = post injury day; fEPSP = field excitatory postsynaptic potential. Please click here to view a larger version of this figure.
Figure 4: Acute neurological impairment in juvenile male rats due to r-mTBI using the ACHI model. The rats underwent eight ACHI procedures at 2 h intervals over 1 day, with a neurological assessment protocol conducted at baseline and after each injury. The NAP consisted of four tasks: startle response, limb extension, beam walk, and rotating beam. Each task was scored out of 3, giving a total possible score of 12 for each session. Data presented as mean ± SEM. (*) indicates p < 0.05. Abbreviations: ACHI = awake closed-head injury; r-mTBI = repeated mild traumatic brain injury; NAP = neurological assessment protocol. Please click here to view a larger version of this figure.
Supplementary Table S1: ACHI procedure animal and impact information. Abbreviation: ACHI = awake closed-head injury. Please click here to download this File.
Supplementary Table S2: Restraint scoring for awake mTBI. Abbreviation: mTBI = mild traumatic brain injury. "Turn around in restraint" refers to the researcher placing the animal in the restraint, before closing the bag around the tail. After the bag is closed, the animal should not be able to turn around. Vocalization and squirming should be scored after the bag is closed. Please click here to download this File.
Supplementary File 1: Cage-side monitoring checklist. Please click here to download this File.
Supplementary File 2: Pain scale and advanced monitoring checklist. Please click here to download this File.
Most preclinical research has utilized models of mTBI that do not recapitulate the biomechanical forces seen in the clinical population. Here, it is shown how the ACHI model can be used to induce r-mTBIs in juvenile rats. This closed-headed model of r-mTBI has significant advantages over more invasive procedures. First, the ACHI does not normally cause skull fractures, brain bleeds, or fatalities, all of which would be contraindications of a "mild" TBI in clinical populations61. Second, the ACHI does not require the use of craniotomies, which is significant because they are known to cause inflammatory responses that can exacerbate symptomologies and neuropathology67. Third, the ACHI does not require the use of anesthesia. This is also significant, as anesthesia can have neuroprotective properties and can impair synaptic plasticity, in addition to learning and memory performance48,49,50,51,68. Finally, the ACHI can produce subtle transient changes in neurological function that can be assessed immediately post injury.
As the ACHI does not normally induce loss of consciousness or apnea, this model mimics mTBI in a significant proportion of the clinical population69,70,71. Despite this, the ACHI model produced a significant reduction in the NAP scores. This reduction persisted with repeated administrations of the ACHI procedure but did not exacerbate sensorimotor impairments within the r-mTBI group. This indicates that the ACHI model induces a mild injury analogous to that observed following concussive or sub-concussive head impacts in clinical populations72,73. A primary advantage of the NAP is the detection of subtle behavioral deficits seen in the acute timeframe following r-mTBI. This quick examination may allow researchers to categorize rats based on their behavioral responses. However, the use of more robust behavioral tests at subacute and chronic time points may be necessary to detect motor, cognitive, and affective symptomologies74,75,76. It is important to note that while there were no differences in NAP scores over the eight injuries, rodent behavior can be influenced by changes in environment and familiarity with the experimenter77,78. Rats should be allowed to acclimate to the procedure room prior to administering r-mTBI or sham injuries. In addition, it is important for one individual to be responsible for administering the impacts to help ensure consistency.
Despite the previously mentioned benefits of the ACHI model, it is not without limitations. First, the paradigm was designed to mimic the cumulation of impacts in a single session and not repetitive injuries following a recovery period. Following injury, the brain resides in a window of cerebral vulnerability that extends from 1 to 5 days post injury in rodents15,79,80. Receiving eight injuries on a singular day does not allow for acute and subacute injury cascades to develop. Therefore, depending on the research question of interest, the injury paradigm may need to be adjusted within the window of vulnerability. Second, while it is beneficial to limit the use of anesthetic, an unintended consequence of the ACHI model is subjecting the rats to restraint stress. It has been shown that exposure to acute and chronic stressors may initiate an inflammatory response, influence a variety of behaviors, and alter synaptic plasticity in the hippocampus81,82,83.
The protocol described above provides a clear-cut method to produce high-quality transverse hippocampal slices from r-mTBI-administered animals with the ACHI model. Additionally, the protocol allows for stable electrophysiological recordings and shows that the hippocampus is still capable of exhibiting synaptic plasticity following r-mTBI, although there may be transient disruptions. With any electrophysiological recordings, slice health is paramount for the ability to record suitable fEPSPs. To preserve brain tissue, prior to slicing, it is imperative that the brain remains ice-cold in carbogenated aCSF. The brain's removal and slicing should be done quickly, but not if this comes at the expense of care. This protocol on juvenile animals utilizes aCSF as cutting solution, but depending on the age of the animal, protective cutting solutions (such as choline-, sucrose-, NMDG-, or glycerol-based solutions) may be required84,85,86.
Field electrophysiological recordings allow researchers to gauge hippocampal synaptic plasticity. However, there are a number of limitations to the technique. The process of slicing the brain has shown to cause changes in spine numbers87, which could affect synaptic plasticity. The use of in vivo recordings would preserve pathways and allow for measurement of synaptic plasticity in anesthetized or live animals88. Additionally, the use of field recordings probes the properties of groups of neurons but does not inform about changes in individual neurons. The use of whole-cell patch-clamp recordings can give temporally detailed information about neuronal properties in response to pharmacological or optogenetic manipulations89. Additionally, the combination of electrophysiological recordings with complementary techniques, such as calcium imaging, Western blotting, immunohistochemistry, or electron microscopy, would allow researchers to gain insight into the mechanisms of action.
Cognitive deficits are commonly reported following r-mTBI, and the current protocol can help investigate some of the underlying physiological processes associated with these deficits. In particular, the mild nature of the ACHI procedure opens up the possibility of examining changes in synaptic physiology across the lifespan of animals that have incurred r-mTBI. The ACHI model appears to be an ecologically valid model of mTBI than can be used to study r-mTBI. Preliminary studies utilizing the ACHI model have showed acute neurological impairment without overt structural damage, administering one, four, and eight repeated injury paradigms61,90. Future studies will examine how r-mTBI can impact synaptic plasticity during developmental periods and in the aging brain. By better understanding the pathophysiology of mTBI and r-mTBI for synaptic function, the hope is to better direct potential therapeutic interventions to help reduce cognitive function.
The authors have nothing to disclose.
We thank all the members of the Christie Laboratory at the University of Victoria, past and present, for their contributions to the development of this protocol. This project was supported with funds from the Canadian Institutes for Health Research (CIHR: FRN 175042) and NSERC (RGPIN-06104-2019). The Figure 1 skull graphic was created with BioRender.
3D-printed helment | Designed and constructed by Christie laboratory (See Specifications in Christie et al. (2019), Current Protocols in Neuroscience) | ||
Agarose | Fisher Scientific (BioReagents) | BP160500 | |
Anesthesia chamber | Home Made | N/A | Plexiglass Container |
Automatic Heater Controller | Warner Electric | TC-324B | |
Axon Digidata | Molecular Devices | 1440A | Low-noise Data Acquisition System |
Balance beam | Can be constructed or purchased (100 cm long x 2 cm wide x 0.75 cm thick) | ||
Calcium Chloride | Bio Basic Canada Inc. | CD0050 | For aCSF |
Camera | Dage MTI | NC-70 | |
Carbogen tank | Praxair | MM OXCD5C-K | Carbon Dioxide 5%, Oxygen 95% |
Clampex Software | Molecular Devices | Clampex 10.5 Version | |
Compresstome Vibrating Microtome | Precisionary | VF 310-0Z | |
Concentric Bipolar Electrode | FHC Inc. | CBAPC75 | |
Dextrose (D-Glucose) | Fisher Scientific (Chemical) | D16-3 | aCSF |
Digital Stimulus Isolation Amplifier | Getting Instruments, Inc. | Model 4D | |
Disodium Phosphate | Fisher Scientific (Chemical) | S373-500 | PBS |
Dissection Tools | |||
Feather Double Edge Blade | Electron Microscopy Sciences | 72002-10 | |
Filter Paper | Whatman 1 | 1001-055 | |
Flaming/Brown Micropipette Puller | Sutter Instrument | P-1000 | |
Hair Claw Clip | Can be obtained from any department store | ||
Home and Recovery Cages | Normal rat cages from animal care unit. | ||
Hum Bug Noise Eliminator | Quest Scientific | 726300 | |
Isoflurane USP | Fresenius Kabi | CP0406V2 | |
Isotemp 215 Digital Water Bath | Fisher Scientific | 15-462-15 | |
Leica Impact One CCI unit | Leica Biosystems | Tip is modified to hold 7mm rubber impact tip | |
Long-Evans rats, male | Charles River Laboratories (St. Constant, PQ) | ||
Low-Density Foam Pad | 3" polyurethane foam sheet | ||
Magnesium Chloride | Fisher Scientific (Chemical) | M33-500 | aCSF |
Male Long Evans Rats | Charles River Laboratories | Animals ordered from Charles River Laboratories, or pups bred at the University of Victoria | |
MultiClamp 700B Amplifier | Molecular Devices | Model 700B | |
pH Test Strips | VWR Chemicals BDH | BDH83931.601 | |
Potassium Chloride | Fisher Scientific (Chemical) | P217-500 | aCSF, PBS |
Potassium Phosphate | Sigma | P9791-500G | PBS |
Push Button Controller | Siskiyou Corporation | MC1000e | Four-axis Closed Loop Controller Push-Button |
Sample Discs | ELITechGroup | SS-033 | For use with Vapor Pressure Osmometer |
Small towel | |||
Sodium Bicarbonate | Fisher Scientific (Chemical) | S233-500 | aCSF |
Sodium Chloride | Fisher Scientific (Chemical) | S271-3 | For aCSF, PBS |
Sodium Phosphate | Fisher Scientific (Chemical) | S369-500 | aCSF |
Soft Plastic Restraint Cones | Braintree Scientific | model DC-200 | |
Stopwatch | Many lab members use their iPhone for this | ||
Table or large cart with raised edges | For NAP and ACHI | ||
Thin Wall Borosilicate Glass (with Filament) | Sutter Instrument | BF150-110-10 | Outside diameter: 1.5 mm; Inside diameter: 1.10 mm; Length: 10 cm |
Upright Microscope | Olympus | Olympus BX5OWI | 5x MPlan 0.10 NA Objective lens |
Vapor Pressure Osmometer | Vapro | Model 5600 | aCSF should be 300-310 mOSM |
Vetbond Tissue Adhesive | 3M | 1469SB | |
Vibraplane Vibration Isolation Table | Kinetic Systems | 9101-01-45 |