The protocol describes mild traumatic brain injury in a mouse model. In particular, a step-by-step protocol to induce a mild midline closed head injury and the characterization of the animal model is fully explained.
Highly reproducible animal models of traumatic brain injury (TBI), with well-defined pathologies, are needed for testing therapeutic interventions and understanding the mechanisms of how a TBI alters brain function. The availability of multiple animal models of TBI is necessary to model the different aspects and severities of TBI seen in people. This manuscript describes the use of a midline closed head injury (CHI) to develop a mouse model of mild TBI. The model is considered mild because it does not produce structural brain lesions based on neuroimaging or gross neuronal loss. However, a single impact creates enough pathology that cognitive impairment is measurable at least 1 month after injury. A step-by-step protocol to induce a CHI in mice using a stereotaxically guided electromagnetic impactor is defined in the paper. The benefits of the mild midline CHI model include the reproducibility of the injury-induced changes with low mortality. The model has been temporally characterized up to 1 year after the injury for neuroimaging, neurochemical, neuropathological, and behavioral changes. The model is complementary to open skull models of controlled cortical impact using the same impactor device. Thus, labs can model both mild diffuse TBI and focal moderate-to-severe TBI with the same impactor.
Traumatic brain injury (TBI) is caused by an external force on the brain, often associated with falls, sports injuries, physical violence, or road accidents. In 2014, the Centers for Disease Control and Prevention determined that 2.53 million Americans visited the emergency department to seek medical help for TBI-related accidents1. Since mild TBI (mTBI) represents the majority of TBI cases, over the past several decades, multiple models of mTBI have been adopted, which include weight drop, piston-driven closed head injury and controlled cortical impact, rotational injury, mild fluid percussion injury, and blast injury models2,3. The heterogeneity of the mTBI models is useful to address the different features associated with mTBI seen in people and to help evaluate the cellular and molecular mechanisms associated with brain injury.
Of the commonly used models of closed head injury, one of the first and most widely used models is the weight drop method, where an object is dropped from a specific height onto the animal's head (anesthetized or awake)2,4. In the weight drop method, the injury's severity depends on several parameters, including craniotomy performed or not, head fixed or free, and the distance and weight of the falling object2,4. One disadvantage of this model is the high variability in the severity of the injury and the high mortality rate associated with respiratory depression5,6. A common alternative is to deliver the impact using a pneumatic or electromagnetic device, which can be done directly on the exposed dura (controlled cortical impact: CCI) or closed skull (closed head injury: CHI). One of the strengths of the piston-driven injury is its high reproducibility and low mortality. However, CCI requires craniotomy7,8, and a craniotomy itself induces inflammation9. Instead, in the CHI model, there is no need for craniotomy. As already stated, each model has limitations. One of the limitations of the CHI model described in this paper is that the surgery is performed using a stereotaxic frame, and the head of the animal is immobilized. While the full head immobilization assures reproducibility, it does not account for movement after the impact that could contribute to the injury associated with a mTBI.
This protocol describes a basic method to perform a CHI impact with a commercially available electromagnetic impactor device10 in a mouse. This protocol details the exact parameters involved to achieve a highly reproducible injury. In particular, the investigator has precise control over the parameters (depth of injury, dwell time, and velocity of impact) to precisely define the injury severity. As described, this CHI model produces an injury that results in bilateral pathology, both diffuse and microscopic (i.e., chronic activation of glia, axonal and vascular damage), and behavioral phenotypes11,12,13,14,15. In addition, the described model is considered mild as it does not induce structural brain lesions based on MRI or gross lesions on pathology even 1 year after the injury16,17.
The experiments performed were approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Kentucky, and both the ARRIVE and the Guide for the Care and Use of Laboratory Animals guidelines were followed during the study.
1. Surgical setup
NOTE: Mice are housed in groups of 4-5/cage, humidity in the housing room is maintained at 43%-47%, and the temperature is maintained at 22-23 ˚C. Mice are given ad libitum access to food and water and exposed to a 12 h/12 h light/dark cycle (7 a.m./7 p.m.).
2. Pre-surgery procedure
3. Surgical procedure
4. Post-surgery care
5. Cleaning
6. Exclusion criteria
This stereotaxic electromagnetic impactor device is versatile. It is used for both an open skull controlled cortical impact (CCI) or a closed head injury (CHI) surgery. Furthermore, the injury severity can be modulated by changing the injury parameters such as impact velocity, dwell time, impact depth, impactor tip, and injury target. Herein is described a CHI surgery using a 5.0 mm steel tip impactor. This injury is considered mild because there are no structural brain lesions. The mortality rate in adult mice is less than 0.9%11,14 and increases slightly to reach ~2.5% in older mice (>8 months old)11. Mortality occurs during the first 2 min because of apnea, which can largely be prevented by carefully monitoring the depth of anesthesia in the seconds before the impact.
The advantage of this CHI model is that the impact produces bilateral diffuse pathology without needing to expose the cortical dural surface (craniotomy). Another feature that makes this an effective TBI model is that less than 1% of mice are excluded from the study due to skull fractures or ear issues following the surgical procedure. Importantly the model produces neuropathological and behavioral impairments with a single impact, which reduces the experimental complexity associated with repetitive mild CHI models15. For instance, a reproducible temporal pattern of microglia and astrocyte morphological changes are identified11 (Figure 2A,B). When validating the model, it is recommended to use the starting ranges of the anterior-posterior coordinates as −1.5 mm ± 0.2 mm and the impact depth as 1.0 ± 0.2 mm. The coordinates may need to be adjusted for the age and strain of the mice, as well as the brand and model of the equipment used. Once the settings have been validated, they should be held constant for an experiment. For validation, the neuropathological characterization of microglia and astrocytes at 3 days post-injury is recommended. Immunohistochemical (IHC) staining was completed following the methods in Bachstetter et al.18. Specifically, 30 μm coronal free-floating sections were stained for glial activation with rabbit anti-GFAP (1:10,000) and for astrocytes using a rabbit anti-IBA1 (1:10,000). An HRP conjugated goat anti-rabbit IgG (1:200) was used to detect both GFAP and IBA-1. Quantification software was used to quantify the staining in each region considered. In addition, at 1-day post-injury, axonal injury markers were found in the neocortex, and changes in mitochondrial metabolism were found by 28 days post-CHI16 (data not shown).
The secondary endpoints for validating the model would be behavioral assays. Reproducible CHI-induced deficits in the radial arm water maze (RAWM)12 and active avoidance13 behaviors were found (Figure 3). The mice were tested in an 8-arm RAWM, a special learning test, as described in Macheda et al.12. Briefly, the mice were tested in a total of 28 trials over a 4 day protocol and had 60 s to locate the platform positioned in the target arm. The total number of trials per day was seven; day 1 and day 2 were considered as training days and day 3 and 4 as testing days. During the training days, the mice were trained to locate the platform, alternating between visible and hidden trials; during the testing days, the platform was hidden during all trials. The experiments were recorded using a camera, and a tracking system was used for behavior analysis (number of errors, total distance, and latency). The mice were tested 2 weeks post-injury. While there was no effect of sex, the CHI mice made more errors to successfully perform the task and reach the platform (Figure 3A). Furthermore, memory impairments have been detected in a 6-arm RAWM test11,14,15,16 as well. Active avoidance, an associative learning-based test, has been used to measure the cognitive deficits associated with this mild model of CHI. The mice were tested using a 5 day protocol and exposed to 50 trials/day13. The mice were trained to avoid a mild foot shock (unconditioned stimulus, US) by associating a conditioned stimulus (CS, light) with it. Over time, the mice learned to avoid the US when the CS was presented. The CHI mice had impaired cognitive function in active avoidance compared to sham mice (Figure 3B). The sham female mice learned significantly faster compared to the males, but the sex did not play a role in CHI mice13. Behavior was recorded using active/passive avoidance software. A reproducible deficit in motor function beyond the first week after the injury has not been detected11.
In this mild TBI model, no gross structural lesions to the brain were found, and a single impact induced bilateral glial activation and changes in microglia morphology. Also, cognitive deficits are associated with this TBI model.
Figure 1: Step 1: Surgical area setup. (A) An example of the surgical area and tools needed to perform CHI surgery (ice pack for the impactor, stereotaxic frame equipped with the impactor, impactor control box, and surgical tools) is shown. (B) A close-up view of the 5 mm steel probe tip, bite bar, and head support apparatus, which illustrates the positioning needed for the midline impact. (C) The head support apparatus is made from a 1 mL latex pipette bulb attached to the tubing by parafilm. A 10 mL syringe is filled with water to inflate the bulb, with a stopcock to keep the bulb inflated once in position. (D) Impactor control box: (1) a large knob to adjust the impact velocity, (2) a dwell counter, (3) an extend/retract toggle switch, (4) a toggle switch that, when pushed down, will deliver the impact. (E) When not in use, the impactor is kept on an ice pack to prevent overheating and possible malfunctioning. (F) A digital stereotaxic display is used for establishing the x (anterior-posterior), y (medial-lateral), and z (dorsal-ventral) coordinates. Step 2: Surgical procedure. (G,H) The anesthetized and shaved mouse is secured into the stereotaxic frame, (I) a midline incision is made to expose the (J) bregma, (K) which is used during surgery to line up the impactor. Step 3: Recovery. (L) The mouse is removed from the stereotaxic frame. After the scalp is closed by stapling or suturing the skin together, it is placed in a clean recovery cage on its side. (M) The mouse is monitored until the mouse rolls over and the righting reflex occurs. Please click here to view a larger version of this figure.
Figure 2: The temporal patterns of astrocyte (GFAP) and microglia (IBA1) morphological changes after a CHI. (A) GFAP staining at low magnification shows the regional increase in staining seen in the cortex of the CHI group. The morphological appearance of the astrocytes is shown in the higher-magnification insets, which were taken from the middle brain sections and from the same regions of the cortex. (B) IBA1-positive staining in the cortex at 1 day, 7 days, and 2 months post-injury shows changes in microglia morphology in the neocortex after the CHI (n = 7-14, 50/50 male/female). The mice (CD-1/129 background) were 8 months old at the time of surgery. This figure has been adapted from 11 and reproduced with permission. Scale bar = 1 mm, 50 µm and 100 µm as indicated in the figure. Please click here to view a larger version of this figure.
Figure 3: CHI-induced memory deficits in RAWM and active avoidance. (A) At 2 weeks post-injury, both the CHI- and sham-operated mice were able to learn the RAWM task, but the CHI mice made more errors compared to the sham mice (*** p < 0.0005); sham (n = 20/20 male/female); CHI (n = 20/20 male/female). The mice (C57BL/6J) were 3-4 months old at the time of surgery. (B) At 4 weeks post-injury, the CHI and sham-operated mice were able to learn the active avoidance task, but the CHI mice avoided fewer foot-shocks compared to the sham mice (*** p = 0.0005; **** p < 0.0001); sham (n = 10/10 male/female); CHI (n = 9/10 male/female). The mice (C57BL/6J) were 3-5 months old at the time of surgery. Data are shown as mean ± SEM. (A) This figure has been adapted from 12 and reproduced with permission. (B) This figure has been adapted from 13 and reproduced with permission. Please click here to view a larger version of this figure.
Several steps are involved in recreating a consistent injury model using the described model. First, it is critical to correctly secure the animal into the stereotaxic frame. The animal’s head should not be able to move laterally, and the skull should be completely flat with bregma and lambda reading the same coordinates. Correctly placing the ear bars is the most difficult aspect of this surgery, and this can only be learned with practice. If the skull is not level, the head should be adjusted before inducing CHI. Failure to adjust the head positioning will cause a skull fracture. To evaluate that the skull is flat, one should look at the gap between the skull and the impact tip from all angles around the tip. Mice with depressed skull fractures should be excluded from experiments, as they have a much stronger inflammatory response and a more severe injury compared to mice that did not suffer skull fractures19. Additionally, mice with skull fractures show more severe TBI outcomes, such as post-traumatic respiratory depression, secondary rebound injury, and eventually death20.
In this study, the animal’s head was secured with ear bars. In particular, only mouse-specific acetal resin ear bars with a tapered point are recommended to be used, not large rat ear bars. It is possible to use non-puncture rubber-tipped ear bars, but these ear bars will compress the skull, altering the biomechanics of the CHI, and are less reproducible. In addition, there is a limitation to using ear bars, as it does not allow for any rotational forces. Nevertheless, the greater reproducibility of the ear bars outweighs the limited number of rotational forces that can be generated if the head is unfixed.
However, fixing the head with ear bars can also cause injury to the ear at impact if the impact forces are all placed at the ears. A head support apparatus placed under the head to displace the forces away from the ears was developed. After testing multiple pillow-like objects, the one that worked the best was the 1 mL latex pipette bulb filled with water. The pipette bulb under the animal’s head can be expanded after the animal is in the stereotaxic frame, allowing it to have a tight fit and provide full support under the head. When placed correctly, there should be no bleeding from the ears or behavioral indications of ear damage (rolling/head tilt) after the injury.
Some versions of the CHI model use a rubber tip probe21,22 or metal helmet23,24 to reduce the occurrence of skull fractures. As long as the 5 mm impactor tip is flush with the skull, there is no need to use any of them. It may be tempting for new users who do not have extensive experience with stereotaxic surgery to induce the injury with the tip not flush with the skull in the medial-lateral plane. If the skull is not level in the medial-lateral plane, it is because the ear bars are not placed correctly. The only fix for this problem is to remove the animal from the impactor and assign the mouse to a sham injury. If the tip is not flush on the anterior-posterior plane, then the height of the bite bar needs to be adjusted and the tip realigned with the bregma. Also, the use of a 5 mm impactor with a flat tip reduces the chance of causing skull fractures19 compared to impactor tips of smaller diameters. Other important factors to consider are the age and weight of the subject, as well as the skull thickness25 and the strains of the mice26.
In people, a mild TBI is not associated with death during the first minutes after the injury. In animals, even a mild injury can cause death. However, in this model, mortality is almost always associated with surgical complications, not the injury alone. The most common reason a mouse would die after the impact is the depth of anesthesia. This could occur if the surgery took longer than expected or if the isoflurane gas was at a higher concentration than needed for that animal. If the animal’s breathing is slow or labored, this could be a sign that the anesthesia depth should be reduced prior to delivering the impact. If the animal’s breath is slow or labored at the time of impact, the animal will likely have apnea and may die.
There are many models of mild TBI. Each has strengths and weaknesses, and this model is no different. As reported, here is described a single hit model of TBI, yet the model has been used to cause a repetitive TBI15. The steps described in this protocol can be repeated to induce a repetitive TBI injury. When evaluating the different TBI models, it is important to consider whether the model has the desired pathology that one is attempting to model. One should also consider how reproducible the model is. It is strongly recommended that the starting point for using this or any TBI model is to independently validate and characterize that the model works as previously reported.
The authors have nothing to disclose.
This work was supported in part by the National Institutes of Health under award numbers R01NS120882, RF1NS119165, and R01NS103785 and the Department of Defense award number AZ190017. The content is solely the responsibility of the authors and does not represent the official views of the National Institutes of Health or the Department of Defense.
9 mm Autoclip Applier | Braintree scientific | ACS- APL | Surgery |
9 mm Autoclip Remover | Braintree scientific | ACS- RMV | Surgery |
9 mm Autoclip, Case of 1,000 clips | Braintree scientific | ACS- CS | Surgery (Staples) |
Aperio ImageScope software | Leica BioSystems | NA | IHC |
BladeFLASK Blade Remover | Fisher Scientific | 22-444-275 | Surgery |
Cotton tip applicator | VWR | 89031-270 | Surgery |
Digitial mouse stereotaxic frame | Stoelting | 51730D | Surgery |
Dumont #7 Forceps | Roboz | RS-5047 | Surgery |
Ear bars | Stoelting | 51649 | Surgery |
EthoVision XT 11.0 | Noldus Information Technology | NA | RAWM |
Fiber-Lite | Dolan-Jeffer Industries | UN16103-DG | Surgery |
Fisherbrand Bulb for Small Pipets | Fisher Scientific | 03-448-21 | Head support apparatus |
Gemini Avoidance System | San Diego Instruments | NA | Active avoidance |
Heating Pad | Sunbeam | 732500000U | Surgery prep |
HRP conjugated goat anti-rabbit IgG | Jackson Immuno Research laboratories | 111-065-144 | IHC |
Induction chamber | Kent Scientific | VetFlo-0530XS | Surgery prep |
Isoflurane, USP | Covetrus | NDC: 11695-6777-2 | Surgery |
Mouse gas anesthesia head holder | Stoelting | 51609M | Surgery |
Neuropactor Stereotaxic Impactor | Neuroscience Tools | n/a | Surgery: Formally distributed by Lecia as impact one |
NexGen Mouse 500 | Allentown | n/a | Post-surgery, holding cage |
Parafilm | Bemis | PM992 | Head support apparatus |
Peanut – Professional Hair Clipper | Whal | 8655-200 | Surgery prep |
Povidone-Iodine Solution USP, 10% (w/v), 1% (w/v) available Iodine, for laboratory | Ricca | 3955-16 | Surgery |
Puralube Vet Oinment,petrolatum ophthalmic ointment, Sterile ocular lubricant | Dechra | 17033-211-38 | Surgery |
Rabbit anti-GFAP | Dako | Z0334 | IHC |
Rabbit anti-IBA1 | Wako | 019-19741 | IHC |
8-arm Radial Arm Water Maze | MazeEngineers | n/a | RAWM |
Scale | OHAUS CS series | BAL-101 | Surgery prep |
Scalpel Handle #7 Solid 6.25" | Roboz | RS-9847 | Surgery |
Sterile Alcohol Prep Pads (isopropyl alcohol 70% v/v) | Fisher Brand | 22-363-750 | Surgery prep |
SumnoSuite low-flow anesthesia system | Kent Scientific | SS-01 | Surgery |
10 mL syringe Luer-Lok Tip | BD Bard-Parker | 302995 | Head support apparatus |
Timers | Fisher Scientific | 6KED8 | Surgery |
Topical anesthetic cream | L.M.X 4 | NDC 0496-0882-15 | Surgery prep |
Triple antibiotic ointment | Major | NDC 0904-0734-31 | Post-surgery |
Tubing | MasterFlex | 96410-16 | Head support apparatus |
Vaporizer Single Channel Anesthesia System | Kent Scientific | VetFlo-1210S | Surgery prep |