A Pediatric Concussion Model in Mice: Closed Head Injury with Long-Term Disorders (CHILD)

Summary

Pediatric concussions can result in long-term sequelae with physiological and psychological perturbations. This pediatric concussion mouse model results in long-term neuroinflammation, white matter alterations, and neuronal modifications with behavioral dysfunction. The model can be adapted to differential impact locations or repeated concussions to assess lifetime changes.

Abstract

Concussions and mild traumatic brain injuries (mTBI) during childhood represent a significant health endangerment, with many patients later in life exhibiting debilitating physiological, neurological, and psychosocial outcomes. The cellular and molecular pathophysiological mechanisms are relatively unknown, and this significant gap effectively precludes investigations into specific therapeutic strategies to mitigate chronic sequelae. No single animal concussion model recapitulates all the features reported in human subjects, and current models are designed to answer specific research questions. We set out to develop a juvenile concussion that recapitulates clinical early, and long-term symptomology encountered, termed Closed Head Injury with Long-term Disorder (CHILD). In this model, a concussive force via an electromagnetic impactor is delivered to the head of a lightly anesthetized unrestrained post-natal day 17 (P17) mouse, allowing free rotation of the head. Mice are placed on a tightly stretched tin foil across a stereotactic device, and the impactor is carefully aligned with the targeted cortical region. The electromagnetic impactor facilitates the selection of impact depth, dwell, and duration to determine injury severity. A strength of this model is that it allows head rotation, an important clinical feature. After impact, mice are immediately monitored for righting time and time to explore their environment, followed by their return to their dam. Increasing the severity of concussion results in intracranial bleeds in a subset of mice. Mice can be routinely monitored for behavior and neuroimaging over their lifespan, as desired. Various injury severities mimic the heterogeneous nature of juvenile concussions. The current standard CHILD model exhibits no skull fracture, no observable conventional neuroimaging change (similar to clinical), but leads to progressive and persistent neuronal death, altered diffusion MRI, modified neuronal activity and plasticity, increased gliosis, and progressive behavioral perturbations with age. In summary, this CHILD model mimics the early and long-term features observed in many clinical concussion patients.

Introduction

Pediatric Traumatic Brain Injury (TBI) is a major health concern as it represents the leading cause of pediatric emergency medical care visits and is associated with debilitating health outcomes later in life. Even mild TBI (mTBI), which makes up 75% of all TBI cases¹, leads to long-term abnormalities, including physiological, psychological, neurological, and cognitive dysfunctions². Despite the high occurrence of pediatric mTBI and its significant associated health risks, the pathological, cellular, and molecular mechanisms that precede the chronic deficits are still underexplored. This gap in knowledge impedes the development of effective treatments and diagnostic methods for pediatric mTBI.

Adult rodent TBI models include weight drop, controlled cortical impact, and fluid percussion injury and can be titrated to elicit moderate to severe injuries (dependent upon outcome measures)^3,4. Many of these adult models have been adapted for pediatric TBI^5,6. During the last decade, models of mTBI have been developed in rodents with closed head injury as well as with rotational models⁷. However, implementation of these models in developing rodents is sparse compared to the adult TBI literature⁸. To address the clinical relevance, encompassing a wide range of severities that model pediatric clinical heterogeneity, this study developed the Closed Head Injury with Long-term Disorder (CHILD) TBI model. CHILD is a non-surgical, unrestrained concussion model using P17 mice, a developmental age that corresponds to the peak of myelination observed in 2-3-year-old children⁹. A concussive force is delivered using an electromagnetic impactor over the somatosensory cortex of a lightly isoflurane-anesthetized mouse. Impacting the unrestrained head results in rotational forces on the brain¹⁰, as reported in many clinical concussion reports. Consistent with clinical mTBI, the CHILD model does not lead to skull fracture or visible abnormality on conventional neuroimaging.

The CHILD model has several advantages. First, both early and long-term pathological features have been observed in CHILD, mimicking the pathological clinical features observed in juvenile concussion patients. It has been observed at late time points gliosis, altered diffusion tensor MRI, and progressive behavioral perturbations with age^10,11, modified neuronal activity, and plasticity^11,12. Interestingly, unsuspected peripheral consequences were also found with cardiac abnormalities in the long term after a single pediatric injury¹³. Second, this model is replicable and showed comparable results between researchers and between labs¹³. Third, the CHILD model is flexible and can be expanded to encapsulate the heterogeneous nature of TBIs. For instance, previous studies have shown that different speeds and depths of impact result in different severities^10,11. Finally, an additional strength of this model is that CHILD can be administered repeatedly and can be easily applied to differential locations to probe responses to repeated injury and identify vulnerable brain regions.

Overall, the CHILD model replicates clinical pediatric mTBI and concussion and serves as a research tool to facilitate the understanding of the pathophysiological characteristics of pediatric mTBI. While no model encapsulates every aspect of the features described in real-life TBI cases, this model can address many of the outstanding research questions that remain unanswered.

Protocol

Experiments were reproduced on 2 sites, the University of Bordeaux (France) and the University of California Riverside (CA, USA).

In Bordeaux, all experiments were performed on male and female P17 C57Bl6J mice bred in the University animal facility. Initial breeders were obtained from a commercial source and raised under standard conditions with a 12/12-h dark/light cycle, a temperature of 21 °C ± 1 °C, and a humidity of 55% ± 10%, and food and water provided ad libitum. Experimental procedures are in accordance with the University of Bordeaux Animal Care Committee regulations, French laws governing laboratory animal use (authorization #29324-2021012118549817 v3), the European Council directives (86/609/EEC), and the ARRIVE guidelines. This protocol, being a mild model of traumatic brain injury, does not produce pain or distress. As such, it does not necessitate any specific pain management outside of regular pain management. Weaning of the pups (TBI, Shams) was effective at P25.

At Riverside, E13 pregnant C57Bl6J female mice were commercially obtained, and all experiments were performed on the resultant litters in both male and female pups aged P17. Animals were raised with an automated 12/12-h dark/light cycle and a temperature of 21 °C. Dams and pups had ad libitum access to food and water. All experimental procedures were approved by the University of California, Riverside Institutional Animal Care and Use Committee and complied with the Animal Welfare Act and Public Health Service policy. All pups (TBI, Sham) in that facility were weaned on P21.

The details of the animals, reagents, and equipment used in this study are listed in the Table of Materials.

1. Preliminary steps

1. To prevent hypothermia, warm up the chamber for anesthesia induction and the recovery cages in advance with warming pads.
2. Set the stereotaxic apparatus with the impactor mounted at a 90° angle.
3. Verify that all parts are properly tightened and secured and that the piston is firmly screwed to the impactor.
4. Cut an aluminum sheet that will fit the stereotaxic apparatus (in this case, 18 cm x 30 cm). The aluminum used here is of standard aluminum quality. Wrap the stereotaxic frame with the foil to form a pad on which the animal will be laid on. Secure the position and tension of the foil with regular lab tape in order to sustain the weight of the animal.
5. Turn on the impactor controller and, in the function of the severity of the injury, set the speed of impact to 2 m.s^-1 for grade 1 (G1) or 3 m.s^-1 for grade 2 (G2) for a more severe impact (modifying this parameter will impact the severity of the concussion).
6. Weight out the animals. At P17, the weight of the animals must be over 5.9 g for C57Bl6 mice. Animals that fall under that weight are excluded.
   NOTE: For some studies, CD1 mice were used, and for P17, the threshold was fixed to be over 6.5 g (personal data).

2. Juvenile mild traumatic brain injury procedure

1. Anesthetize the mouse with a 5 min exposure to isoflurane. Set the gas composition to a 2.5% mixture (in this case, the carrier is a synthetic air mixture with 21% oxygen) with a flow rate of 1.5 L/min.
   NOTE: From this step, all the actions must be done swiftly, as unimpacted animals will wake up within 5 min.
2. Position the mouse head under the impactor. The pup should be spread out at all lengths. The tip is 3 mm in diameter and covered with rubber to minimize the effect of the metal on the skull.
3. Lower the impactor by switching the knob to the Extend position.
4. Roughly lower the impactor and position the mouse so that the impactor is just between the ears. Then, slide the mouse so that the piston tip is moved forward to the equivalent of one piston tip, and one piston tip to the left of the animal (see Figure 1).
   NOTE: Sham animals are positioned slightly away from the piston (no part of the animal should be impacted).
5. Lower the impactor so that it contacts the head of the animal (Figure 1E2).
6. Set the impactor to Retract (Figure 1E3).
7. In order to determine the severity of the injury, move the piston 1 mm down for grade 1 (G1). For more severe impact grade 2 (G2), move the piston 3 mm down (Figure 1E4).
8. Set the dwell time to 0.1 s regardless of the severity (modifying this parameter will not impact the severity of the concussion).
9. Press Impact on the right knob (Figure 1E5).
10. Rapidly set the left knob to the off position so that the system doesn't overheat. See the flow chart for more information (Figure 2).

Figure 1: Positioning of the impactor tip. (A) The impactor tip is first positioned on the midline between the ears. (B) The tip is then moved 3 mm forward (the equivalent of a tip diameter). (C) Finally, the piston tip is moved 3 mm on the left side of the mouse. (D,E) Possible positions of the impactor tip in regard to the magnetic barrel. (D) The impactor has 3 positions in relation to the piston body (neutral, retracted, or extended/impact) with 5 mm range between them. (E) Positions of the impactor tip in relation to the skull surface at the different stages of the procedure. (1) Position of the impactor in the neutral position; (2) the piston is moved to gently touch the skull of the animal (step 2.5); (3) the impactor is then set to retract (step 3.6); (4) the whole body of the impactor is moved of x mm down (step 2.7). For G1 grade, a speed of impact (y) of 2 m.s^-1 with 1 mm depth (x) for G1 or a speed of impact (y) of 3 m.s^-1 with 3 mm (x) depth for G2. (5) Once the lever impact is pressed, the impactor tip moves x mm below the skull surface at a y speed during dwell time upon moving back to the neutral position. Please click here to view a larger version of this figure.

Figure 2: CHILD flow chart. Flow chart detailing the steps to induce CHILD in PND17 mice color-coded to denote different phases of concussion induction. Photographs of the setup include (A) the entire CHILD induction setup (anesthesia device is not visible), including the Impact controller, the stereotactic device with the impactor and tip positioned above taut tin foil, and the recovery box located on top of a heating pad. (B) Rubber impact tip (3 mm) with scale. (C) The stereotactic device with tin foil stretched across the stereotactic frame and the impactor solenoid with the impact tip positioned above the foil. (D) Recovery chamber sitting on the heat pad. (E) The impact controller with two control switches, highlighted as in the flow chart (red box = extend/retract/off; purple box= impact toggle). Please click here to view a larger version of this figure.

3. Post-trauma follow-up

NOTE: The following steps can be videorecorded for offline analysis, allowing the experimenter to perform CHILD on multiple animals in a row. Nonetheless, regular and repeated assessments of animal well-being should be performed.

1. Note the duration of apnea, if present (immediately after impact), and if there was a head rotation during impact.
2. Position the mouse on its right flank in a recovery cage.
3. Note the righting time of the mouse (rearing on 4 legs) and the time of resuming an exploratory behavior. Typically, sham mice awake within 3-4 min and resume an exploratory behavior 3-4 min later (Figure 3). CHILD mice awake within 7-12 min after impact (Figure 3A) and resume exploration within 12-18 min after impact (Figure 3B), according to the sex and impact severity (Figure 3A,B).
   NOTE: If the impacted animal doesn't resume its exploratory behavior within 30 min, the animal is excluded.
4. Once fully recovered, return the animals to their home cage. At this point, the behavior of sham and CHILD mice (G1 or G2 severity) should be exactly similar, and animal groups shouldn't be distinguishable within the litter.
   NOTE: The animals can be maintained up to 18 months post-injury¹⁴ in conditions indicated at the beginning of the methods. Behavioral outcomes assessments can be performed using a battery of tests in association with MRI measurements, miniscope imaging, histology, and immunohistochemistry¹². Immunohistochemistry for GFAP and NeuN can be used at early time points to validate the model and the presence of tissue changes.

4. Immunohistochemistry

NOTE: One day post-injury, animals from the G2 and sham groups (n = 4 mice for G2 and n = 4 for Sham) were anesthetized and transcardially perfused with 4% paraformaldehyde (PFA) prepared in phosphate-buffered saline (PBS) 0.1 M, pH 7.4 at 4 °C. Brains were extracted¹⁰ from the skull and immersed in the same PFA solution overnight, and then transferred to PBS with sodium azide (0.1%) before cutting. Coronal sections were then prepared at 50 µm thickness using a vibratome. For long storage, sections were placed in cryoprotective medium (30% ethylene glycol and 20% glycerol in PBS) at -20 °C until use. To characterize the impact of the concussion on brain gliosis and neuronal loss, immunohistochemistry targeting GFAP and NeuN was performed as described here below.

1. Wash the brain sections 4 times for 10 min in PBS.
2. Transfer the sections in blocking solution (1% BSA, 0,3% Triton X-100 in PBS) for 10 min at room temperature (RT).
3. Incubate sections overnight at 4 °C with the following primary antibodies diluted in blocking solution: chicken anti-GFAP (1:3000) and rabbit anti-NeuN (1:500).
4. Wash the slices 4 times for 10 min in PBS under agitation.
5. Then, transfer the sections in blocking solution containing the secondary fluorescent antibodies diluted 1:1000 for 1 h at RT, Alexa-Fluor-488-nm goat anti-rabbit, and Alexa-Fluor-568-nm goat anti-chicken.
6. Wash then the sections 4 times for 10 min in PBS under agitation.
7. Mount the sections on slides using a mounting medium and add a cover slip. Keep the slides at 4 °C until the image acquisition.
8. Aquire immunofluorescence using an epifluorescence microscope and micromanager software. Use a stitching plugin with a 10x lens to obtain tiled images of ipsilateral and contralateral hemispheres.
   NOTE: Negative control staining where the primary antibody or secondary antibody was omitted showed no detectable labeling. For more detailed image analysis, Fiji software¹⁵ can be used to quantify the immunostaining parameters in various regions of interest¹⁶.

Results

The CHILD model is a relatively rapid and reproducible model of mTBI in pediatric mice, also allowing for variations in severity and location. The CHILD mice, as expected, exhibited significantly longer recovery times from closed head injury (Figure 3). When cohorts (n = 36 female sham, n = 140 male sham, n = 30 female G1, n = 55 male G1, n = 32 female G2, n = 141 male G2) were aggregated, a delay in righting time was observed (Figure 3A, 3-way ANOVA p ...

Discussion

Critical steps in the protocol
The CHILD protocol described here requires the use of an electromagnetic impactor (Figure 2A) that (1) standardizes the impact parameters (speed, depth, duration), (2) ensures reproducibility within and across laboratories, and (3) is readily available commercially. Another critical component is the impactor tip that strikes the rodent's head. A variety of possibilities can be used, ranging from silicone to metal tips, but this protoc...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
BSA	Sigma Aldrich Chimie S.a.r.l 80 Rue de Luzais L'lsle-d'Abeau Chesnes - BP701 38297 Saint-Quentin-Fallavier Cedex France	A7030
C57Bl6J mice	Charles River Laboratory France CRLF 9 All. Moulin Berger, 69130 Écully, France		Bordeaux
Electromagnetic impactor	Leica Biosystems, Richmond, IL USA	Impact One stereotaxic impactor
Epifluorescence microscope	Olympus, BX41, Center Valley, PA
Ethylen glycol	Sigma Aldrich Chimie S.a.r.l 80 Rue de Luzais L'lsle-d'Abeau Chesnes - BP701 38297 Saint-Quentin-Fallavier Cedex France	102466
GFAP primary antibody	Millipore, Billerica, MA	AB5541	chicken anti-GFAP
GFAP secondary antibody	Molecular Probes, Invitrogen	A11034	goat anti-rabbit AlexaFluor  488
Glycerol	Sigma Aldrich Chimie S.a.r.l 80 Rue de Luzais L'lsle-d'Abeau Chesnes - BP701 38297 Saint-Quentin-Fallavier Cedex France	G5516
Heavy duty aluminium foil	Kirkland	RK611	obtained at Cosco
Micromanager software	(NIH, USA
NeuN primary antibody	AbCam, Discovery Drive, Cambridge Biomedical Campus, Cambridge, CB2 0AX, UK	AB128886	rabbit anti-NeuN
NeuN secondary antibody	Molecular Probes, Invitrogen	A11041	goat anti-chicken AlexaFluor 568
PBS	Sigma Aldrich Chimie S.a.r.l 80 Rue de Luzais L'lsle-d'Abeau Chesnes - BP701 38297 Saint-Quentin-Fallavier Cedex France	P5493
PFA	Sigma Aldrich Chimie S.a.r.l 80 Rue de Luzais L'lsle-d'Abeau Chesnes - BP701 38297 Saint-Quentin-Fallavier Cedex France	1.04003
Sodium azide	Sigma Aldrich Chimie S.a.r.l 80 Rue de Luzais L'lsle-d'Abeau Chesnes - BP701 38297 Saint-Quentin-Fallavier Cedex France	26628-22-8
Stereotaxic frame - base	KOPF instruments 7324 Elmo St, Tujunga, CA 91042, États-Unis	900-C
Stereotaxic frame - electrode manipulator	KOPF instruments 7324 Elmo St, Tujunga, CA 91042, États-Unis	960
Stereotaxic frame - u-frame	KOPF instruments 7324 Elmo St, Tujunga, CA 91042, États-Unis	900-U
Triton x-100	Sigma Aldrich Chimie S.a.r.l 80 Rue de Luzais L'lsle-d'Abeau Chesnes - BP701 38297 Saint-Quentin-Fallavier Cedex France	X100
Vectashield	Vector laboratories	H-1000-10	Mounting medium
Vibratome	Leica Biosystems, Richmond, IL USA	VTS1000
	Jackson laboratory The Jackson Laboratory 600 Main Street Bar Harbor, ME USA 04609		UCR