Traumatic Brain Injury in Young Rats Leads to Progressive Behavioral Deficits Coincident with Altered Tissue Properties in Adulthood

Abstract

Traumatic brain injury (TBI) affects many infants and children, and results in enduring motor and cognitive impairments with accompanying changes in white matter tracts, yet few experimental studies in rodent juvenile models of TBI (jTBI) have examined the timeline and nature of these deficits, histologically and functionally. We used a single controlled cortical impact (CCI) injury to the parietal cortex of rats at post-natal day (P) 17 to evaluate behavioral alterations, injury volume, and morphological and molecular changes in gray and white matter, with accompanying measures of electrophysiological function. At 60 days post-injury (dpi), we found that jTBI animals displayed behavioral deficits in foot-fault and rotarod tests, along with a left turn bias throughout their early developmental stages and into adulthood. In addition, anxiety-like behaviors on the zero maze emerged in jTBI animals at 60 dpi. The final lesion constituted only ∼3% of brain volume, and morphological tissue changes were evaluated using MRI, as well as immunohistochemistry for neuronal nuclei (NeuN), myelin basic protein (MBP), neurofilament-200 (NF200), and oligodendrocytes (CNPase). White matter morphological changes were associated with a global increase in MBP immunostaining and reduced compound action potential amplitudes at 60 dpi. These results suggest that brain injury early in life can induce long-term white matter dysfunction, occurring in parallel with the delayed development and persistence of behavioral deficits, thus modeling clinical and longitudinal TBI observations.

Introduction

T raumatic brain injury (TBI) in children is a major cause of death and disability (Kuppermann et al., 2009; Schneier et al., 2006), and it has been suggested that developmental neuroplasticity assists in the recovery of younger individuals (Giza and Prins, 2006; Jane, 1996). However, clinical and experimental reports indicate that TBI results in lasting structural and functional damage. Clinically, TBI to the frontal cortex has been associated with disruptions in motor function, problem solving, spontaneity, memory, judgment, impulse control, and social behavior (Levin et al., 1987). Although the underlying cellular and molecular pathophysiology is not fully understood, post-TBI behavioral deficits such as aggressive behavior, anxiety, and depression, as well as an increased risk for epilepsy may last for months to years (Fujii and Ahmed, 2002; Giza and Prins, 2006; Lippert-Gruner et al., 2006). Only a small number of experimental TBI studies have evaluated and shown extended behavioral deficits (Adelson et al., 2000; Huh and Raghupathi, 2007; Prins and Hovda, 1998). Little is known about changes in the tissue properties in association with behavioral modifications.

Several clinical and experimental observations have shown that white matter tracts are highly vulnerable to TBI-induced secondary injury, and that their damage may contribute to cognitive deficits after TBI (Huh and Raghupathi, 2007; Levin et al., 2008; Wagner et al., 2007). Post-natal day (P) 17 in rats, which corresponds to the juvenile population, is a critical myelination period, as it marks the peak of myelin basic protein (MBP) synthesis in the developing rat brain (Akiyama et al., 2002; Bjelke and Seiger, 1989). Previous studies have shown changes in myelination after environmental stress in rats at P17, including changes in MBP (Kodama et al., 2008). Early life stress, such as premature weaning at P17, has been associated with increased anxiety and increased MBP in adulthood (Kodama et al., 2008). To date, experimental juvenile TBI studies have not focused on the timeline of the behavioral deficits seen in association with alterations in white and gray matter tissue properties using a combinatorial approach of magnetic resonance imaging (MRI), immunohistochemistry, and electrophysiology.

The clinical literature led us to hypothesize that an early-life injury such as TBI would elicit enduring behavioral deficits associated with functional and morphological changes in the brain tissue that persist into adulthood. We used a single controlled cortical impact (CCI) injury to the parietal cortex of P17 rats, at an age of active myelin sheath formation, and evaluated the animals over a period of 60 days post-injury (dpi). We carried out a large battery of behavioral tests to assess a wide spectrum of motor and cognitive functions, and each of the tests that we employed measured a unique aspect of the spectrum of deficits reported after juvenile TBI (jTBI) in humans. We used MRI to evaluate anatomical changes (areas of lesion and corpus callosum [CC]). Tissue was collected to assess morphological and molecular changes in gray and white matter, as well as for in vitro axonal conductivity. We present findings of extended behavioral deficits coincident with significant changes in white matter tract properties observed in adulthood, and characterized using MRI, immunohistochemistry, and electrophysiology.

Methods

Animals

All protocols and procedures were approved by the Institutional Animal Care and Use Committee of Loma Linda University, and are in compliance with the U.S. Department of Health and Human Services Guide. Juvenile male Sprague-Dawley rats (P17; Harlan, Indianapolis, IN) were housed with their dams on a 12-h light-dark cycle at constant temperature and humidity. Pups were weaned at P24, housed two rats per cage, and fed with standard lab chow and water ad libitum. All groups of animals used in the study had identical sham or CCI surgical parameters. One group of animals (n=8 sham, n=7 CCI) was studied longitudinally for MRI (3, 30, and 60 dpi), and all behavioral tests (3, 7, 30, and 60 dpi). At 60 dpi, a final immunohistochemistry study was carried out on sham (n=5) and CCI (n=6) animals. Additional groups of animals were evaluated at each time point for immunohistochemistry at 3 dpi (n=3 sham, n=4 CCI), 7 dpi (n=5 sham, n=5 CCI), and 30 dpi (n=5 sham, n=5 CCI), and for electrophysiology at 60 dpi (n=4 sham, n=4 CCI).

Juvenile traumatic brain injury model

CCI was induced in juvenile rats as previously described in our adult experiment (Obenaus et al., 2007). Briefly, P17 juvenile rats were anesthetized with isoflurane (Webster Veterinary Supply, Inc., Sterling, MA) and placed in a mouse stereotaxic frame (David Kopf Instruments, Tujunga, CA). Following a midline skin incision over the skull, a 5-mm craniotomy was performed over the right fronto-parietal cortex (bregma: 1 mm anterior-posterior and 2 mm medial-lateral). CCI was accomplished using a 3-mm rounded-tip metal impactor fixed to an electromechanical actuator and centered over the exposed dura at a 20° angle to be parallel to the cortical surface (Leica Microsystems Company, Richmond, IL). The CCI was delivered at a 1.5-mm depth from the cortical surface with impact duration of 200 msec at a velocity of 6 m/sec. The surgical site was sutured after recording any bleeding or herniation of cortical tissues. Body temperature was maintained at 37°C during surgery. Buprenorphine (0.01 mg/kg; dilution: 0.01 mg/mL) was injected subcutaneously for pain relief before the animals were returned to their dams. The sham animals underwent identical procedures except for the cortical impact.

Magnetic resonance imaging and analysis

MRI was performed in vivo at 3 and 30 dpi followed by high-resolution ex vivo imaging at 60 dpi as previously described (Obenaus et al., 2007). For in vivo MRI, the rats were anesthetized using isoflurane (1–2%), placed in a temperature-controlled cradle containing a volume radio frequency (RF) coil, and imaged on a 11.7-T MRI (8.9 cm bore), or on a larger-bore (40 cm) 4.7-T MRI (Bruker Biospin, Billerica, MA), based on the size of the animal. T2-weighted imaging (T2WI) to determine TBI volume was acquired using: a repetition time/echo time (TR/TE)=3278.3/20 msec; 256×256 matrix; 1-mm slice thickness; 1-mm interleaved; average of two acquisitions; number of slices=20; and field of view (FOV)=2.2 cm, for a total imaging time of 28 min on 11.7-T MRI; and a TR/TE=3563.4/20 msec; 256×256 matrix; 1-mm slice thickness; 1-mm interleaved; average of two acquisitions; number of slices=25; and FOV=2.8 cm, for a total imaging time of 30 min on 4.7-T MRI. The utilization of two different field strengths did not alter the measurements performed in our experiments (Obenaus et al., 2011).

For ex vivo MRI, the animals were transcardially perfused with 4% paraformaldehyde (PFA) prepared in phosphate-buffered saline (PBS), and the brains were immersed in the same solution overnight, and then transferred to PBS with sodium azide (0.1%) before ex vivo scanning using the 11.7-T magnet. T2WI was acquired using: a TR/TE=3278.3/20 msec; 256×256 matrix; 1-mm slice thickness; 1-mm interleaved; average of two acquisitions; number of slices=20; and a FOV=2.2 cm, for a total imaging time of 28 min. We previously reported no volumetric differences between in vivo and ex vivo MRI (Obenaus et al., 2011). Volumetric analyses of the total brain volume and percent lesion/brain volume were performed on T2WI using Cheshire image processing software (Hayden Image Processing Group, Waltham, MA), and Amira (Mercury Computer Systems, Visage Imaging, Inc., San Diego, CA). Lesioned tissue was defined as those regions containing abnormal hypo- and/or hyper-intense signals. All tissue below the skull level was included, but in the rare instances when tissue herniated and extended above the level of the skull, it was excluded. Injury volume was adjusted using total brain volume to correct for individual differences and age-related brain growth. Area measurements of the CC at 60 dpi were selected from the MR images at bregma level+1 mm, anterior to the lesion, and included all visible portions of the CC. The coronal section with the same visible anatomical landmarks was chosen for analysis for all animals by a blinded experimenter.

Behavioral testing

All behavioral tests at each time point were carried out on sham and CCI animals within a 3-h morning time block (8–11 am). Sham and CCI animals were interleaved in testing sequence.

Foot-fault

Foot-fault testing was carried out on an elevated wire mesh (2.5×30 cm rectangular holes/grid spacing) raised 76 cm above the floor at 1, 3, 7, and 60 dpi. The rats were placed in the middle of the wire mesh and its movements were video-recorded for a period of 60 sec in two separate trials 30 min apart (Schmanke et al., 1996). When a rodent's paw slipped completely through the wire mesh, it was considered as an individual fault. The average foot-fault score was calculated from the total number of faults from two 60-sec trials, and a percentage was calculated versus sham levels.

Beam balance

A square acrylic glass beam balance (61 cm long) labeled in 2.5-cm increments was employed in these experiments. Juvenile animals were tested on a 0.65-cm-wide beam at 1, 3, and 7 dpi, and adult animals were tested on a 2.5-cm-wide beam at 30 and 60 dpi according to size requirements. The animals were placed at the midpoint of the beam, perpendicular to the longitudinal axis, and allowed to walk unrestricted in either direction for 60 sec, with two trials 20 min apart. The number of falls, total time spent on the beam, and the distance covered by each animal was recorded.

Rotarod

Rotarod evaluation was performed on all the animals at 30 and 60 dpi (Rotamex-5; Columbus Instruments, Columbus, OH). A rotating 7-cm-wide spindle, accelerating at a rate of 2 RPM every 5 sec until the animal fell off, was used to evaluate performance during two trials 15 min apart. Latency to fall was recorded as a measure of motor coordination and balance (Recker et al., 2009), and the two fall latencies were averaged.

Open-field

Open-field testing assessed general exploratory behavior and activity levels at 30 and 60 dpi. Each animal was placed in an empty plastic bin (50 cm long, 36 cm wide, 45 cm high), and activity was video-recorded for 30 min (Noldus Ethovision; Noldus Information Technology, Inc., Leesburg, VA). Total distance traveled was assessed as a measure of overall activity level.

Zero maze

An elevated zero maze apparatus was used to evaluate anxiety levels at 30 and 60 dpi. The circular apparatus was composed of a 10-cm-wide track (100 cm outer diameter) with walls (30 cm high) enclosing half of the track, with the other half remaining open and brightly lit. The animals were given one trial of 5-min duration and the percentage of time spent in the enclosed half was recorded. Spending more time in the dark quadrants of the track is generally associated with increased anxiety-like behaviors.

Morris water maze (MWM)

Cued and spatial water maze performance was assessed over 3 days as a measure of learning and spatial memory functions at 30 and 60 dpi. A computerized tracking device (Noldus Information Technology) recorded the rodent's swim path (Hartman et al., 2005). The water maze consisted of a 110-cm-diameter metal tank filled with opaque water. An escape platform (11-cm diameter) was adjusted so that the platform's surface was 2 cm above the water surface for cued testing, and submerged 1 cm below the water surface for the spatial learning and memory tasks. In the cued task, the platform was placed in one of four quadrants, and the location of the platform was changed for each block of trials. The animals were given 10 trials (60 sec maximum) in 5 blocks of 2 consecutive trials with a 5-sec inter-trial interval. For each trial, the rat was released into the tank opposite the platform location and allowed to search for the platform. If the rat had not located the platform within 60 sec, it was guided to it. The animals were allowed to remain on the platform for 15 sec after each trial.

For the spatial testing, the platform remained in the same location for all 10 trials and the rat was released into the tank at one of four release points and allowed to find the platform. A probe trial was administered 1.5 h later, in which the platform was removed from the tank, and the rats were allowed to swim for 60 sec. The percentage of time spent in the quadrant where the platform was previously located was calculated. Left/right turn bias and swim speed was also calculated during the probe trial with the percentage of left and right turns measured for each rat.

Tissue processing and immunohistochemistry

Rats were transcardially perfused with 4% PFA prepared in PBS at 3, 7, 30, and 60 dpi, the brains were immersed in the same solution overnight, and then were transferred to PBS with sodium azide (0.1%). The brains were immersed in 30% sucrose at 4°C for 48 h, and then frozen on dry ice and stored at −20°C. Sham and CCI brains were cut as coronal cryostat sections (50 μm) throughout the entire length of the brain (Leica CM1850; Leica Microsystems GmbH, Wetzlar, Germany). All antibody incubations were carried out in PBS (Fisher Scientific, Pittsburgh, PA) containing 0.25% Triton X-100 and 0.25% bovine serum albumin (BSA) (both from Sigma-Aldrich Co., St. Louis, MO). After washes in PBS, the sections were blocked for 1.5 h in PBS with 1% BSA, and then incubated overnight at 4°C with mouse anti-neuronal nuclei (mouse anti-NeuN, 1:500; Chemicon International, Temecula, CA), anti-myelin basic protein (mouse anti-MBP, 1:1000 for classical or 1:200 for infrared analysis; Chemicon International, and rabbit anti-MBP, 1:1000 for classical or 1:200 for infrared analysis; Abcam Inc., Cambridge, MA), oligodendrocyte marker anti-CNPase (1:1000 for classical or 1:400 for infrared analysis; Abcam Inc), or anti-neurofilament-200 (NF200, 1:1000 for classical or 1:250 for infrared analysis; Sigma-Aldrich Co) antibodies. After rinsing in PBS, the sections were incubated for 2 h at room temperature with goat anti-mouse secondary antibody coupled with Alexa-Fluor 488 (1:500 or 1:1000; Molecular Probes, Invitrogen, Carlsbad, CA), or with goat anti-mouse secondary antibody coupled with Alexa-Fluor 800, and goat anti-rabbit secondary antibody coupled with Alexa-Fluor-680 for infrared analysis (1:1000; Molecular Probes, Invitrogen), and subsequently washed in PBS 3×10 min. Sections for classical immunofluorescence were mounted on glass slides and cover-slipped with anti-fading medium (Vectashield; Vector Laboratories, Burlingame, CA) containing 4,6-diamino-2-phenylindole (DAPI). Sections for infrared analysis were mounted on glass slides and dried. Negative control staining, for which the primary antibody was omitted, showed no detectable labeling. Immunostaining was carried out the same day with the same antibody mix for both sham and CCI animals for each time point.

Automated counting of NeuN-positive nuclei was performed as previously described (Badaut et al. 2011) in the dorsal parietal cortex (close to injury), and in the piriform cortex (distant from the injury), in coronal sections that exhibited a visible TBI lesion cavity. NeuN is a specific marker of neuronal nuclei only, allowing the opportunity to count the number of positive neurons stained with NeuN, but not other cell types such as astrocytes, oligodendrocytes, or microglia (Mullen et al., 1992). The region of interest (ROI) for the ipsilateral parietal cortex (ipsi-parietal cortex) was drawn close to the lesion cavity, and the contralateral parietal cortex (contra-parietal cortex) was a mirror image of the ipsi-parietal cortex outline in the contralateral hemisphere. Similar regions were drawn in sham animals at the same anatomical regions. All values were collected using an epifluorescence microscope (model BX41; Olympus, Center Valley, PA) and each ROI contained 80–95 different fields (422×338 μm) quantified using Mercator software (Explora-Nova, La Rochelle, France). The threshold and morphological user-defined parameters were selected to maximize visualization of NeuN-positive staining in the regions of interest, and parameters were kept consistent for all animals. The Mercator program automatically counted only NeuN-positive nuclei (see Fig. 4A). Accuracy of counting was previously tested on slices from control rats stained with NeuN, and no significant differences were observed between the two hemispheres (data not shown).

NF200, MBP, and CNPase coronal sections that were immunostained with infrared secondary antibodies were scanned using the same parameters for sham and CCI with a Licor-Odyssey scanner at 21 μm/pixel resolution (Licor Bio-Science, Lincoln, NE). The infrared method proposed to quantify the immunohistochemistry has recently been described by different authors (Bloch et al., 2011; Wong, 2004), and by our group (Badaut et al., 2007,2011). For surface area measurements of the CC with NF200 (see Fig. 5C) and MBP (see Fig. 6E), an average area was obtained from fields-of-interest drawn with the Licor-Odyssey analysis software around the entire CC on at least three coronal sections located anterior to the injury site at a distance from the disrupted CC directly below the injured area. Tissue processing after MRI and prior to immunohistochemistry (sucrose fixation, freezing, and cryosectioning) may result in a different degree of decrease in CC area from immunohistochemistry compared to MRI. Alternatively, slice thickness on MRI (1 mm) compared to immunohistochemistry (50 μm) could also account for these differences. For infrared analysis of specific regions, integrated intensity (I.I.) values were obtained from 4–8 coronal sections spaced at least 1.2 mm apart throughout the longitudinal brain axis (as available per brain region). For NF200 and CNPase, identical fields-of-interest were drawn using Licor-Odyssey software on both ipsilateral and contralateral hemispheres, only for an analysis of the medial CC (near the longitudinal fissure, adjacent to the injury), and lateral CC (dorsal to the rhinal fissure, ventral to the injury), and a global I.I. average value was calculated for the CC only. For MBP, identical fields-of-interest were drawn using Licor-Odyssey software on both ipsilateral and contralateral hemispheres for the following brain regions: anterior commissure, medial CC (near the longitudinal fissure, adjacent to the injury), lateral CC (dorsal to the rhinal fissure, ventral to the injury), cingulate cortex, medial parietal cortex, lateral parietal cortex (dorsal to the rhinal fissure, ventral to the injury), and temporal/entorhinal cortex (ventral to the rhinal fissure, ventral to the injury). For MBP, statistical comparisons revealed no within-group differences across bregma levels, or ipsilateral versus contralateral hemispheres, meaning there was similar immunoreactivity for all white matter structures, and similar immunoreactivity for all gray matter structures within the individual sham and CCI groups. Therefore, a global I.I. average value was calculated for the white matter tracts (the anterior commissure, and the medial and lateral CC), and the cortical gray matter (the cingulate, medial and lateral parietal, and temporal/entorhinal cortices).

Electrophysiology

At 60 dpi, a separate cohort of rats was anesthetized with 3.5% isoflurane, decapitated, and their brains were dissected to prepare coronal slices of the CC. Slices (400 μm thick) were cut using a vibratome (Electron Microscopy Sciences, Hatfield, PA) in ice-cold artificial cerebrospinal fluid (ACSF) composed of (in mM): 130 NaCl, 3.5 KCl, 1.25 NaH₂PO₄×H₂O, 10 MgSO₄×7 H₂O, 0.50 CaCl₂, 24 NaHCO₃, and 10 glucose, and saturated with carbogen (95% O₂+5% CO₂) at a pH of 7.4. Slices were prepared from a brain region anterior to the injury site (bregma+0.5 mm), transferred to a recording chamber, and superfused for at least 80 min with carbogen-saturated ACSF with the Mg²⁺ concentration reduced to 1.5 mM, and the Ca²⁺ concentration elevated to 2.0 mM. During all electrophysiological recordings, the slices were continuously perfused at a rate of 2–3 mL/min and maintained at 25°C. A stimulation electrode was placed in the callosal fiber tract ipsilateral to the lesion ∼0.5 mm lateral to the longitudinal fissure to deliver square current pulses 0.05 msec in duration at 0.02 Hz. Compound action potentials (CAP) were recorded with glass electrodes (tip resistance 1–3 MΩ) filled with 3 M NaCl and positioned in the contralateral CC, 1.0 mm from the stimulation electrode, and the depth was adjusted to yield a maximal CAP (∼140 μm depth). The responses were amplified (Axoclamp; Axon Instruments, Foster City, CA) and digitized at a sampling rate of 10 kHz, and analyzed using Clampfit (Axon Instruments) and MiniAnalyses (Synaptosoft, Decatur, GA) software. The CAP is characterized as a biphasic waveform produced by fast-conducting, myelinated axons representing the N1 peak, and slower, unmyelinated axons representing the N2 peak. The excitability of the myelinated fibers was assayed by evaluating the amplitude of the N1 peak of the CAP at incrementally increasing stimulation intensities (0.5-mA increments), until the maximal CAP amplitude was reached. Maximal CAP amplitude was used from each animal and averaged.

Statistical analyses

All data are presented as mean±standard error of the mean (SEM), and statistical analyses were done using GraphPad InStat version 3.05 (GraphPad Software, San Diego, CA), and Sigmastat/Sigmaplot (SPSS Inc., Chicago, IL). For behavioral data, a Kolmogorov and Smirnov test was first performed to assess the gaussian distribution, and parametric data were analyzed using unpaired Student t-tests and analysis of variance (ANOVA) testing, followed by Tukey-Kramer for multiple comparisons, while non-parametric data were evaluated with Wilcoxon and Kruskal-Wallis tests to assess group differences. Repeated-measures ANOVAs were used in cases for which longitudinal data were collected on the same groups. MRI and immunohistochemical findings were assessed by unpaired Student t-tests for normally distributed data, or Mann-Whitney U analysis.

Results

All jTBI rats survived the CCI injury (P17) from 1 day to 2 months post-TBI (adulthood; P77). There were no significant weight differences between the CCI and sham groups during the study.

Behavioral testing

Because the TBI lesion was located in the parietal cortex, which contains somatosensory processing areas, a series of motor tests were performed to evaluate the functional impact of jTBI 2 months after the initial injury. We found significant impairments on the foot-fault task, with an increased number of faults at 7 dpi (p<0.05), and 60 dpi (p<0.001; Fig. 1A) compared to sham animals. In the CCI animals, we observed temporal changes from 1 dpi to 60 dpi (p<0.001 by repeated ANOVA) in the number of faults (Fig. 1A), supporting the concept of progressive motor impairments over time. The beam balance results show impaired performance in CCI animals, with less distance traveled compared to sham animals from 1 to 60 dpi (*p<0.05, **p<0.01; Fig. 1B), and no changes within each group's performance over time. Sensorimotor coordination was assessed using the rotarod (Fig. 1C), and revealed that CCI animals were unable to stay on the rotating cylinder as long as sham animals at 30 dpi (p<0.05). CCI animals appeared to recover from this deficit by 60 dpi, indicating impaired coordination that resolved over time (Fig. 1C). Analysis of turn bias from MWM tests demonstrated a significant left turn bias in CCI rats at 30 dpi (p<0.05), which was maintained at 60 dpi (p<0.05; Fig. 1D). Taken together, these results reveal impaired sensorimotor skills of CCI animals that persisted until adulthood.

FIG. 1.

(A) Sensorimotor functions were tested with the foot-fault test. There was a significant increase in the number of faults in controlled cortical impact (CCI) rats compared to their sham counterparts at 7 and 60 dpi (*p<0.05, **p<0.01). (B) Balance and coordination skills were measured with the beam balance test. The distance covered by CCI animals was significantly decreased compared to sham animals at 1, 3, 7, and 60 dpi. (C) Balance and coordination skills were tested with the rotarod test. Fall latency was significantly lower in CCI animals at 30 dpi (*p<0.05). (D) Turn bias measurements indicated that the percentage of left turns (contralateral to the injury) at 30 and 60 dpi was significantly increased in CCI compared to sham animals (*p<0.05; SEM, standard error of the mean).

We also administered a comprehensive battery of tests to assess exploratory behavior (open field), anxiety-like behaviors (zero maze), and learning and memory (MWM), abilities at 30 and 60 dpi. In open-field testing, the total distance traveled in 30 min by the CCI group was significantly less than that seen in sham animals at 30 dpi (p<0.05), but not at 60 dpi (Fig. 2A). These and the above rotarod data (Fig. 1C), suggest that certain aspects of motor function recovered after TBI, whereas others did not. Analysis of the first 3 min of the open-field test, as a marker of exploratory behavior in a novel environment (Fig. 2B), showed that CCI animals moved significantly less when initially placed into the novel environment compared to the sham groups at 60 dpi (p<0.05), consistent with decreased motor/exploratory behaviors. Both CCI and sham animals explored less during the first 3 min at 60 dpi than at 30 dpi (p<0.001), suggesting an effect of general maturation and/or repeated exposure to the open field. Similarly, the zero maze test revealed no group differences in anxiety-like behaviors (measured as increased time spent in the enclosed half of the track) at 30 dpi, but a significant increase in anxiety-like behaviors in CCI animals at 60 dpi (p<0.05; Fig. 2C). Our open-field data suggest reduced motor performance at 30 dpi, whereas reduced exploratory behavior may be related to the emerging anxiety-like behaviors seen in the elevated zero maze at 60 dpi.

FIG. 2.

(A) During open-field testing, the overall distance traveled over 30 min (10 blocks of 3 min each) revealed a significant decrease in activity in the controlled cortical impact (CCI) group compared to sham animals at 30 dpi (*p<0.05). (B) The first 3-min block of activity was evaluated in sham and CCI animals, as this time would normally be used to explore the new environment into which they have been placed. At 60 dpi (but not at 30 dpi) there was a significant difference in the distance traveled in CCI compared to sham animals (*p<0.05), when the CCI animals showed decreased exploratory activity. (C) The zero maze was used to assess anxiety-like behaviors, and in this test more time spent in the dark is thought to correlate with increased anxiety. We observed a significant increase in the time spent in the dark at 60 dpi (*p<0.05). (D) Cumulative distance traveled by the rats to find the platform at 30 and 60 dpi was no different between the CCI and sham groups in the Morris water maze spatial test.

Learning and memory abilities after jTBI were evaluated on the cued and spatial versions of the MWM at 30 and 60 dpi (Fig. 2D). Although both groups improved over time, no performance differences in overall swim speed, cued learning, or spatial learning/memory were observed between sham and CCI animals at either time point.

MRI-derived lesion volumes and surface area of the corpus callosum

MRI at 3, 30, and 60 dpi was used to non-invasively measure lesion volumes (Fig. 3), and the area of the CC (Fig. 5A and B). The relative lesion volume measured by the T2WI may be influenced by edema formation, inflammation, and cell death. Sham animals had few or no abnormalities at the site of the craniotomy, whereas CCI animals demonstrated brain injury as evidenced by increased tissue edema (Fig. 3A, arrows at 3 and 30 dpi). No prominent midline shift was observed in the MR images in this model, and there were only a few CCI animals that exhibited herniation of brain tissue. At 30 and 60 dpi, MRI revealed a cortical cavity that circumscribed the TBI site in the fronto-parietal cortex (Fig. 3A), and a 3D reconstruction (Fig. 3B) revealed the rostro-caudal extent of the lesion cavity in the parietal cortex 60 days after jTBI. Lesion volumes were significantly increased in CCI animals compared to sham controls at 3 dpi (3.5% versus 0.4% of brain volume; p<0.001), 30 dpi (3.0% versus 0.3%; p<0.001), and 60 dpi (2.9% versus 0.1%; p<0.001; Fig. 3C). Lesion volumes, as defined using the parameters of hyper- and hypo-intensity described in the methods section, decreased across the three time points in the CCI and sham groups (p<0.05, Fig. 3C). Decreased lesion volume may be a consequence of decreased edema and inflammation over time. MRI at 60 dpi was used to analyze the surface area of the CC, measured at bregma+1 mm anterior to the site of the injury in CCI animals, and in a comparable coronal section in sham animals, and showed a global decrease of 21% compared to the sham group (p<0.05) at 60 dpi (Fig. 5A and B).

FIG. 3.

(A) Magnetic resonance imaging (MRI) at 3, 30, and 60 dpi identifies the lesion (arrows) as an increase in the signal intensity at 3 and 30 dpi, and as a hole at 60 dpi, corresponding to the formation of the lesion cavity. (B) A three-dimensional reconstruction of the brain (gray) and lesion (red) at 3 dpi extends to include ∼3% of the total brain volume. (C) The lesion volumes were significantly larger in controlled cortical impact (CCI) rats compared to their sham counterparts receiving craniotomy only (**p<0.001) at all time points during our 60-day observation period. Color image is available online at www.liebertonline.com/neu

Immunohistochemical changes following CCI

The consequences of jTBI for gray and white matter were studied at 60 dpi using immunostaining for a neuronal marker (NeuN), neuronal process marker (NF200), oligodendrocyte marker (CNPase), and myelin basic protein (MBP), both at the level of and remote from the site of injury. In sham animals, an intact cortex was observed (Figs. 4A, 5A left, and 6A left), whereas a cavity at the lesion site extending to the level of the CC was seen in the CCI animals (arrow in Fig. 4C). This lesion impacted a relatively small portion of the parietal cortex (in accord with the results obtained from MRI), and may be directly responsible for some of the observed behavioral deficits. In Figure 4, NeuN staining reflects the numerical index of NeuN-positive nuclei used to assess neuronal numbers, and DAPI staining shows no difference in cell density in images co-labeled with NeuN. The DAPI-positive cells without NeuN are non-neuronal cells in the cortex, such as astrocytes, microglia, and endothelial cells. Quantitative analyses of NeuN-positive cells near the injury, and in the ipsilateral and contralateral parietal cortex, revealed a 35% decrease in CCI compared to sham animals (p<0.05; Fig. 4B, D, and E). Quantification of NeuN in the piriform cortex, a cortical region distant from the injury site, did not show any changes in the number of neurons between groups (688±117 NeuN-positive cells/mm² in sham versus 545±60 NeuN-positive cells/mm² in CCI animals). At 60 dpi, the NeuN staining intensity did not change significantly (Fig. 4A and C), in contrast to studies showing increased brain NeuN during the first days after injury (Liu et al., 2009). Therefore, while the intensity level, or actual brightness of NeuN staining was unchanged, the NeuN-positive nuclei reflecting neuronal number was decreased after injury. At 60 days, the observed loss of parietal neurons may be a contributing factor to sensorimotor dysfunction.

FIG. 4.

(A–D) Neuronal nuclei (NeuN) immunohistochemistry at 60 dpi in the sham (A and B) and controlled cortical impact (CCI; C and D) groups showed the lesion cavity in the parietal cortex (arrow in C). The number of NeuN-positive neurons (green) was decreased in CCI (D) compared to sham (B) animals in the parietal cortex (Par Cx) at close proximity to the injury site, in tissue adjacent to the cavity. Nuclei of all cell types were stained with DAPI (blue in B and D), and automated counting of the NeuN-positive cells in the parietal cortex (E) showed a significant reduction in CCI compared to sham animals in the ipsilateral and contralateral hemispheres (*p<0.05; scale bars in A and C=200 μm; in B and D=50 μm; DAPI, 4,6-diamino-2-phenylindole). Color image is available online at www.liebertonline.com/neu

FIG. 5.

(A) Slices from magnetic resonance imaging (MRI) and neurofilament-200 (NF200) immunostaining show a decreased corpus callosum (CC) size on MRI (green and yellow outlines), and NF200 immunostaining (arrows). (B) Quantification of the CC area from the MRI at 60 dpi (mm²) showed a significant decrease in CCI animals (*p<0.05). (C) Quantification of the CC area from the NF200 infrared immunostaining (% of sham animals) shows a significant decrease in the CCI group at 7, 30, and 60 dpi (*p<0.05). (D–G) Classical immunostaining with NF200 images at higher magnification in the corpus callosum (CC, red dashed outlines), and overlying cortex (Cx), showed similar morphological staining patterns, and no changes between the sham (D and E) and CCI (F and G) groups. (H) Infrared (IR) quantification of NF200 staining intensity in the CC was unchanged between groups at 60 dpi (scale bar in A=1 mm; in D and F=500 μm; in E and G=50 μm; A.U., arbitrary units). Color image is available online at www.liebertonline.com/neu

FIG. 6.

Myelin basic protein (MBP) immunostaining in the sham (A, B, and G) and controlled cortical impact (CCI; A, C, F, and G) animals from bregma +1.7 mm to −6.0 mm. (A) MBP immunoreactivity using infrared antibodies with mouse anti-MBP (ms-MBP) was increased in the CCI animals compared to sham animals at the anteroposterior bregma level in white matter tracts. This global increase of MBP was also seen distant from the lesion cavity (asterisk in A). (B and C) Higher-magnification images of classical MBP immunostaining show higher intensity in the corpus callosum (CC) compared to the striatum (STR) and cortex (CX), in (B) sham and (C) CCI animals, and increased MBP staining in the CC white matter of CCI animals (insets in B and C). (D) Quantification of infrared MBP staining showed a significant increase in the white matter tracts (CC and anterior commissure) of CCI animals compared to sham animals at 3 and 60 dpi (*p<0.05). (E) CC area measurements of MBP immunostaining showed a decrease in CCI animals compared to sham animals (p<0.05). (F) Double-immunolabeling with infrared MBP antibodies recognizing two different epitopes, mouse anti-MBP (ms-MBP, green) and rabbit anti-MBP (rb-MBP, red), showed co-localization in the merged images. (G) The intensity of the rb-MBP was increased in the CCI compared to the sham group, as indicated by ms-MBP (asterisks in A, F, and G indicate the lesion cavity; scale bars in A, F, and G=1 mm; in B and C=200 μm; in insets in B and C=50 μm). Color image is available online at www.liebertonline.com/neu

Similarly to the MRI analysis (Fig. 5A and B), area measurements of the CC from infrared NF200-stained coronal sections (Fig. 5A and C) at 60 dpi showed a significant decrease after 7 dpi in the CC of CCI animals (31±1% of sham animals; p<0.01), that was still present at 60 dpi (42±4% of sham animals, p<0.005; Fig. 5C). CC area measurements from MBP immunostaining also showed a similar decrease in CCI animals compared to sham animals (p<0.05; Fig. 6E). CC area changes seen on MRI were highly correlated with measurements derived from infrared NF200 (r²=0.505, p<0.05), or infrared MBP (r²=0.433, p<0.05). Differences in the size of the decrease in CC area between CCI and sham animals as derived from MRI and immunostaining are potentially due to differences in tissue processing and analysis methods. However, it is clear that both quantification methods demonstrate a significant decrease in CC area at 60 dpi following CCI.

In addition to these gross morphological changes seen in the CC surface area following jTBI, infrared NF200 (Fig. 5), infrared CNPase (Fig. 7), and infrared MBP (Fig. 6), immunoreactivities were quantified to address the question of whether they were altered. Quantification of NF200 infrared staining intensity in the CC was not significantly changed between the sham and CCI groups at 60 dpi (Fig. 5A and H). Classical immunostaining with NF200, including higher-magnification images of the CC, showed similar morphological staining patterns between the sham and CCI groups (Fig. 5D–G). NF200 appears more punctate in the gray matter cortex in both sham and CCI animals. In the white matter, there are no obviously distorted fiber tracts, or interrupted fiber tracts, that are commonly observed in other white matter injuries and models (e.g., optic nerve injury, Wang et al., 2011; spinal cord injury Choo et al., 2009; Iannotti et al., 2011). In the CC, CNPase staining intensity was not significantly changed (Fig. 7A and F), in accordance with our observations at higher magnification showing no gross changes in the pattern of staining between the sham and CCI groups (Fig. 7B–E). Classical CNPase immunolabeling identified occasional positive cells in the CC that were consistent in shape and size with expected oligodendrocyte morphology (Fig. 7B–E). Although no changes were detected with these antibodies and quantification methods, it remains possible that other axonal changes occurred at the ultrastructural or molecular level.

FIG. 7.

(A) CNPase immunolabeling along white matter tracts is shown in representative coronal sections near the lesion (bregma −1.8 mm), and anterior to the lesion cavity (bregma +0.5 mm), with bright levels of CNPase staining with infrared (IR) antibodies localized to white matter tracts in the corpus callosum (CC; groups of red arrows). (B–E) Classical CNPase staining at higher magnification shows similar staining patterns in the CC (red dashed outlines), and on individual oligodendrocyte cell bodies (arrows), and processes (arrowheads) of (B and C) sham, and (D and E) controlled cortical impact (CCI) animals. (F) Quantification of infrared staining in the CC across several coronal slices shows no statistically significant differences between the sham and CCI groups for CNPase staining levels (CX, cortex; asterisk in A indicates lesion cavity; scale bar in A=1 mm; in B and D=200 μm; in insets in C and E=20 μm; A.U, arbitrary units). Color image is available online at www.liebertonline.com/neu

However, in contrast to the NF200 and CNPase staining, the immunostaining and higher-magnification images for MBP in Figure 6 reveal stark differences between the sham and CCI groups. As expected during brain development (Akiyama et al., 2002; Bjelke and Seiger, 1989), sham animals demonstrated a significant global increase in MBP immunoreactivity between the ages of 20 and 60 days in the CC (increased 25±4%, p<0.005, data not shown). Quantification of infrared MBP immunoreactivity in the CCI group (% of sham animals) revealed a biphasic change, with an initial increase (25±5%) in CCI compared to sham animals at 3 dpi, normalization at 7 and 30 dpi, and then an increase (33±9%) at 60 dpi, in white matter tracts (Fig. 6D). MBP immunoreactivity was increased in the CCI group at 3 dpi and 60 dpi in all anterior-to-posterior white matter tracts (average values of the anterior commissure and medial and lateral CC, from bregma +1.7 to −6.0 mm; Fig. 6A and D), and was also increased in CCI animals in the CC structure alone (average values of the medial and lateral CC only, from bregma +1.7 to −6.0 mm; p<0.05, data not shown). MBP staining in the cortex and anterior commissure alone were higher in the CCI group than in the sham group from bregma +1.7 to −6.0 mm, but did not reach statistical significance (data not shown). MBP staining was consistent across bregma levels within groups, and no differences were detected for comparisons of individual white matter structures within each individual sham or CCI group. Mosaic images and higher magnification of classical immunostaining demonstrate increased MBP after jTBI at 60 dpi (Fig. 6B and C). The coronal slices in sham animals show uniform MBP staining throughout individual white matter locations, including the CC (Fig. 6B). In contrast, CCI animals have a higher staining intensity level in the CC (Fig. 6C). Similar results were obtained using a second antibody raised in rabbit against MBP, with a different epitope located from amino acid 150 to the N-terminus, in comparison to the antibody raised in mouse against MBP, that recognizes an epitope from amino acids 116–131. Both antibodies showed similar staining patterns and increases in CCI (Fig. 6F–G).

Corpus callosum compound action potentials (CAP) following CCI

To assess the functional consequences of jTBI on the CC, we evaluated electrical transmission within the CC of CCI and sham animals at 60 dpi. N1 amplitudes of CAP reflects excitability of fast-conducting myelinated axons (Fig. 8), and represents the sum of all individual action potentials (Bolton and Carter, 1980; Velumian et al., 2011). The average N1 amplitudes in CCI animals were significantly reduced compared to responses recorded in sham-operated animals (p<0.005; Fig. 8). This suggests an impairment of axonal conductance in the CC of CCI animals that may be associated with a decrease in the thickness of the CC (surface area), and increased MBP (immunoreactivity).

FIG. 8.

Electrophysiology in the corpus callosum (CC) at 60 dpi shows N1 amplitudes recorded after compound action potentials (CAP) were evoked. The stimulating electrode (S) was placed on the side of injury (or sham surgery), located approximately 1 mm from the recording electrode (R). Quantification shows that N1 amplitudes were significantly decreased in controlled cortical impact animals compared to sham animals (*p<0.005).

In summary, our data indicate that the CCI is a focal injury at the time of impact, with formation of a cavity under the site of the impact, but immunohistochemistry and electrophysiology data suggest that the resulting jTBI continues to pathologically affect the brain in a diffuse pattern, primarily along white matter tracts.

Discussion

Using a CCI model of brain injury in juvenile rats, we have demonstrated for the first time that TBI in the developing brain is associated with acute sensorimotor deficits that persist into adulthood, and also showed the delayed appearance at 60 dpi of abnormal anxiety-like behaviors. The morphological changes within the gray matter (the presence of a cavity and reduced number of neurons), and the white matter tracts (decreased area of the CC as seen on MRI and histology) were associated with increased MBP immunostaining and reduced trans-callosal electrical conductance remote from the lesion site. Our findings with the zero maze test are consistent with the accepted notion that anxiety-like behaviors involve several brain regions, such as the amygdala, and the parietal and frontal cortical areas. We believe our global analysis of gray and white matter tissue properties also demonstrate that a local impact evolves to a global change that may affect overall brain function. Specifically, our results suggest that TBI in juvenile rats produces long-lasting changes in gray and white matter properties, and impairments of electrical signaling in major axonal pathways, that may explain the observed sensorimotor deficits and anxiety-like behaviors.

Our observations in this jTBI rodent model corroborate clinical reports that highlight the persistence of neurological, cognitive, behavioral, and psychosocial sequelae after childhood TBI (Goold and Vane, 2009; Lippert-Gruner et al., 2006). In human subjects, these impairments last from a few months to 50 years post-injury, and impact activities of daily living (Brenner et al., 2007; Yeates et al., 2002). A wide range of deficits have been clinically described, including slowed information processing, impaired judgment, attention-deficit/hyperactivity disorder, hampered reasoning and problem-solving skills, mood disorders, anxiety, aggression, and anti-social behaviors (Brenner et al., 2007; Cattelani et al., 1998; Max et al., 2004; Prigatano, 2008).

A focal injury to the parietal sensorimotor cortex in P17 rats resulted in a 35-mm³ lesion volume (∼3% of total brain volume) at 60 dpi (Fig. 3B), which is comparable to previously published data (Prins et al., 2005). The lesion volume remained relatively stable over 60 dpi (Fig. 3C), despite the emergence of specific motor and behavioral impairments. In fact, motor dysfunction was observed within the first week and persisted until 60 dpi (Fig. 1). Early increases in foot-faults and decreased distance covered on beam balance tests are similar to results described in a closed head injury model using P17 mice (Adelson et al., 2000). The recovery that we observed in rotarod testing in the CCI group at 60 dpi may be due to a combination of learning effects, age-related brain development, and reduced sensitivity of this particular test to small deficits in motor planning, proprioception, and coordination, which other tests such as foot-fault can detect (Barreto et al., 2010). In agreement with other investigators (Prins and Hovda, 1998), we did not find major deficits in spatial learning and memory using water maze testing at 30 and 60 dpi. However, jTBI induced the development of anxiety-like behaviors that emerged only in adulthood (Fig. 2C), which was revealed by the longer time spent in the enclosed part of the zero maze, combined with a reduction in exploratory activity in the first 3 min of open-field testing (Fig. 2B). These results are similar to clinical observations of the delayed onset of anxiety in adult patients after TBI (Whelan-Goodinson et al., 2009). The acute decrements in motor function (i.e., increased of incidence of foot-faults and asymmetric turning behavior) are likely the result of the initial lesion and local neuronal cell death in the sensorimotor cortex (Nishibe et al., 2010; Figs. 1 –4). The persistence and progressive worsening of these motor functions could possibly be due to the cavitation of the motor cortex, as well as to impaired axonal conductance associated with the global changes seen in white matter tracts (Figs. 1 and 4 –8). As noted above for the rotarod test, the absence of differences between the groups in open-field testing at 60 dpi (Fig. 2A) may be due to learning or habituation effects resulting from repeated testing (Russell and Williams, 1973).

Clinical imaging studies have emphasized the vulnerability of white-matter pathways, such as the CC and anterior commissure, after TBI (Holshouser et al., 2005; Wilde et al., 2006; Wu et al., 2010). These structural changes may alter the level of MBP, a primary protein involved in myelin sheath formation. Several sets of researchers have studied MBP in human serum and cerebrospinal fluid (CSF) following pediatric and adult TBI, as a potential biomarker of injury severity (Berger et al., 2005; Sandler et al., 2010). For example, serum MBP was increased in children for up to 2 weeks following TBI (Berger et al., 2005). Thus we sought to explore the relationship between white matter tract changes as assessed with MRI, histology, and electrophysiology, and behavioral changes in adulthood after jTBI. Morphological analysis of white matter tracts using MRI and histology revealed decreased callosal thickness starting at 7 dpi that persisted to 60 dpi, findings that are similar to the white matter loss observed in humans (Kraus et al., 2007; Wu et al., 2010). Our data show that the CC is damaged in this CCI model at the site of the impact, but also at some distance from the lesion site, with no recovery seen over time (Figs. 3 –8).

The permanently decreased area of the CC we saw over the 60-day observation period (as measured with MRI, NF200, and MBP), contrasts with the temporal callosal changes in MBP staining that we measured by infrared quantitative immunohistochemistry (Figs. 5 and 6). We found early increased MBP staining at 3 dpi, which may reflect injury-induced myelin changes. Indeed, increased MBP levels were previously reported in an adult rat TBI model, and attributed to MBP fragmentation at 3 dpi (Liu et al., 2006). MBP staining was normalized to sham levels at 7 dpi and 30 dpi, despite the significant decreases in CC thickness (Figs. 5 and 6). In adult TBI models, early myelin fragmentation is due to increased calpain activity that also normalizes 7 days after TBI (Liu et al., 2006). However, our MBP staining normalization was transient, and by 60 dpi we observed a global increase in MBP staining in white matter tracts throughout the anterior-posterior extent of the brain in CCI compared to sham animals (bregma +1.7 to −6.0 mm; Fig. 6A). We also showed an absence of CNPase and NF200 staining changes in the CC, in parallel with increased MBP staining at 60 dpi. Thus, increased MBP staining is likely not a result of anincreased number of oligodendrocytes (CNPase) in the white matter. It remains possible that MBP fragmentation may play a role, as may overexpression associated with an attempt by the brain to compensate for myelin loss (Ihara et al., 2010), but the exact mechanism needs further exploration in this CCI model. Previous studies indicate that glucocorticoids may influence MBP, as they are known to enhance the rate of myelin formation (Chan et al., 1998), and are elevated after brain injury as part of the innate immune response (Glezer and Rivest, 2004).

In conjunction with decreased callosal white matter thickness and increased MBP staining, we found electrophysiological evidence of deficient CC conductivity after jTBI compared to sham animals (Fig. 8), suggesting a possible functional basis for some of the behavioral deficits. The decreased CAP suggests that the increase of MBP observed in CCI animals is non-functional and insufficient to compensate for the loss of axons after CCI. Our findings are similar to a recent report in mice showing that myelin proteolipid protein (PLP) overexpression was associated with decreased axonal conductance velocity (Tanaka et al., 2009). Interestingly, Kodama and associates (2008) used the same mouse anti-MBP antibody, and found increased MBP protein levels in adulthood following early weaning. They ascribed their MBP increase to the 21.5-kDa isoform only, which suggests that the increased MBP observed in their study and in our jTBI model may not necessarily be localized in the myelin sheath. This hypothesis is supported by our electrophysiological data that showed decreased axonal conductance in spite of increased MBP staining, but additional electron microscopy studies would be needed to confirm the precise localization. Further, our data showing decreased CC thickness post-TBI are consistent with a decrease in the number of fibers that may contribute to a decrement in function.

Morphological and functional changes in white matter tracts after jTBI may explain the persistence of motor dysfunction and the development of anxiety-like behaviors in adulthood. Clinically, non-mechanical stressors early in life are associated with long-term behavioral changes, and can also result in altered myelin development and reduced size of white matter tracts. For example, children with post-traumatic stress disorder have a smaller CC volume compared to healthy subjects when they reach adulthood (Bremner et al., 1995). In rodent models, early weaning of rat pups at P17 induces an increase in MBP and development of anxiety-like behaviors (Kodama et al., 2008). Our induction of a cortical mechanical-stress (CCI) injury at the same developmental age also induced an increase in MBP immunoreactivity and development of anxiety-like behaviors that emerged during adulthood. In the rodent, P14–P21 is a critical maturation period for central axonal myelination (Akiyama et al., 2002), and it is likely that a stressor (early weaning or TBI) during this critical period may induce long-term changes in myelination and behavior. The developmental changes seen in myelin proteins (MBP and PLP) were associated with a physiological response to stress and emotional states, including anxiety and depression in adulthood (Kodama et al., 2008; Ono et al., 2008; Tanaka et al., 2009). Thus, despite the presumed neuroplasticity of the juvenile brain after jTBI, there is likely insufficient cortical reorganization or compensation to completely reverse injury-induced deficits.

Conclusion

Our findings showed that a focal TBI results in lasting behavioral deficits (decrements in motor function and late emergence of anxiety-like behaviors), similar to those seen in human TBI patients, as well as global alterations in white matter (decreased thickness of white matter tracts and increased MBP immunoreactivity), leading to long-term changes in neuronal function (decreased neuronal number and axonal conduction). A mechanical stress to the cortex during the sensitive developmental window of myelin formation resulted in long-term changes in the morphology of white matter tracts, as seen with changing MBP immunoreactivity that appears concurrently with decreased axonal output and behavioral changes. However, the precise molecular mechanisms and changes in oligodendrocytes underlying MBP increases in adulthood after TBI have yet to be defined, and suggest new avenues for the development of novel pharmacological therapies.

Footnotes

Acknowledgments

The authors thank Mr. Dane W. Sorensen for his technical help in this work, and Monica Rubalcava at the Advanced Imaging and Microscopy facility at Loma Linda. This work was supported in part by the National Institutes of Health (R01HD061946), the Pediatric Research Fund, the Swiss Science Foundation (FN 31003A-122166 and IZK0Z3-128973), the Department of Pediatrics, and a NASA Cooperative Agreement (NCC9-149) to the Radiobiology Program, Department of Radiation Medicine, at Loma Linda University.

Author Disclosure Statement

No competing financial interests exist.

References

Adelson

P.D.

, Dixon

C.E.

, Kochanek

P.M.

2000. Long-term dysfunction following diffuse traumatic brain injury in the immature rat. J. Neurotrauma, 17:273–282.

Akiyama

, Ichinose

, Omori

, Sakurai

, Asou

2002. Study of expression of myelin basic proteins (MBPs) in developing rat brain using a novel antibody reacting with four major isoforms of MBP. J. Neurosci. Res., 68:19–28.

Badaut

, Ashwal

, Adami

, Tone

, Recker

, Spagnoli

, Ternon

, Obenaus

2011. Brain water mobility decreases after astrocytic aquaporin-4 inhibition using RNA interference. J. Cereb. Blood Flow Metab., 31:819–831.

Badaut

, Ashwal

, Tone

, Regli

, Tian

H.R.

, Obenaus

2007. Temporal and regional evolution of aquaporin-4 expression and magnetic resonance imaging in a rat pup model of neonatal stroke. Pediatr. Res., 62:248–254.

Bloch

, Kaeser

, Sadeghi

, Rouiller

E.M.

, Redmond

D.E.

Jr. , Brunet

J.F.

2011. Doublecortin-positive cells in the adult primate cerebral cortex and possible role in brain plasticity and development. J. Comp. Neurol., 519:775–789.

Barreto

, Huang

T.T.

, Giffard

R.G.

2010. Age-related defects in sensorimotor activity, spatial learning, and memory in C57BL/6 mice. J. Neurosurg. Anesthesiol., 22:214–219.

Berger

R.P.

, Adelson

P.D.

, Pierce

M.C.

, Dulani

, Cassidy

L.D.

, Kochanek

P.M.

2005. Serum neuron-specific enolase, S100B, and myelin basic protein concentrations after inflicted and noninflicted traumatic brain injury in children. J. Neurosurg., 103:61–68.

Bjelke

, Seiger

1989. Morphological distribution of MBP-like immunoreactivity in the brain during development. Int. J. Dev. Neurosci., 7:145–164.

Bolton

C.F.

, Carter

K.M.

1980. Human sensory nerve compound action potential amplitude: variation with sex and finger circumference. J. Neurol. Neurosurg. Psychiatry, 43:925–928.

10.

Bremner

J.D.

, Randall

, Scott

T.M.

, Bronen

R.A.

, Seibyl

J.P.

, Southwick

S.M.

, Delaney

R.C.

, McCarthy

, Charney

D.S.

, Innis

R.B.

1995. MRI-based measurement of hippocampal volume in patients with combat-related posttraumatic stress disorder. Am. J. Psychiatry, 152:973–981.

11.

Brenner

L.A.

, Dise-Lewis

J.E.

, Bartles

S.K.

, O'Brien

S.E.

, Godleski

, Selinger

2007. The long-term impact and rehabilitation of pediatric traumatic brain injury: a 50-year follow-up case study. J. Head Trauma Rehabil., 22:56–64.

12.

Cattelani

, Lombardi

, Brianti

, Mazzucchi

1998. Traumatic brain injury in childhood: intellectual, behavioural and social outcome into adulthood. Brain Inj., 12:283–296.

13.

Chan

J.R.

, Phillips

L.J.

2nd , Glaser

1998. Glucocorticoids and progestins signal the initiation and enhance the rate of myelin formation. Proc. Natl. Acad. Sci. USA, 95:10459–10464.

14.

Choo

A.M.

, Liu

, Dvorak

, Tetzlaff

, Oxland

T.R.

2009. Modeling spinal cord contusion, dislocation, and distraction: Characterization of vertebral clamps, injury severities, and node of Ranvier deformations. J. Neurosci. Meth., 181:6–17.

15.

Fujii

, Ahmed

2002. Psychotic disorder following traumatic brain injury: a conceptual framework. Cogn. Neuropsychiatry, 7:41–62.

16.

Giza

C.C.

, Prins

M.L.

2006. Is being plastic fantastic? Mechanisms of altered plasticity after developmental traumatic brain injury. Dev. Neurosci., 28:364–379.

17.

Glezer

, Rivest

2004. Glucocorticoids: protectors of the brain during innate immune responses. Neuroscientist, 10:538–552.

18.

Goold

, Vane

D.W.

2009. Evaluation of functionality after head injury in adolescents. J. Trauma, 67:71–74discussion 74.

19.

Hartman

R.E.

, Izumi

, Bales

K.R.

, Paul

S.M.

, Wozniak

D.F.

, Holtzman

D.M.

2005. Treatment with an amyloid-beta antibody ameliorates plaque load, learning deficits, and hippocampal long-term potentiation in a mouse model of Alzheimer's disease. J. Neurosci., 25:6213–6220.

20.

Holshouser

B.A.

, Tong

K.A.

, Ashwal

2005. Proton MR spectroscopic imaging depicts diffuse axonal injury in children with traumatic brain injury. Am. J. Neuroradiol., 26:1276–1285.

21.

Huh

J.W.

, Raghupathi

2007. Chronic cognitive deficits and long-term histopathological alterations following contusive brain injury in the immature rat. J. Neurotrauma, 24:1460–1474.

22.

Iannotti

C.A.

, Clark

, Horn

K.P.

, van Rooijen

, Silver

, Steinmetz

M.P.

2011. A combination immunomodulatory treatment promotes neuroprotection and locomotor recovery after contusion SCI. Exp. Neurol., 230:3–15.

23.

Ihara

, Polvikoski

T.M.

, Hall

, Slade

J.Y.

, Perry

R.H.

, Oakley

A.E.

, Englund

, O'Brien

J.T.

, Ince

P.G.

, Kalaria

R.N.

2010. Quantification of myelin loss in frontal lobe white matter in vascular dementia, Alzheimer's disease, and dementia with Lewy bodies. Acta Neuropathol., 119:579–589.

24.

Jane

J. A.

, Francel

P.C.

1996. Age and outcome of head injury. Neurotrauma. Narayan

R.K.

, Wilberger

J.E.

, Povlishock

J.T.

McGraw-Hill: New York, 723–741.

25.

Kodama

, Kikusui

, Takeuchi

, Mori

2008. Effects of early weaning on anxiety and prefrontal cortical and hippocampal myelination in male and female Wistar rats. Dev. Psychobiol., 50:332–342.

26.

Kraus

M.F.

, Susmaras

, Caughlin

B.P.

, Walker

C.J.

, Sweeney

J.A.

, Little

D.M.

2007. White matter integrity and cognition in chronic traumatic brain injury: a diffusion tensor imaging study. Brain, 130:2508–2519.

27.

Kuppermann

, Holmes

J.F.

, Dayan

P.S.

, Hoyle

J.D.

Jr. , Atabaki

S.M.

, Holubkov

, Nadel

F.M.

, Monroe

, Stanley

R.M.

, Borgialli

D.A.

, Badawy

M.K.

, Schunk

J.E.

, Quayle

K.S.

, Mahajan

, Lichenstein

, Lillis

K.A.

, Tunik

M.G.

, Jacobs

E.S.

, Callahan

J.M.

, Gorelick

M.H.

, Glass

T.F.

, Lee

L.K.

, Bachman

M.C.

, Cooper

, Powell

E.C.

, Gerardi

M.J.

, Melville

K.A.

, Muizelaar

J.P.

, Wisner

D.H.

, Zuspan

S.J.

, Dean

J.M.

, Wootton-Gorges

S.L.

2009. Identification of children at very low risk of clinically-important brain injuries after head trauma: a prospective cohort study. Lancet, 374:1160–1170.

28.

Levin

H.S.

, Amparo

, Eisenberg

H.M.

, Williams

D.H.

, High

W.M.

Jr. , McArdle

C.B.

, Weiner

R.L.

1987. Magnetic resonance imaging and computerized tomography in relation to the neurobehavioral sequelae of mild and moderate head injuries. J. Neurosurg., 66:706–713.

29.

Levin

H.S.

, Wilde

E.A.

, Chu

, Yallampalli

, Hanten

G.R.

, Li

, Chia

, Vasquez

A.C.

, Hunter

J.V.

2008. Diffusion tensor imaging in relation to cognitive and functional outcome of traumatic brain injury in children. J. Head Trauma Rehabil., 23:197–208.

30.

Lippert-Gruner

, Kuchta

, Hellmich

, Klug

2006. Neurobehavioural deficits after severe traumatic brain injury (TBI) Brain Inj., 20:569–574.

31.

Liu

, Schafer

D.P.

, McCullough

L.D.

2009. TTC, Fluoro-Jade B and NeuN staining confirm evolving phases of infarction induced by middle cerebral artery occlusion. J. Neurosci. Methods, 179:1–8.

32.

Liu

M.C.

, Akle

, Zheng

, Kitlen

, O'Steen

, Larner

S.F.

, Dave

J.R.

, Tortella

F.C.

, Hayes

R.L.

, Wang

K.K.

2006. Extensive degradation of myelin basic protein isoforms by calpain following traumatic brain injury. J. Neurochem., 98:700–712.

33.

Max

J.E.

, Lansing

A.E.

, Koele

S.L.

, Castillo

C.S.

, Bokura

, Schachar

, Collings

, Williams

K.E.

2004. Attention deficit hyperactivity disorder in children and adolescents following traumatic brain injury. Dev. Neuropsychol., 25:159–177.

34.

Mullen

R.J.

, Buck

C.R.

, Smith

A.M.

1992. NeuN, a neuronal specific nuclear protein in vertebrates. Development, 116:201–211.

35.

Nishibe

, Barbay

, Guggenmos

, Nudo

R.J.

2010. Reorganization of motor cortex after controlled cortical impact in rats and implications for functional recovery. J. Neurotrauma, 27:2221–2232.

36.

Obenaus

, Dilmac

, Tone

, Tian

H.R.

, Hartman

, Digicaylioglu

, Snyder

E.Y.

, Ashwal

2011. Long-term magnetic resonance imaging of stem cells in neonatal ischemic injury. Ann. Neurol., 69:282–291.

37.

Obenaus

, Robbins

, Blanco

, Galloway

N.R.

, Snissarenko

, Gillard

, Lee

, Curras-Collazo

2007. Multi-modal magnetic resonance imaging alterations in two rat models of mild neurotrauma. J. Neurotrauma, 24:1147–1160.

38.

Ono

, Kikusui

, Sasaki

, Ichikawa

, Mori

, Murakami-Murofushi

2008. Early weaning induces anxiety and precocious myelination in the anterior part of the basolateral amygdala of male Balb/c mice. Neuroscience, 156:1103–1110.

39.

Prigatano

G.P.

2008. The problem of not developing normally and pediatric neuropsychological rehabilitation: the Mitchell Rosenthal Lecture. J. Head Trauma Rehabil., 23:414–422.

40.

Prins

M.L.

, Hovda

D.A.

1998. Traumatic brain injury in the developing rat: effects of maturation on Morris water maze acquisition. J. Neurotrauma, 15:799–811.

41.

Prins

M.L.

, Fujima

L.S.

, Hovda

D.A.

2005. Age-dependent reduction of cortical contusion volume by ketones after traumatic brain injury. J. Neurosci. Res., 82:413–420.

42.

Recker

, Adami

, Tone

, Tian

H.R.

, Lalas

, Hartman

R.E.

, Obenaus

, Ashwal

2009. Rodent neonatal bilateral carotid artery occlusion with hypoxia mimics human hypoxic-ischemic injury. J. Cereb. Blood Flow Metab., 29:1305–1316.

43.

Russell

P.A.

, Williams

D.I.

1973. Effects of repeated testing on rats' locomotor activity in the open-field. Anim. Behav., 21:109–111.

44.

Sandler

S.J.

, Figaji

A.A.

, Adelson

P.D.

2010. Clinical applications of biomarkers in pediatric traumatic brain injury. Childs Nerv. Syst., 26:205–213.

45.

Schmanke

T.D.

, Avery

R.A.

, Barth

T.M.

1996. The effects of amphetamine on recovery of function after cortical damage in the rat depend on the behavioral requirements of the task. J. Neurotrauma, 13:293–307.

46.

Schneier

A.J.

, Shields

B.J.

, Hostetler

S.G.

, Xiang

, Smith

G.A.

2006. Incidence of pediatric traumatic brain injury and associated hospital resource utilization in the United States. Pediatrics, 118:483–492.

47.

Tanaka

, Ma

, Tanaka

K.F.

, Takao

, Komada

, Tanda

, Suzuki

, Ishibashi

, Baba

, Isa

, Shigemoto

, Ono

, Miyakawa

, Ikenaka

2009. Mice with altered myelin proteolipid protein gene expression display cognitive deficits accompanied by abnormal neuron-glia interactions and decreased conduction velocities. J. Neurosci., 29:8363–8371.

48.

Velumian

A.A.

, Wan

, Samoilova

, Fehlings

M.G.

2011. Contribution of fast and slow conducting myelinated axons to single-peak compound action potentials in rat spinal cord white matter preparations. J. Neurophysiol., 105:929–941.

49.

Wagner

A.K.

, Postal

B.A.

, Darrah

S.D.

, Chen

, Khan

A.S.

2007. Deficits in novelty exploration after controlled cortical impact. J. Neurotrauma, 24:1308–1320.

50.

Wang

, Hamm

R.J.

, Povlishoch

J.T.

2011. Traumatic axonal injury in the optic nerve: evidence for axonal swelling, disconnection, dieback, and reorganization. J. Neurotrauma, 28:1185–1198.

51.

Whelan-Goodinson

, Ponsford

, Johnston

, Grant

2009. Psychiatric disorders following traumatic brain injury: their nature and frequency. J. Head Trauma Rehabil., 24:324–332.

52.

Wilde

E.A.

, Bigler

E.D.

, Haider

J.M.

, Chu

, Levin

H.S.

, Li

, Hunter

J.V.

2006. Vulnerability of the anterior commissure in moderate to severe pediatric traumatic brain injury. J. Child Neurol., 21:769–776.

53.

Wong

S.K.

2004. A 384-well cell-based phospho-ERK assay for dopamine D2 and D3 receptors. Anal. Biochem., 333:265–272.

54.

T. C.

, Wilde

E. A.

, Bigler

E. D.

, Li

, Merkley

T. L.

, Yallampalli

, McCauley

S. R.

, Schnelle

K. P.

, Vasquez

A. C.

, Chu

, Hanten

, Hunter

J. V.

, Levin

H. S.

2010. Longitudinal Changes in the Corpus Callosum following Pediatric Traumatic Brain Injury. Dev. Neurosci., 32:361–373.

55.

Yeates

K.O.

, Taylor

H.G.

, Wade

S.L.

, Drotar

, Stancin

, Minich

2002. A prospective study of short- and long-term neuropsychological outcomes after traumatic brain injury in children. Neuropsychology, 16:514–523.