LM has received research funding from Bristol-Myers Squibb and has served on the scientific advisory board of iPierian. He is a co-inventor on patent applications/patents relating to tau reduction that are owned by the Gladstone Institutes (U.S. Patent Publication Number 20120198573; U.S. Patent Publication Number 20140065206; and EP Patent No. 2145014), all of which are entitled “Agents that Reduce Neuronal Overexcitation”. This does not alter the authors' adherence to PLOS ONE policies on sharing data and materials.
Conceived and designed the experiments: JC GY GM SM RP LM. Performed the experiments: JC RC GY XW KH GM SM. Analyzed the data: JC SM RP LM. Contributed reagents/materials/analysis tools: SM. Wrote the paper: JC LM.
Because reduction of the microtubule-associated protein Tau has beneficial effects in mouse models of Alzheimer's disease and epilepsy, we wanted to determine whether this strategy can also improve the outcome of mild traumatic brain injury (TBI).
We adapted a mild frontal impact model of TBI for wildtype C57Bl/6J mice and characterized the behavioral deficits it causes in these animals. The Barnes maze, Y maze, contextual and cued fear conditioning, elevated plus maze, open field, balance beam, and forced swim test were used to assess different behavioral functions. Magnetic resonance imaging (MRI, 7 Tesla) and histological analysis of brain sections were used to look for neuropathological alterations. We also compared the functional effects of this TBI model and of controlled cortical impact in mice with two, one or no
Repeated (2-hit), but not single (1-hit), mild frontal impact impaired spatial learning and memory in wildtype mice as determined by testing of mice in the Barnes maze one month after the injury. Locomotor activity, anxiety, depression and fear related behaviors did not differ between injured and sham-injured mice. MRI imaging did not reveal focal injury or mass lesions shortly after the injury. Complete ablation or partial reduction of tau prevented deficits in spatial learning and memory after repeated mild frontal impact. Complete tau ablation also showed a trend towards protection after a single controlled cortical impact. Complete or partial reduction of tau also reduced the level of axonopathy in the corpus callosum after repeated mild frontal impact.
Tau promotes or enables the development of learning and memory deficits and of axonopathy after mild TBI, and tau reduction counteracts these adverse effects.
Each year, 1.7 million Americans suffer mild traumatic brain injury (TBI) and half of them experience an acute loss of consciousness (LOC) as a result of the injury
Histopathological studies of
Recent evidence suggests that mild TBI may share pathogenic mechanisms with AD, including aberrant network excitability, cytoskeletal disruption, and inflammation
We used male and female C57Bl/6J mice at 4–6 months of age. Mice were group housed on a regular 12-h light/dark cycle. Food (PicoLab Rodent Diet 20,5053) and water were provided
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All procedures were approved by the Animal Care and Use Committee of the University of California, San Francisco and all efforts were made to minimize suffering.
Mice were randomly assigned to undergo TBI or sham surgery. Three types of mild TBI were used: frontal impact (2-hit vs 1-hit) and controlled cortical impact (1-hit). After injury, mice were injected with buprenorphine (0.01 mg/kg SC) and checked hourly for 6 hours and then daily. They were allowed to recover for a minimum of 2 weeks before the initiation of behavioral testing.
We adapted a rat model of diffuse cortical injury
Mice were induced for 2 min with 3% isoflurane and maintained under this anesthesia using a nose cone for the duration of the procedure. Bupivacaine (8 mg/kg SC) was administered to the scalp. After shaving, a linear incision was made in the midline followed by a 2.5 mm circular craniotomy. The CCI device (Hatteras Instruments, Caray, NC) was attached to a stereotactic frame and positioned 1.5 mm lateral and 2.3 mm posterior to Bregma. Injury was inflicted using a 1.5-mm circular, flat impactor tip traveling at a speed of 3 m/s and penetrating to a depth of 1.5 mm for 150 ms. After injury, the craniotomy bone was replaced and the scalp closed using dermabond cement. Sham-injured animals underwent anesthesia and craniotomy but not cortical impact.
Mouse cohorts underwent a maximum of three of the behavioral tests described below. In instances where more than one test was performed on the same group of mice, the order of testing was selected to minimize test interactions. Testing in the open field and elevated plus maze preceded testing in the Barnes maze, and testing in the Y-maze and on the balance beam preceded contextual and cued fear conditioning. The forced swim test was carried out in a mouse cohort that did not undergo other behavioral tests.
The maze consisted of a circular platform (91.4 cm diameter) with 20 holes around the periphery (5.1 cm diameter) with an escape box attached to the bottom of one of the holes and shallow boxes attached to the bottom of the other holes. The lights were kept bright (650 lux) to motivate mice to find and enter the escape box. Visual extra-maze cues were present on 3 walls of the room at a 1.5–1.8 m distance from the maze. For all trials, mice were placed individually in a cylindrical black start chamber in the center of the maze for 10 s, which was then lifted to start the test. During an adaptation period, mice were guided to the escape tunnel and allowed to stay there for 2 min. During a spatial acquisition period, a total of 10 acquisition trials (2 trials per day with an inter-trial interval of 15 min) were performed; mice were allowed to explore the maze freely for 3 min. Each trial ended when the mouse entered the escape tunnel or after 3 min had elapsed. Mice that did not find the tunnel were guided to it. All mice were allowed to remain in the tunnel for 1 min. During the probe trial conducted 1 day after the last training trial, the escape tunnel was replaced by a shallow box and mice were allowed to explore the maze for 90 s. Mice were video recorded and the time (“latency”) and path length (“distance”) taken to the target location during the probe trials were measured. For mice that did not reach the target location, total testing time (90 s) and total distance moved were used for analysis in lieu of latency and distance to target.
Mice were tested in a 3-day paradigm as described
Mice were tested for a total of 10 min in a dimly lit room. The test was initiated by placing mice at the intersection between the open and closed arms. Basic locomotor activity and percent of time spent in open versus closed arms were recorded as described
Mice were tested for total movements and rearings as described
Mice were trained to traverse a square beam measuring 6 mm×6 mm×61 cm and tested using a square beam measuring 3 mm×3 mm×61 cm. The total number of foot slips and latency to cross the beam were recorded as described
The test was initiated by placing mice into one arm of the Y-maze. Total movements and the number and percent of alternations were recorded for 6 min and analyzed as described
Mice were placed in a clear polycarbonate cylinder measuring 31 cm in diameter and 76 cm in height filled to 48 cm with room-temperature tap water. Time to immobilization after immersion was recorded up to a 6 min maximum as described
Mice were anesthetized with isoflurane while MRI data were acquired with a 7T MRI scanner (Agilent/Varian, Santa Clara, CA) using a 3D gradient echo sequence to produce T1 and T2* weighted images as described
Twelve months after 2-hit injury, mice were anesthetized with Avertin (tribromomethanol, 250 mg/kg) and perfused transcardially with 0.9% saline for 1 min. Brains were removed and post-fixed in 4% paraformaldehyde (PFA) at 4°C for 24 hrs, followed by incubation in 30% sucrose for 1–3 days at 4°C. Hemi-brains were then sectioned coronally to a thickness of 30 µm using a freezing microtome (Leica SM 2000R). Sections were stained with the following antibodies: mAPP (Millipore, 22C11, dilution 1∶5000), GFAP (Millipore, MAB360 Clone GA5, dilution 1∶2000), and Iba-1 (Wako Chemicals, 019-19741, 1∶1000). An avidin-biotin complex kit (Vector Laboratories) and 3,3′-diaminobenzidine tetrahydrochloride (Vector Laboratories) were used to visualize antibody labeling. Silver staining was performed using the Bielschowsky method as described
Levels of axonopathy were determined by counting mAPP-positive profiles throughout the body of the corpus callosum in four sections per mouse. To detect astrocytosis and microgliosis, three non-overlapping areas (100 µm2 each) in the body of the corpus callosum were randomly selected in four sections per mouse and the average percent area occupied by GFAP or Iba-1 immunoreactivity was determined using ImageJ software (NIH), as described
Statistical analyses were performed using SPSS 21 (IBM, Armonk, NY) and JMP (SAS, Cary, NC). Differences among multiple means were assessed by one-way or two-way ANOVA followed by post-hoc comparisons between groups by Tukey-Kramer test. Differences between two means were assessed with the Student's t-test. Learning curves in the Barnes maze were assessed with a linear mixed effects model and fitted using the SPSS package MIXED. The model included the following effects:
In humans, most concussive injuries occur in the anterior-posterior axis of the brain, a process that was not adequately simulated by previously available mouse models. In contrast, the Maryland TBI model for rats does simulate an angular, frontal impact and produces robust behavioral deficits
Spatial learning and memory were assessed four weeks after the injury with the Barnes maze test. During the acquisition phase of this test, 1-hit mice learned the task as well as controls (
Wildtype mice (n = 9–10 per group) received a 1-hit or a 2-hit frontal impact injury or sham treatments, and were tested in the Barnes maze one month later. (A, B) Learning curves of the 1-hit (A) and 2-hit (B) groups, reflecting the time it took mice to find the target, averaged from 2 trials per day. Only the 2-hit group differed significantly from sham-treated controls (p = 0.0062 by linear mixed effects model analysis). (C) The 2-hit group and sham-treated controls showed a comparable latency to target during the first trial on the first training day. (D) Probe trial administered 24 h after the last training trial. *p<0.05, **p<0.01 vs corresponding sham group or as indicated by bracket. Sh, Sham; H, Hit. Data are means ± SEM.
Learning and memory retention were further assessed in a probe trial administered 24 hours after the last training trial. Whereas 1-hit mice performed at control levels, 2-hit mice were impaired, taking significantly longer to reach the target hole than controls and 1-hit mice (
To compare their fear responses and associative learning and memory, we tested 2-hit mice and controls in a 3-day fear conditioning paradigm 4 weeks after injury. Similar results were obtained in both groups of mice in regards to all outcome measures examined (
Wildtype mice (n = 8–10 per group) received a 2-hit frontal impact injury or sham treatment, and underwent cued and contextual fear conditioning 1 month later. (A) On day 1, the sham-treated and 2-hit groups showed comparable amounts of freezing at baseline as well as during and between training trials. BL, baseline (3 min); CS, conditional stimulus (auditory stimulus followed by foot shock); IT, interval between CS (2 min); End, period following last CS (2 min). See
Wildtype mice (n = 8–10 per group) received a 2-hit frontal impact injury or sham treatment, followed by assessment in different behavioral tests. (A–D) Open field activity 2 weeks post-injury. (E–H) Behavior in elevated plus maze 4 weeks post-injury. Student's t test revealed no significant differences between the 2-hit and sham treated groups for any of the tests and outcome measures. Data are means ± SEM.
Wildtype mice (n = 8–10 per group) received a 2-hit frontal impact injury or sham treatment, followed by behavioral assessment. (A–B) Balance beam performance 3 weeks post-injury. (C–E) Y-maze activity 2 weeks post-injury. (F–G) Forced swim test 5 days (F) and 6 months (G) post-injury. Student's t test revealed no significant differences between the 2-hit and sham treated groups for any of the tests and outcome measures. Data are means ± SEM.
Some 2-hit mice and controls (n = 6 per group) were subjected to brain imaging by 7T MRI 2 days after the second injury. This analysis revealed no evidence for contusions, hemorrhage, ischemia, or hippocampal injury in 2-hit mice (
Wildtype mice (n = 6 per group) received a 2-hit frontal impact injury or sham treatment. Their brains were imaged by MRI two days after the second injury. (A–B) Representative T1 weighted coronal images of sham treated (A) and 2-hit injured (B) mice.
To assess the effects of complete and partial Tau reduction on functional deficits caused by 2-hit frontal impact, we subjected 4–6-month-old
Similar results were obtained in a probe test administered 24 hours after the last training trial, with 2-hit mice showing significant impairments relative to sham-injured controls on the
To determine whether Tau reduction is also beneficial in another model of TBI, we subjected an independent cohort of 4–6-month-old
To assess whether 2-hit frontal injury leads to long term neurodegenerative changes, we analyzed a cohort of
To determine whether 2-hit frontal injury causes prolonged astrocytosis or microgliosis, we obtained brain sections from wildtype mice 12 months after injury and immunostained them for the astroglial marker GFAP or the microglial marker Iba-1. Two-hit and 2-sham mice showed similar levels of GFAP and Iba-1 immunoreactivity in the corpus callosum (
Wildtype mice (n = 8–10 per group) received a 2-hit frontal impact injury or sham treatment, underwent behavioral testing 2–6 weeks later, and were analyzed histologically 12 months after the initial injury. Coronal brain sections were immunostained for GFAP or Iba-1(A, B) Representative images (A) and quantitation (B) of GFAP immunoreactivity in the corpus callosum. (C, D) Representative images (C) and quantitation (D) of Iba-1 immunoreactivity in the corpus callosum. Sh, Sham, H, Hit. Scale bar: 80 µm. Data are means ± SEM.
Our study demonstrates that partial reduction of endogenous tau can protect against spatial learning and memory deficits and chronic axonopathy caused by mild TBI. These findings were obtained in a new mouse model of mild, repetitive frontal impact injury that we developed based on a rat model reported by Kilbourne et al.
Our impact model has several advantages over existing mouse models of mild TBI. First, it generates a frontal impact that does not cause radiological abnormalities on MRI. Second, it eliminates complications that can be associated with skin incision and craniotomy, including infection, unintended cerebral injury, and postoperative wound pain. Third, preservation of the calvarium and overlying soft tissue also increases the options for employing diagnostic and therapeutic interventions such as electroencephalographic (EEG) monitoring, placements of drug infusion pumps, and radiological imaging.
During injury, the vector of force occurs along the anterior-posterior plane of the head and parallel to the skull base. This directionality reproduces common human injury mechanisms occurring during falls, motor vehicle accidents, and sports- or combat-related concussions. Unlike other models that impact the skull near the vertex, the new model does not compress the brain against the skull base, minimizing collateral injury to the brainstem and cerebellum that can cause confounding deficits in survival, coordination and balance. Like many forms of human TBI, it primarily impacts frontal brain structures and creates brain shifts as well as linear and angular shearing forces that place long, white matter tracts at risk
It is interesting in this regard that reducing tau, which can be detected in dendritic structures but is located primarily within axons
Neuropathologically, our 2-hit model caused prolonged axonopathy, detectable in the corpus callosum of wildtype mice 12 months after injury, consistent with previous studies demonstrating neurodegenerative alterations after mild TBI in mice
From a therapeutic perspective, it is encouraging that even partial reduction of tau was able to prevent spatial learning and memory deficits and neurodegenerative changes after mild, repetitive frontal impact injury, particularly in light of the recent demonstration that acute cerebral tau reduction in wildtype mice with antisense oligonucleotides is both feasible and well tolerated
(TIF)
We thank Kaspar Kaledjian, M.D., Vladimir V. Gerzanich, Ph.D., and J. Marc Simard, M.D., Ph.D., for advice on the frontal impact model, John Carroll for preparation of graphics, Sharon Lee for technical assistance, and Amy Cheung and Monica Dela Cruz for administrative assistance.