Recovery of Neurological Function Despite Immediate Sleep Disruption Following Diffuse Brain Injury in the Mouse: Clinical Relevance to Medically Untreated Concussion

Abstract

Study Objective:

We investigated the relationship between immediate disruption of posttraumatic sleep and functional outcome in the diffuse brain-injured mouse.

Design:

Adult male C57BL/6 mice were subjected to moderate midline fluid percussion injury (n = 65; 1.4 atm; 6-10 min righting reflex time) or sham injury (n = 44). Cohorts received either intentional sleep disruption (minimally stressful gentle handling) or no sleep disruption for 6 h following injury. Following disruption, serum corticosterone levels (enzyme-linked immunosorbent assay) and posttraumatic sleep (noninvasive piezoelectric sleep cages) were measured. For 1-7 days postinjury, sensorimotor outcome was assessed by Rotarod and a modified Neurological Severity Score (NSS). Cognitive function was measured using Novel Object Recognition (NOR) and Morris water maze (MWM) in the first week postinjury.

Setting:

Neurotrauma research laboratory.

Measurements and Results:

Disrupting posttraumatic sleep for 6 h did not affect serum corticosterone levels or functional outcome. In the hour following the first dark onset, sleep-disrupted mice exhibited a significant increase in sleep; however, this increase was not sustained and there was no rebound of lost sleep. Regardless of sleep disruption, mice showed a time-dependent improvement in Rotarod performance, with brain-injured mice having significantly shorter latencies on day 7 compared to sham. Further, brain-injured mice, regardless of sleep disruption, had significantly higher NSS scores postinjury compared with sham. Cognitive behavioral testing showed no group differences among any treatment group measured by MWM and NOR.

Conclusion:

Short-duration disruption of posttraumatic sleep did not affect functional outcome, measured by motor and cognitive performance. These data raise uncertainty about posttraumatic sleep as a mechanism of recovery from diffuse brain injury.

Citation:

Rowe RK; Harrison JL; O'Hara BF; Lifshitz J. Recovery of neurological function despite immediate sleep disruption following diffuse brain injury in the mouse: clinical relevance to medically untreated concussion. SLEEP 2014;37(4):743-752.

INTRODUCTION

Traumatic brain injury (TBI) is a major cause of death and disability throughout the world, with few pharmacological treatments available for individuals suffering from the lifelong neurological morbidities associated with TBI. Vascular, cellular, and molecular pathological processes initiated at the time of injury can compound the injury and manifest into functional impairments. In the United States alone, the Centers for Disease Control and Prevention estimated that between the years 2002 and 2006 there were on average 52,000 deaths, 275,000 hospitalizations, and 1,365,000 emergency department visits related to TBI each year.¹ Beyond this, it is estimated that as high as 42% of TBIs are not included in these statistics, because approximately 1.2-4.2 million survivors of mild TBI do not seek medical attention.² A common neurological consequence of mild TBI is excessive sleepiness immediately following injury.³ However, discordant opinions suggest that individuals should not be allowed to sleep or should be frequently awakened following mild brain injury. The intentional sleep disruption used outside of medical care (e.g., at home) contrasts the unintentional sleep disruption associated with diagnostic procedures in a clinical setting. This controversy is not supported by peer-reviewed biomedical literature, and the effect of sleep disruption immediately following TBI upon functional recovery is not understood.

For the first time, this study investigates the contribution of acute posttraumatic sleep on the recovery of neurological function after diffuse TBI. Posttraumatic sleep may be beneficial to recovery from injury, because prevailing hypotheses suggest the function of sleep is restorative, conservative, and adaptive.^4,5 To investigate the relationship between TBI and acute sleep, the current study uses gentle handling to disrupt sleep after midline fluid percussion injury (mFPI), an animal model of concussion.⁶ Following mFPI, with and without sleep disruption, mice can be evaluated for performance in cognitive, neurological, and motor function, using standard behavioral tests.^7–9

Our previous findings indicated an increase in sleep during the first 6 h following diffuse brain injury in mice—a period we have defined as posttraumatic sleep.¹⁰ During this period, brain-injured mice are responsive, capable of movement, and eat and groom themselves, indicating they are not in a comatose state of unresponsiveness. Following TBI, secondary injury processes, including the production of cytokines, are triggered, some of which have dual roles as sleep regulatory substances.^11–13 Similar increases in cytokine signaling have been observed across experimental models and in human TBI, highlighting their involvement in pathological and reparative processes triggered by injury.^14–17 The increases in sleep-promoting cytokines suggest posttraumatic sleep is a natural process. Whether this natural process is beneficial to functional outcome remains to be seen.

Clinical recommendations and at-home practices with regard to sleep after TBI cover an array of interventions including total deprivation, frequent awakening, and encouraging sleep. Experimentally, multiple techniques can disrupt or deprive sleep. Sleep deprivation is the complete disruption of one or both types of sleep (rapid eye movement [REM], nonrapid eye movement [NREM]), compared with sleep disruption, in which minimal sleep may occur. Deprivation of REM sleep can be achieved using methods that operate when muscle tone is lost as REM sleep begins. In these methods, muscle tone is required for an animal to balance on a platform in water; with the onset of atonia, the animal would fall off the platform into water and awaken. Frequently used REM deprivation methods include the flower pot and variations of the multiple platform method,¹⁸ which also may compromise slow wave sleep.¹⁹ Acute injury-induced motor deficits prevented the use of these deprivation methods. Thus, total disruption of posttraumatic sleep was implemented, allowing only minimal sleep during the disruption period. Experimental methods for disrupting sleep include forced wheel running and gentle handling. Investigations have shown forced and voluntary exercise can positively and negatively affect behavioral and histological outcomes following brain injury.^20–25 To disrupt sleep, we used the gentle handling method along with cage tapping whenever animals began falling asleep,²⁶ which disrupts all stages of sleep.

We are extending our investigations into sleep as a natural response to TBI. Increased posttraumatic sleep, paired with inconsistent recommendations for sleep after TBI, make understanding the role of posttraumatic sleep an important public health concern. Understanding the effect of posttraumatic sleep on functional outcome could inform home care recommendations for the large number of TBI survivors not seeking medical attention. We hypothesize that posttraumatic sleep disruption would result in poor functional outcome following TBI. To test this hypothesis, we used mFPI to model diffuse TBI in mice, disrupted posttraumatic sleep for 6 h postinjury, and assessed cognitive, motor, and sensorimotor functional outcome over 1 w postinjury. The goal of this study is to add knowledge for evidence-based clinical recommendations in the treatment of mild TBI.

METHODS

Animals

Male C57BL/6 mice (Harlan Laboratories, Inc., Indianapolis, IN) were used for all experiments (n = 109). The animals were housed in a 12 h light/12 h dark cycle at a constant temperature (23°C ± 2°C) with food and water available ad libitum according to the Association for Assessment and Accreditation of Laboratory Animal Care International. Animals were acclimated to their environment following shipment for at least 3 days prior to any experiments. After surgery, animals were evaluated daily for postoperative care by a physical examination and documentation of each animal's condition. Animal care was approved by the Institutional Animal Care and Use Committees at the University of Kentucky and the St. Joseph's Hospital (Phoenix, AZ).

Housing

All mice used in this study were singly housed. Mice used for the Novel Object Recognition (NOR) and Morris water maze (MWM) studies were housed in standard individually ventilated cages. Mice for all other studies were housed in the noninvasive sleep-monitoring cage system (Signal Solutions, Lexington, KY).

Midline Fluid Percussion Injury

Mice (20-24 g) were subjected to mFPI consistent with methods previously described.²⁷ Animal numbers are indicated in the Results section and figure legends for individual studies. Mice were anesthetized using 5% isoflurane in 100% oxygen for 5 min and the head of each animal was placed in a stereo-taxic frame with continuously delivered isoflurane at 2.5% via nosecone. While the animal was anesthetized, body temperature was maintained using a Deltaphase^® isothermal heating pad (Braintree Scientific Inc., Braintree, MA). A midline incision was made exposing bregma and lambda, and fascia was removed from the surface of the skull. A trephine (3 mm outer diameter) was used for the craniotomy, centered on the sagittal suture between bregma and lambda without disruption of the dura. An injury cap prepared from the female portion of a Luer-Lok needle hub (Becton Dickinson, Franklin Lakes, NJ) was fixed over the craniotomy using cyanoacrylate gel and methyl methacrylate (Hygenic Corp., Akron, OH). The incision was sutured at the anterior and posterior edges and topical lidocaine ointment was applied. The injury cap was closed using a Luer-Lok cap (Becton Disckinson, Franklin Lakes, NJ) and animals were placed in a heated recovery cage and monitored until ambulatory before being returned to the sleep cage.

For injury induction 24 hours postsurgery, animals were reanesthetized with 5% isoflurane delivered for 5 min. The cap was removed from the injury hub assembly, and the craniotomy was visually inspected through the hub. The hub was then filled with normal saline and attached to a tube connected to the male end of the fluid percussion device (Custom Design and Fabrication, Virginia Commonwealth University, Richmond, VA). An injury of moderate severity (1.4 atm) was administered by releasing the pendulum onto the fluid-filled cylinder. Sham-injured animals underwent the same procedure, except the pendulum was not released. Animals were monitored for the presence of a forearm fencing response and righting reflex times were recorded for the injured animals as indicators of injury severity.²⁸ The righting reflex time is the total time from the initial impact until the animal spontaneously rights itself from a supine position. The fencing response is a tonic posturing characterized by extension and flexion of opposite arms that has been validated as an overt indicator of injury force magnitude.²⁸ The injury hub was removed and the brain was inspected for uniform herniation and integrity of the dura. Animals in whom the dura was compromised were excluded from all studies as technical failures. The incision was cleaned using saline and closed using sutures. Moderate brain-injured animals had righting reflex recovery times greater than 6 min and a positive fencing response. Sham-injured animals recovered within 20 sec. After spontaneous righting, animals were placed in a heated recovery cage and monitored until ambulatory (approximately 5 to 15 min) before being returned to the sleep cage or subjected to sleep disruption. Adequate measures were taken to minimize pain or discomfort.

Sleep Disruption

Mice were randomly assigned to a sleep disruption or no sleep disruption group. The mice in the no sleep disruption group were returned to their individual cages as soon as they were ambulatory (approximately 5 to 15 min) following the brain injury or sham injury. Mice in the sleep disruption group were placed in individual cages and continuously sleep disrupted for 6 h postin-jury, the duration over which the mFPI mice sleep in excess of sham controls, as previously observed in our model.¹⁰ Sleep disruption was accomplished using a minimally stressful gentle handling method,^26,29,30 which included tapping on the cages or gently touching the animal when visible signs of sleep (immo-bile, eyes closed) were present.²⁶ After the disruption period, mice were returned to their individual cages and sleep activity was measured continuously for 24 h (see following paragraphs).

Sleep Recordings

The noninvasive sleep cage system (Signal Solutions, Lexington, KY) used in this study consisted of 16 separate units that could simultaneously monitor the sleep and wake states, as previously published.¹⁰ Each cage unit housed a single mouse inside 7 × 7-inch walled compartments with attached food and water structures.³¹ The cages had open bottoms resting on polyvinylidine difluoride (PVDF) sensors serving as the cage floor.³¹ The noninvasive high-throughput PVDF sensors were coupled to an input differential amplifier, and pressure signals were generated and classified by an algorithm (see following text) as motions consistent with either wake activity or the inactivity and regular breathing movements associated with sleep.³¹ Briefly, sleep was characterized primarily by periodic (3 Hz) and regular amplitude signals recorded from the PVDF sensors, typical of respiration from a still mouse. In contrast, signals characteristic of wake were both the absence of characteristic sleep signals and higher amplitude, irregular spiking associated with volitional movements. The piezoelectric signals in 2-sec epochs were classified by a linear discriminant classifier algorithm based on frequency and amplitude to assign a binary label of “sleep” or “wake”.³¹ Mice sleep in a polycyclic manner (more than 40 sleep episodes per hour).³² For experimental studies, mouse sleep was quantified as the minutes spent sleeping per hour, presented as a percentage for each hour. Sleep activity data were binned over specified time periods (e.g. 5 min, 1 h) to calculate the average of percent sleep. Data were binned by length of individual sleep bouts to calculate median bout length.

Corticosterone Assay

Immediately following 6 h of sleep disruption, mice were euthanized (between 14:00 and 17:00) by an overdose of sodium pentobarbital, and a cardiac blood sample was collected. A separate group of animals was not exposed to sleep disruption prior to blood collection. The blood samples were centrifuged (3,000 rpm, 8 min) and the serum was stored at -20°C. A commercially available competitive immunoassay was followed according to the manufacturer's protocol for the quantitative determination of corticosterone (no. 900-097; Assay Designs, Inc. Ann Arbor, MI). The kit uses an anti-corticosterone polyclonal antibody to bind standards and samples. The enzyme reaction generates a yellow color that is inversely proportional to the corticosterone concentration and is read on a microplate reader (405 nm). All samples were diluted 1:5 (80%) in order to stay within the sensitivity of the assay (32-20,000 pg/mL). Data are presented as levels of serum corticosterone concentration in ng/mL.

Behavioral Testing

Behavioral testing was performed on two cohorts of animals. One cohort was tested on the Rotarod (Economex Rotarod system, Columbus Instruments, Columbus, OH) and Neurological Severity Score (NSS). A separate cohort was tested by NOS and the MWM.

Neurological Severity Score

Posttraumatic neurological impairments were assessed using an eight-point NSS paradigm adapted from those previously used in experimental models of TBI.^16,33–35 Animals were tested at selected time points postinjury (1, 3, 5, and 7 days). One point was given for failure on an individual task, and no points were given if an animal completed a task successfully. Mice were observed for hindlimb flexion, startle reflex, and seeking behavior (presence of these behaviors was considered successful task completion). Mice traversed in sequence, 3-, 2-, and 1-cm beams. The beams were elevated and mice were given 1 min to travel 30 cm on the beams. The task was scored as a success if the mouse traveled 30 cm with normal forelimb and hindlimb position (forelimb/hindlimb did not hang from the beam). Mice were also required to balance on a 0.5-cm beam and a 0.5-cm round rod for 3 sec in a stationary position with front paws between hind paws. Nonparametric data are presented as a composite score ranging from 1 to 8 representing performance on all tasks combined. High final NSS scores were indicative of task failure and interpreted as neurological impairment.

Rotarod

Sensorimotor function was assessed using the Economex Rotarod system from Columbus Instruments. Animals were pretrained (60 sec at 4 rpm for three trials) for 3 consecutive days including the day of the craniotomy. Animals were then tested at selected time points postinjury. Animals were tested either immediately and 1 hour after injury or tested on postin-jury days 1, 3, 5, and 7. For the test, animals were placed on the rod with a starting speed of 4 rpm, and the speed was continuously accelerated up to 28 rpm over 5 min (for two trials), as previously published.³⁶ The trial ended when the mouse fell from the rod or 5 min elapsed. Data are presented as latency to fall in seconds and total distance traveled in centimeters.

Morris Water Maze

Learning ability was assessed in the MWM using a paradigm similar to those previously used in experimental models of TBI.^35,37–40 The 1-m diameter MWM was filled with water (19-21°C) and nontoxic white paint (Rich Art Co., Northvale, NJ) was added to hide the platform (6.3 cm diameter) that was submerged 0.5 cm. At selected time points postinjury (3, 4, 5, and 6 days), mice were tested in sets of four trials per day. Mice started from one of four starting points (north, south, east, west) and used visual cues placed on the walls outside the tank to locate the platform. All trials were monitored using overhead video/tracking software (EZVideo version 5.51DV, Accuscan Instruments Inc., Columbus, OH). The latency of the mouse to find the platform was recorded as well as the distance traveled. If a mouse did not find the platform within the 70-sec trial, it was placed on the platform for 10 sec. Data are presented as latency to find the hidden platform in seconds.

Novel Object Recognition

Cognitive impairment was tested using the NOR test, as previously published.^41,42 The test consisted of three phases: habituation, training, and testing. On day 3 postinjury, mice were placed in an open field (42 cm, 21 cm, and 21 cm) for 1 h of habituation. Mice were removed and two identical objects were placed in opposing quadrants of the field for the training phase. Mice were placed in the center of the open field and given 5 min to explore the objects. Following training, mice were returned to their home cages. Testing began 4 h after training. One familiar object and one novel object were placed in opposing quadrants of the field. Mice were placed into the center and given 5 min to explore. On day 7 postinjury, mice were given 10 min of habituation to their previously used open field. After habituation, mice were removed and the familiar object from training and a novel object (distinct from the object on day 3) were placed in opposing quadrants of the field and mice were given 5 min to explore. For training and testing the percentage of time spent exploring the novel object was quantified. Exploration of an object included the mice sniffing, touching, or climbing onto an object while facing the object. If an animal climbed onto an object and sniffed into the air, this time was not calculated into the exploration of the novel object. During training and testing trials, mice were required to spend a minimum combined 10 sec exploring objects. If this time was not met, trial time was extended for that animal until 10 sec of exploration was achieved (sham: 1 of 12, 30 sec; FPI: 5 of 24, mean 61 ± 17.5 sec). Data are presented as percent of total exploration time spent exploring the novel object.

Statistical Analysis

Data are shown as mean ± standard error of the mean and were analyzed using Prism 6 (Graphpad Software, La Jolla, CA). Differences in righting reflex times were determined by t-test. Differences in Rotarod performance immediately following TBI were determined with a repeated measure two-way analysis of variance (ANOVA) followed by Sidak multiple comparison test. Percent sleep following disruption was analyzed using a repeated-measures two-way ANOVA. Differences in rebound sleep were determined using a one-way ANOVA followed by Tukey post hoc analysis. Differences in functional performance over time post-injury measured by the Rotarod, MWM, and NOR all were determined using a two-way ANOVA, followed by Tukey post hoc analysis as needed. Nonparametric NSS data were analyzed by Kruskal-Wallis ANOVA, followed by Dunn comparison post hoc test (see Results). Statistical significance was assigned when P < 0.05.

RESULTS

Immediate Neurological Deficits Following Diffuse TBI

We have previously reported a suppression of the righting reflex response in rats following mFPI,²⁸ as an injury-induced deficit. Diffuse brain injury resulted in a significant suppression of the righting reflex in brain-injured mice compared with anesthetized, uninjured shams (t₃₁ = 3.351, P = 0.0021; sham n = 28, injury n = 33; Figure 1A). To assess acute vestibulomotor deficits following diffuse TBI, we used the Rotarod task immediately and 1 h after the return of the righting reflex.^43–45 There was a significant injury-dependent motor deficit measured by latency on the Rotarod task in brain-injured mice compared to uninjured shams (F(1, 11) = 83.93, P < 0.0001; Figure 1B) both immediately and 1 h after return of the righting reflex (sham n = 6, injury n = 7; Figure 1B).

Figure 1.

Diffuse traumatic brain injury (TBI) led to immediate neurological deficits. (A) Immediately following experimental diffuse TBI in mice, there was significant suppression of the righting reflex (mean ± standard error of the mean [SEM]; t(31) = 3.351, P = 0.0021; sham n = 28, injury n = 33). (B) Immediately following diffuse TBI there was a significant motor deficit measured by latency in the Rotarod task. A repeated-measures two-way analysis of variance showed a significant decrease in latency on the Rotarod in brain-injured mice compared with uninjured shams (mean ± SEM; F(1, 11) = 83.93, P < 0.0001). Sidak multiple comparison test showed a significant difference between brain-injured mice compared with uninjured shams both immediately after injury, and 1 h after injury (asterisk, P < 0.05; sham n = 6, injury n = 7).

Intentional Sleep Disruption Following Diffuse TBI Did Not Result in a Rebound of Lost Sleep But Alters Activity Response to Dark Onset

Corticosterone levels were measured at the conclusion of the 6-h disruption period as an indicator of stress related to the sleep disruption. Sleep disruption did not significantly alter corticosterone levels in cardiac blood samples in either sleep disruption shams or sleep disruption injured mice impaired to no disruption shams and no disruption brain-injured mice (Figure 2A). Brain-injured mice had significantly lower levels of corticosterone compared with uninjured shams regardless of sleep disruption at the conclusion of the testing period (F(1,8) = 7.57, P = 0.0250; Figure 2A). As intended, the sleep disruption method developed for these studies did not adversely affect corticosterone levels.

Figure 2.

Intentional sleep disruption following diffuse traumatic brain injury (TBI) did not result in altered corticosterone levels or a rebound of lost sleep but altered activity response to dark onset. (A) Sleep disruption for 6 h following diffuse TBI did not alter levels of corticosterone in cardiac blood samples. In both no disruption and sleep disruption groups, the brain-injured mice (black bars) had significantly lower levels of corticosterone compared with uninjured shams (white bars) (mean ± standard error of the mean [SEM]; F(1,8) = 7.57, P = 0.0250). There was no significant difference between sleep disruption and no disruption groups. (B) Following 6 h of intentional sleep disruption (see Methods), a two-way analysis of variance (ANOVA) showed no significant change in percent sleep between groups (mean ± SEM; F(3, 32) = 2.187, P = 0.1087) indicating no rebound of lost sleep. Bar indicates light/dark transition. Box in panel B is enlarged in panel C. (C) At dark onset, a one-way ANOVA showed an effect of sleep disruption (mean ± SEM; F(3, 32) = 0.9386, P = 0.0024). Tukey post hoc test indicated a difference between no disruption sham compared to both sleep disruption brain-injured and sleep disruption sham mice (asterisk, P < 0.05; no disruption sham n = 7, no disruption injury n = 8, sleep disruption sham n = 8, sleep disruption injury n = 13).

Following 6 h of intentional sleep disruption, using the gentle handling method,²⁶ there was no significant change in percent sleep over 6 h between sleep disruption brain-injured and sleep disruption shams compared with no disruption brain-injured and no disruption shams (F(3, 32) = 2.187, P = 0.1087; no disruption sham n = 7, no disruption injury n = 8, sleep disruption sham n = 8, sleep disruption injury n = 13; Figure 2B). This indicated that the sleep lost during the 6 h of intentional disruption was not recovered by the sleep-disrupted groups, at least in terms of total sleep time. However, there was a sleep disruption effect on activity response to dark onset at 10 h postinjury (19:00) (F(3, 32) = 0.9386, P = 0.0024; Figure 2C). There was a significant difference between no disruption sham compared with both sleep disruption brain-injured and sleep disruption sham mice (Figure 2C). Both the sleep disrupted brain-injured and sleep disrupted shams slept significantly more following dark onset compared with the no disruption brain-injured and no disruption shams. In the absence of sleep disruption, uninjured and brain-injured animals showed increased wake activity following dark onset, as is typical for nocturnal rodents.

Diffuse TBI Resulted in Neurological Impairments Independent of Acute Sleep Disruption

Overall, brain-injured mice showed significant neurological impairments measured by the NSS compared with uninjured shams, independent of sleep disruption (no disruption sham n = 12, no disruption injury n = 13, sleep disruption sham n = 10, sleep disruption injury n = 13; Figure 3). On postin-jury days 1, 3, and 5, both sham groups had significantly lower NSS scores compared with both brain-injured groups (Day 1 KW(4,48) = 27.45, P < 0.0001; Day 3 KW(4,48) = 18.99, P = 0.0003; Day 5 KW(4,48) = 15.63, P = 0.0013; Day 7 KW(4,48) = 16.36, P = 0.001; Kruskal-Wallis statistic with Dunn comparison post hoc; Figure 3). There was no significant difference in neurological impairments measured by the NSS between the sleep disruption brain-injured mice and no disruption brain-injured mice at any postinjury time point.

Figure 3.

Diffuse traumatic brain injury resulted in neurological impairments independent of acute sleep disruption. Overall, brain-injured mice showed significant neurological impairments measured by the Neurological Severity Score (NSS) compared with uninjured shams independent of sleep disruption. Sleep disruption shams showed a significantly lower NSS score compared to both injury groups (sleep disruption injury, no sleep disruption injury) on postinjury days 1,3, and 5 (asterisk; P < 0.05). On postinjury day 7, sleep disruption shams had a significantly lower NSS score compared to sleep disruption brain-injured mice (asterisk; P < 0.05). No disruption shams had a significantly lower NSS score compared with both injury groups (sleep disruption injury, no disruption injury) on postinjury days 1 and 7 (number sign; P < 0.05). No disruption shams had significantly lower NSS scores compared with no disruption brain-injured mice on postinjury days 3 and 5 (number sign; P < 0.05). See Results for statistics. (No disruption sham n = 12, no disruption injury n = 13, sleep disruption sham n = 10, sleep disruption injury n = 13).

Diffuse TBI Reduced Motor Performance on the Rotarod Task Independent of Acute Sleep Disruption

To assess motor function we used the Rotarod task, as previously published.³⁶ Following diffuse brain injury, there was a significant injury-dependent effect on latency to stay on the Rotarod (F(3, 44) = 3.367, P = 0.0268) and a time-dependent improvement in latency (F(3, 132 = 41.60, P < 0.0001; Figure 4A; no disruption sham n = 12, no disruption injury n = 13, sleep disruption sham n = 10, sleep disruption injury n = 13). There was a significant injury and disruption effect on Rotarod latency between sleep disruption brain-injured mice compared with no disruption sham on postinjury day 7 (Figure 4A). There was no significant difference in latency on the Rotarod task between the sleep disruption brain-injured mice and no disruption brain-injured mice (F(1,24) = 0.5033, P = 0.4849). Further analysis of Rotarod performance showed a significant injury-dependent effect on distance traveled (F(3, 44) = 4.009, P = 0.0132) and a time-dependent improvement in distance traveled (F(3, 132) = 34.61, P < 0.0001; Figure 4B). There was a significant injury effect between no disruption sham compared with both injury groups (no disruption and sleep disruption) on postinjury days 5 and 7. There was no significant difference in distance traveled on the Rotarod task between the sleep disruption brain-injured mice and no disruption brain-injured mice (F(1,24) = 0.2592, P = 0.6154).

Figure 4.

Diffuse traumatic brain injury altered motor performance on the Rotarod task independent of acute sleep disruption. (A) A two-way analysis of variance (ANOVA) showed a significant injury-dependent effect on latency to stay on the Rotarod (mean ± standard error of the mean [SEM]; F(3, 44) = 3.367, P = 0.0268) and a time-dependent improvement in latency (mean ± SEM; F(3, 132 = 41.60, P < 0.0001). Tukey post hoc test indicated a difference between sleep disruption brain-injured mice compared to no disruption sham on postinjury day 7 (asterisk, P < 0.05). There was no significant difference in latency on the Rotarod task between the sleep disruption brain-injured mice and no disruption brain-injured mice (F(1,24) = 0.5033, P = 0.4849). (B) A two-way ANOVA showed a significant injury-dependent effect on distance traveled (mean ± SEM; F(3, 44) = 4.009, P = 0.0132) and a time-dependent improvement in distance traveled (mean ± SEM; F(3, 132) = 34.61, P < 0.0001). Tukey post hoc test indicated a difference between no disruption sham compared to both injury groups (no disruption and sleep disruption) on postinjury days 5 and 7 (asterisk, P < 0.05). There was no significant difference in distance traveled on the Rotarod task between the sleep disruption brain-injured mice and no disruption brain-injured mice (F(1,24) = 0.2592, P = 0.6154). (No disruption sham n = 12, no disruption injury n = 13, sleep disruption sham n = 10, sleep disruption injury n = 13).

Acute Sleep Disruption Following Diffuse TBI Did Not Alter Cognitive Performance

Performance on the MWM task indicated all groups had a significant time-dependent improvement in latency to find the hidden platform (F(3, 96) = 17.40, P < 0.0001; Figure 5A; no disruption sham n = 6, no disruption injury n = 12, sleep disruption sham n = 6, sleep disruption injury n = 12). There was no significant injury-dependent effect or sleep disruption effect on cognitive performance measured by latency to find the platform at the selected acute postinjury time points (days 3, 4, 5, 6). Performance on the novel object recognition task showed no significant injury-dependent effect or sleep disruption effect on recall of the familiar object at the selected acute postinjury time points (days 3, 7). All groups showed signifi-cant increase in novel object exploration at 3 days postinjury compared to training (F(2,64) = 15.61, P < 0.001; Figure 5B; no disruption sham n = 6, no disruption injury n = 12, sleep disruption sham n = 6, sleep disruption injury n = 12). There was no significant injury effect or sleep disruption effect on novel object recognition.

Figure 5.

Acute sleep disruption following diffuse traumatic brain injury did not alter cognitive performance on the Morris water maze or Novel Object Recognition (NOR) tasks. (A) A two-way analysis of variance showed a significant time-dependent improvement in latency to the platform (mean ± standard error of the mean [SEM]; F(3, 96) = 17.40, P < 0.0001). There was no significant injury-dependent effect or sleep disruption effect on cognitive performance at the selected acute postinjury time points. (No disruption sham n = 6, no disruption injury n = 12, sleep disruption sham n = 6, sleep disruption injury n = 12). (B) There was no significant injury-dependent effect or sleep disruption effect on cognitive performance at the selected acute postinjury time points. All groups showed a time-dependent learning of the familiar object (mean ± SEM; F(2,64) = 15.61, P < 0.001). There was no significant injury effect or disruption effect. (No disruption sham n = 12, no disruption injury n = 13, sleep disruption sham n = 10, sleep disruption injury n = 13).

DISCUSSION

In the diffuse brain-injured mouse, immediate disruption of posttraumatic sleep does not worsen injury-induced motor or cognitive deficits. Our previous studies showed a signifi-cant increase in percent sleep of brain-injured mice compared to uninjured shams over the first 6 h following diffuse TBI.¹⁰ Disrupting the 6 h of posttraumatic sleep was hypothesized to worsen functional outcome after midline fluid percussion injury, because sleep, in general, is reparative and restorative.^4,5,46–48 In the current study, we show that gentle handling for 6 h postin-jury to disrupt sleep had no effect on stress measured by corticosterone levels at the end of the disruption period and did not result in rebound of lost sleep or worsened functional outcome. It remains possible that sleep disruption leads to an early surge in corticosterone, which was missed by measuring after 6 h of disruption, regardless of functional consequences. Our data provide the first evidence that sleep disruption (immediate, short duration) does not affect functional outcome following diffuse brain injury in mice.

Currently, investigations into acute sleep disruption after TBI are lacking, despite the common practice of disturbing sleep acutely following a concussion. Previous attempts to investigate sleep disruption as a neuroprotective intervention following TBI have been presented in short communication, but methodological issues compromise the interpretation of the data.⁴⁹ Following ischemia, however, acute intervention has been shown to be neuroprotective in the rat.⁵⁰ Whisker stimulation in the immediate postinfarct period (within the first hour) can protect the cortex, suggesting a role for cortical activation through sensory stimulation in determining outcome following middle cerebral artery occlusion.⁵⁰ Contradictory data after focal cerebral ischemia in the rat show that sleep deprivation for 12 h using a gentle handling protocol increased infarct area.⁵¹ Also, sleep disturbance (12 hours a day for 3 consecutive days) following focal cerebral ischemia in the rat worsened behavioral outcome assessed using the single pellet-reaching test through 35 days postischemia.⁵² The findings of these studies suggest a role of sleep modulating recovery processes following stroke.^51,52 Thus, sleep disturbance immediately following ischemia can worsen outcome, but activating discrete circuits through whisker stimulation can be neuroprotective,^50–52 suggesting that additional systemic effects of behavioral sleep deprivation (e.g., elevated temperature and blood pressure) may counteract protective effects of local circuit activation. In this study, we pursued sleep disruption in a clinically relevant manner to model the acute period of posttraumatic sleep disruption and found limited effect on functional outcome.

In midline fluid percussion, injury-induced histopathology is uncomplicated by contusion, cavitation, or overt hemorrhage^53,54 and microscopically the injury is characterized by traumatic axonal and vascular injury.^55–60 Diffuse brain injury also leads to significant changes in behavioral and histological outcome.^16,36 We have previously shown that TBI increases sleep in mice over the first 6 h postinjury regardless of injury severity.¹⁰ Because our previous data show that mild and moderate diffuse TBI result in equivalent increases in sleep, the current study did not include mild TBI. Sleep bouts in brain-injured mice are longer in duration than in uninjured mice.¹⁰ Our previous report of the raw data produced by the non-invasive sleep monitoring cages showed sleep was interrupted by high amplitude and frequency signals corresponding to volitional movement in both brain-injured mice and uninjured shams.¹⁰ Interruptions of sleep bouts by volitional movement indicate the brain-injured animals terminate sleep bouts in a manner similar to that of uninjured mice, suggesting that brain-injured mice are responsive, capable of movement, and not in a comatose state of unresponsiveness.

For our method of sleep disruption, we used gentle handling to continually disrupt sleep for 6 h. Immediate motor deficits following our diffuse injury model (Figure 1) make it impractical to use the flower pot or multiple platform method for sleep disruption, because the brain-injured animals cannot perform the task (i.e. balance on the flower pot). Previous studies have shown forced and voluntary exercise can positively and negatively affect behavioral and histological outcomes following brain injury,^20–25 and for this reason, animals were not disrupted with forced wheel running. For example, treadmill exercise following fluid percussion injury in the rat reduced injury-induced seizures but did not protect against neuronal injury.²⁵ Similarly, voluntary wheel running following medial frontal cortical contusions in rats exacerbated TBI-induced deficits.²⁴ Our data also indicate that our method of sleep disruption did not significantly alter stress levels, as indicated by stable corticosterone levels at the conclusion of the disruption period. Corticosterone, a glucocorticoid, is released following stress, and exposure to high-level glucocorticoids may affect brain plasticity and recovery.^61,62 To this end, a sleep disruption protocol minimized stress and thereby affect outcomes.

Immediately following the 6-h disruption period, percent sleep was measured to determine whether sleep-disrupted mice displayed a rebound of lost sleep. A previous study using a similar sleep disruption protocol for 6 h prior to ischemia in rats showed a significant rebound of sleep in the dark cycle following insult.⁶³ Sleep was disrupted prior to insult to test if sleep rebound is neuroprotective after ischemia. However, in the current studies sleep-disrupted mice did not show a rebound in sleep to compensate for the loss of 6 h of sleep. Despite the lack of rebounded sleep, short-term disruption was insufficient to adversely affect outcome. It is likely that both rodent and man can recover from transient sleep disruption after brain injury without significant functional consequence. A recent study has shown sleep deprivation prior to injury accelerated recovery from TBI in rats⁶⁴; however, due to lack of suitable controls, it is unclear whether the sleep deprived state at the time of injury, or the sleep recovery after the injury is relevant to the acceleration of recovery. Further studies are needed to determine if sleep deprivation prior to injury is inherently beneficial.

Clinical data suggest injury-induced circadian rhythm disturbances may contribute to the pathophysiological sequelae of TBI^65–67 and inhibit neurogenesis, potentially compromising recovery.⁶⁸ Lateral fluid percussion injury has been shown to alter circadian clock gene expression without producing injury-dependent changes in activity during the first dark cycle postinjury.⁶⁹ Similarly, in our study, we did not observe injury-dependent changes in activity during the first dark cycle. However, sleep- disrupted brain-injured and uninjured shams exhibit a delayed activity response during the hour following the first dark onset postinjury (10 h postinjury), sleeping significantly more than nonsleep-disrupted mice. Sleep disrupted mice slept 10-20% more during this hour, amounting to 6-12 min more sleep than the nonsleep-disrupted mice. Sleep patterns of the disrupted mice become indistinguishable from the nondisrupted mice by the first light onset (data not shown).

We found that sleep disruption immediately following brain injury did not affect functional outcome. It has previously been shown that experimental diffuse TBI results in significant motor impairment.^36,70 We observed injury-induced sensorimotor impairments measured with the Rotarod and NSS tests. Acute sleep disruption, however, neither exacerbated nor attenuated the injury-induced deficits. In our literature evaluation, we found no published reports on the effect of acute sleep disruption on sensorimotor performance in mice.

Similarly, diffuse TBI in mice has been shown to significantly impair cognition as early as 2 days postinjury, with deficits lasting as long as 90 days postinjury.^70,71 However, cognitive outcome measures in mice following midline fluid percussion injury have not been previously reported. No cognitive impairment following diffuse TBI was measured using the MWM or NOR task, within the first week postinjury. Following sleep disruption, there was no significant change in cognitive performance. It has previously been reported that learning in the MWM is not affected by 6 h of rapid eye movement sleep deprivation.⁷² Also, sleep disruption before acquisition had no effect on spatial learning and memory components of rodent cognition.^73,74 These findings are in line with our data indicating 6 h of sleep disruption did not alter cognitive function.

Taken together, our data indicated that acute sleep disruption following diffuse TBI did not worsen functional outcome, which precluded further analysis of cellular repair benefits of sleep. These results were not unexpected, because previous studies have reported that 6 h of sleep disruption did not affect functional outcome independent of brain injury.^72,74 A possible limitation of this study was the disruption of posttraumatic sleep only. Extending the duration of sleep disruption may negatively affect outcome; however, chronic sleep disruption would reduce clinical relevance for the large at-home, nonmedically treated population. To address the controversy of whether non-hospitalized individuals should sleep or be frequently awakened following TBI, 6 h of sleep disruption following TBI is translationally relevant. Further investigation is needed to determine the effect of sleep disruption on other aspects of posttraumatic symptomology, including somatic and emotional function.

CONCLUSION

The current study demonstrated that disrupting acute post-traumatic sleep following diffuse TBI did not worsen functional outcome. The sleep lost during the disruption period was not recovered by increased sleep time, and there were no long-lasting circadian disturbances detected under 12:12 light:dark conditions. Further studies are needed to fully understand the cellular benefit or detriment, if any, of acute posttraumatic sleep on recovery following TBI, as well as other neurological conditions.

DISCLOSURE STATEMENT

Dr. O'Hara is a principal owner of Signal Solutions (Lexington, KY). The company makes and sells the sleep-wake monitoring system (software and hardware) used in this study. The other authors have indicated no financial conflicts of interest.