Dexmedetomidine-mediated sleep phase modulation ameliorates motor and cognitive performance in a chronic blast-injured mouse model

Abstract

Multiple studies have shown that blast injury is followed by sleep disruption linked to functional sequelae. It is well established that improving sleep ameliorates such functional deficits. However, little is known about longitudinal brain activity changes after blast injury. In addition, the effects of directly modulating the sleep/wake cycle on learning task performance after blast injury remain unclear. We hypothesized that modulation of the sleep phase cycle in our injured mice would improve post-injury task performance. Here, we have demonstrated that excessive sleep electroencephalographic (EEG) patterns are accompanied by prominent motor and cognitive impairment during acute stage after secondary blast injury (SBI) in a mouse model. Over time we observed a transition to more moderate and prolonged sleep/wake cycle disturbances, including changes in theta and alpha power. However, persistent disruptions of the non-rapid eye movement (NREM) spindle amplitude and intra-spindle frequency were associated with lasting motor and cognitive deficits. We, therefore, modulated the sleep phase of injured mice using subcutaneous (SC) dexmedetomidine (Dex), a common, clinically used sedative. Dex acutely improved intra-spindle frequency, theta and alpha power, and motor task execution in chronically injured mice. Moreover, dexmedetomidine ameliorated cognitive deficits a week after injection. Our results suggest that SC Dex might potentially improve impaired motor and cognitive behavior during daily tasks in patients that are chronically impaired by blast-induced injuries.

Keywords: blast injury model; chronic blast injury; cognitive behavior; dexmedetomidine; motor behavior; sleep/wake cycle disturbance.

Figure 1

Timeline of the longitudinal analysis of the sleep/wake cycle and behavior in SBI and control mice. The schematic illustrates the timeline of the repeated tests applied to blast injured and control mice before and several weeks after blast injury. The top line shows the time scale divided into periods. Seven days before period one, animals were implanted with a wireless transmitter (gray box). Then in period 1 (5 days), mice were exposed to a battery of motor and cognitive tests, including novel object recognition (NOR), Y maze(Y-M), and balance beam. After this evaluation, the sleep/wake cycle was monitored for 24 h on three consecutive days (period 2). Next, on day 0 of period 3, mice were subjected to a blast and sham procedure, respectively, while animals were anesthetized. We monitored EEG and EMG activity immediately after SBI and for six consecutive days. Later in period 4, we repeated motor and cognitive testing (blue box). After that, we studied EEG/EMG activity in period 5 for 8 days, and mice executed the rotarod (RR). We repeated the cognitive and motor tasks for the third time and for 5 days (period 6). Subsequently, while we were monitoring EEG/EMG activity (period 7), we injected Dex (0.15 mg/kg) within the first hour of the light cycle, and 16 h later, mice ran on the rotarod (orange box). Lastly, we evaluated cognitive and motor behavior a week after Dex injection (period 8).

Figure 2

Blast injured mice exhibit an excessive sleep pattern during the acute stage post-injury. (A) Representative scatter plot of the delta-alpha scores obtained from SBI and sham mice during day 2 after trauma (period 3, P3) and normalized to the delta/alpha ratio before blast injury (period 2; P2). Data were clustered into two groups using the K-means function: mice with a high (HD) and low (LD) delta/alpha ratio. The cluster centroids (cyan cross) represent the average of delta and alpha power relative scores, and the pink circle depicts 90% of the mean distance between the centroid of the cluster containing control animals and LD. Carmine dots represent SBI mice assigned to the cluster of low delta/alpha ratio (n = 8); brown dots, SBI mice assigned high ratio (n = 8), and blue dots correspond to the control group (n = 11). (B) Delta alpha ratio differences between SBI-HD (n = 7), SBI-LD (n = 8), and sham mice (n = 11) in NREM, REM, and Wake state before and after blast induced-injury (dashed gray line). (C) NREM, REM, and awake episode duration over 24 h in SBI-HD (n = 7), SBI-LD (n = 8) and sham mice (n = 11) and normalized to averaged episode duration before SBI. (D) Average peak domain amplitude (mean ± S.E.M.) in microvolts of the detected spindles in 24-h periods and normalized to the averaged peak found before SBI represented in relative units (rel.un.). (E) Averaged intra-spindle frequency distribution (mean ± S.E.M.) in detected spindles during 24 h in SBI-HD, SBI-LD and control mice days before and after causing SBI. We employed Kruskal Wallis and Dunn's test with Bonferroni correction, *p < 0.05, **p < 0.01, ***p < 0.001. Carmine asterisks represent differences between sham and SBI-LD, brown asterisks depict differences between SBI-HD and sham mice, and the orange asterisk display differences between SBI-HD and SBI-LD groups.

Figure 3

SBI chronically affects memory tasks. (A) Novel object preference by control (n = 22) and SBI mice (n = 30) exploring a novel object 1 (period 4) and 4 weeks (period 6) after SBI. Data were normalized to novel object preference before injury and shown in relative units (rel.un.). We applied the Mann–Whitney U test for a two-group comparison. (B) Exhibits the preference of control (n = 22) and SBI mice (n = 30) for the “new” arm of the Y Maze as the time spent in the “new” arm vs. “familiar” (fam.) arm ratio at a week and 4 weeks respectively. Data were normalized to Y Maze execution before SBI and, therefore, represented in relative units. Since the data was non-parametric, we applied the Mann–Whitney U test for two-group comparison *p < 0.05, **p < 0.01, and non-significant statistical differences (N.S).

Figure 4

SBI chronically alters fine skill motor execution. (A) Illustrates the velocity reached by control (n = 22) and SBI (n = 30) while crossing the wooden beam a week (period 4) and 4 weeks (period 6) after SBI. Data were normalized to velocity reached before SBI. Data are shown in relative units (rel.un.). In addition, the data shows the balance beam performance when the SBI group was split into SBI-HD (n = 7) and SBI-LD (n = 8) and contrasted to control (n = 22). (B) The performance trajectories on a rotarod of control (n = 21), SBI-HD (n = 6), and SBI-LD (n = 7) after 3 weeks post-SBI (period 5). Data is presented in average durations (mean ± S.E.M.) of 5 trials per individual per day. We used the Kruskal–Wallis test and Dunn's with post hoc Bonferroni correction to determine the effect of blast injury on fine motor skills. *p < 0.05, **p < 0.01, ***p < 0.001, and non-significant statistical differences (N.S). Brown asterisks depict differences between SBI-HD and sham mice, and the orange asterisk display differences between SBI-HD and SBI-LD group.

Figure 5

Dex-mediated sleep phase modulation ameliorates motor tasks. (A) Depicts changes in the intra-spindle frequency tracked for 24 h 3 days before and 3 days while injecting Dex injection (0.15 mg/Kg) in control mice injected with vehicle (n = 5), control mice injected with Dex (n = 4), SBI mice injected with vehicle (n = 4), and SBI mice injected with Dex (n = 5). (B) The upper panel shows the averaged (mean ± S.E.M.) run time on rotarod in seconds (s) of control (n = 21) and SBI mice (n = 27). In addition, we showed the run times 3 days before, during, and a day after Dex treatment (lower panel) including control mice injected with Dex (n = 11), control mice injected with vehicle (n = 10), SBI mice injected with Dex (n = 17) and SBI mice injected with vehicle (n = 11). (C) Displays box plots of the delta/alpha ratio scores obtained from control-vehicle (n = 5), control-Dex (n = 4), SBI-vehicle (n = 4), and SBI-Dex (n = 5) during the NREM, REM, and awake states. Each state was divided into 12 h of light and dark cycles. We applied Kruskal–Wallis and Dunn's test with Bonferroni correction to establish the effect of Dex on intra-spindle frequency, rotarod performance, and delta-alpha ratio. *p < 0.05, ***p < 0.001, and non-significant statistical differences (N.S).

Figure 6

Cognitive tasks improve a week after Dex injection. (A) Delta/alpha ratio differences between SBI-Dex (n = 5), SBI-control (n = 4), Control-vehicle (n = 5), and Control-Dex (n = 4) during the NREM detected over 24 h and measured 3 days before SBI and compared to the ratio obtained a day after withdrawing Dex. Data were normalized to averaged ratio before SBI. (B) Alpha power measured 3 days before injury and compared to the alpha power measured a day after withdrawing Dex. Data were normalized to averaged alpha power before SBI. (C) Novel object task conducted at week 7 (period 8) in control (n = 22) and SBI (n = 28) mice, right side of the panel shows differences between SBI-vehicle (n = 12) and SBI-Dex (n = 16) groups observed before Dex injection (period 6) and 7 days post-Dex injection (period 8). Data were normalized to time spent exploring the novel object before injury and are shown in relative units (rel.un.). (D) Exhibits the velocity reached by control (n = 22) and SBI (n = 28) mice while performing on the balance beam at week 7 (period 8). The right side of the panel shows the differences between SBI-vehicle (n = 12) and SBI-Dex groups (n = 16) before Dex injection (period 6) and 7 days after Dex injection (period 8). Data were normalized to velocity reached before injury and shown in relative units (rel.un.). We employed the Mann–Whitney U test for two-group comparison to establish differences between control and SBI groups as well as in SBI groups with and without Dex injection while executing novel object recognition and balance beam tasks. *p < 0.05, and non-significant statistically differences (N.S).