Gamma frequency entrainment rescues cognitive impairment by decreasing postsynaptic transmission after traumatic brain injury

Abstract Introduction The relationship between oscillatory activity in hippocampus and cognitive impairment in traumatic brain injury (TBI) remains unclear. Although TBI decreases gamma oscillations and 40 Hz light flicker improves TBI prognosis, the effects and mechanism of rhythmic flicker on TBI remain unclear. Aims In this study, we aimed to explore whether light flicker could reverse cognitive deficits, and further explore its potential mechanisms in TBI mouse model. Methods The Morris water maze test (MWM), step‐down test (SDT), and novel object recognition test (NOR) were applied to evaluate the cognitive ability. The local field potential (LFP) recording was applied to measure low gamma reduction of CA1 in hippocampus after TBI. And electrophysiological experiments were applied to explore effects of the gamma frequency entrainment on long‐term potentiation (LTP), postsynaptic transmission, and intrinsic excitability of CA1 pyramidal cells (PCs) in TBI mice. Immunofluorescence staining and western blotting were applied to explore the effects of 40 Hz light flicker on the expression of PSD95 in hippocampus of TBI mice. Results We found that 40 Hz light flicker restored low gamma reduction of CA1 in hippocampus after TBI. And 40 Hz, but not random or 80 Hz light flicker, reversed cognitive impairment after TBI in behavioral tests. Moreover, 40 Hz light flicker improved N‐methyl‐D‐aspartate (NMDA) receptor‐dependent LTP (LTPNMDAR) and L‐type voltage‐gated calcium channel‐dependent LTP (LTPL‐VGCC) after TBI treatment. And gamma frequency entrainment decreased excitatory postsynaptic currents (EPSCs) of CA1 PCs in TBI mice. Our results have illustrated that 40 Hz light flicker could decrease intrinsic excitability of PCs after TBI treatment in mice. Furthermore, 40 Hz light flicker decreased the expression of PSD95 in hippocampus of TBI mice. Conclusion These results demonstrated that 40 Hz light flicker rescues cognitive impairment by decreasing postsynaptic transmission in PCs after TBI treatment in mice.


| INTRODUC TI ON
Oscillatory amplitude, frequency, and its related brain rhythms were involved in some diseases including traumatic brain injury (TBI), Alzheimer's disease, and stroke. [1][2][3][4] A level of 8-14-Hz oscillatory activity contributes to a better attenuation in TBI patients during paced auditory serial addition test (PASAT). Although Yang et al., have reported that TBI decreases gamma oscillations and 40 Hz light flicker improves TBI prognosis, the effects and mechanism of rhythmic flicker on TBI remain unclear.
Our previous study has shown that TBI could induce cognitive deficits accompanying the changes in excitability and excitatory transmission in the hippocampus. 5 The hippocampal gamma frequency entrainment has a close relationship with cognitive ability, and abnormal gamma oscillations were found in neurological disorders. 6,7 Furthermore, alterations of gamma oscillations after hippocampal local field potential (LFPs) cross-frequency coupling are related to long-term potentiation (LTP) reduction in rats. 8 Only low gamma, rather than theta or high gamma, has been found in a mouse model of hippocampal damage. 9 However, the role of hippocampal oscillation in the learning and memory after TBI in mice remains unclear. In this study, we aim to explore the effects of light flicker on cognitive ability in TBI mice and explore its potential mechanisms.
Brain oscillations could be manipulated by 40 Hz light flicker in mouse models of Alzheimer's disease and two-vessel occlusion models in mice and then improve cognitive abilities in those mice. 7,10 These results suggested that 40 Hz light flicker could reverse cognitive deficits by manipulating brain oscillations. Thus, we aim to explore the potential mechanisms of 40 Hz light flicker on cognitive ability after TBI. Impairment of hippocampal LTP, reduction of spontaneous synaptic transmission, and intrinsic excitability of CA1 pyramidal cells (PCs) were found in cognitive deficits after TBI. 5 However, the effects of 40 Hz light flicker on synaptic transmission, LTP, and cognitive ability remain unclear. In this study, we aim to explore rhythmic flicker on cognitive ability and its possible mechanism in TBI mice.
Behavioral tests including the Morris water maze (MWM), stepdown test (SDT) and novel object recognition test were applied to detect the cognitive ability, and electrophysiology was used to evaluate the synaptic transmission, intrinsic excitability of PCs and LTP in TBI mice. We found that 40 Hz light flicker can reverse cognitive deficits with decreasing spontaneous postsynaptic transmission, intrinsic excitability of PCs and CA3-CA1 LTPs in TBI mice. Mouse skull was exposed after being anesthetized using 2.5% isoflurane, and drilled in the middle of the brain. The herringbone sulcus was drilled and kept the dura mater intact during performing operation of TBI. The weight drop of 30 g was set at 3 centimeters above the mouse brain. For the control group, mice were applied to the same procedure without the weight drop injury.
The visual stimulation was performed according to previous study with minor modification. 11 All mice were placed in a chamber illuminated by a 40 Hz, 80 Hz, or random flicker which was illuminated by a light-emitting diode (LED) bulb. The six LED lamps were arranged with parallel circuits in order to equally distribute the 40 Hz LED light to the four transparent cages, which housing the four experiment groups at the same time. The 1 h session of visual stimulation was performed 12 h after surgery, and the stimulation was performed twice a day with 12 h intervals that lasted to 25 days post-TBI ( Figure S2A). Moreover, 80 Hz or random flicker visual stimulation was conducted with using the same protocol.

| Behavioral experiments
The MWM was performed on days 15-18 after TBI according to the previous study ( Figure S2A,B). 12 A circular pool with 130 cm diameter, 50 cm height was filled with water. A platform with 8 cm diameter was hidden under the surface of the water during the learning phase for 3 days. In the learning phase, all the mice were allowed to find the hidden platform for three consecutive days, 4 trials per day with a 60-s interval. For each trail, the mice were allowed to start from one of the four quadrants and ended when the mice climbed on the hidden platform by navigation with four visual cues ( Figure S2B).
The mice were guided to the hidden platform when the mice could not climb on the platform up to 60 s. Memory ability was tested 24 h after the learning phase by removing the platform. Each mouse was allowed to swim freely for 2 min and monitored by using a computer video-tracking system (Huaibei Zhenghua Biologic Apparatus Facilities Limited Company).
The SDT was carried out 1 day after MWM ( Figure S2A,C) according to previous study. 13 The shock box was a 60 cm × 10 cm × 10 cm plastic box, and the floor of the box was assembled of parallel 0.1 cm-caliber stainless steel bars with a spacing of 0.5 cm. Four rubber platforms (diameter: 10 cm, height: 4.5 cm) were put in the middle of the shock box. All the mice were placed on the bottom of the shock box and received an electric shock (36 V, AC) on day 19 for 5 min in the training phase. The animals were punished by an electric shock when the mice exposed to stainless steel bars. On day 20, the mice were placed on the rubber platform, and the latency that the time for mice to step down from the rubber platform for the first time was measured, which was applied to measure the memory retention for mice.
The novel object recognition test (NOR) test was performed to detect the memory ability on days 21-24 after TBI ( Figure S2A,D).
All the mice in this test were subjected to three phases: habituation phase, familiarization phase, and discrimination phase. The habituation phase was performed before the TBI surgery. The mouse was allowed to move for 5 min, and its exploration time on each object was recorded. The distance between the mouse's nose and the object being identified should not exceed 2 cm or the contact with the nose is considered inquiring behavior. In the familiarization phase, the presence of walking around the identified object is not considered to be an inquiry behavior. The discrimination phase   Figure S1C). Then the microwire array electrode was fixed to the brain skull bone by using skull screws and reinforced by using dental cement.

| Intracranial electrode implantation surgeryand LFP recording
Mice were adapted in a 60 cm × 10 cm × 10 cm plastic box for 10 min per day from day 21 to day 25 after TBI until the first LFP recording. The plastic box was cleaned by using 75% ethanol before LFP recording in each mouse. The microwire array electrode was connected to a helium balloon and the mice were allowed to move freely in the plastic box during LFP recording. In order to reduce the impact of mouse movement on the values of theta waves, 14 the states of approach, exploration, and other low-speed walking were applied in this study for LFP recording. The microwire array electrode was connected to the OmniPlex Neural Recording Data Acquisition System (Plexon Inc.). Meanwhile, a camera above the plastic box was applied to record the mouse movement during the LFP recording. Moreover, an LED bulb was set in front of the plastic box to perform visual stimulation. The LFPs in the CA1 region were measured for at least 20 min for each mouse. Each mouse received the same pattern of 40 Hz visual stimulation with 2 h break between two mice. A total of six mice for each group were applied for statistical analysis according to the previous study. 7 Briefly, the sampling rate of LFP signals were set at 1000 Hz with a bandpass filter set at 05-300 Hz. The NeuroExplorer software (Version 5, Nex Technologies) was used for power spectral analyses. The ratio of power spectral density (PSD) between PSD exploration and PSD approach in theta, low gamma, 40 Hz region, and high gamma were measured.

| Electrophysiology
All the mice were sacrificed on day 25 after TBI ( Figure

| Immunofluorescence staining
Mice were decapitated, then brain was harvested on day 25 after TBI ( Figure S2A). Immunofluorescence staining was conducted as previously described. 15 In brief, the brain tissues were fixed in 4% paraformaldehyde at 4°C for 48 h and dehydrated in ethanol and embedded in paraffin. The coronal mouse brain sections were ob- were taken using a laser scanning confocal microscope (Leica).

| Western blotting
Mice were decapitated, then hippocampus was harvested on day 25 after TBI ( Figure S2A). The tissues were homogenized at 4°C by

| Statistical analysis
All data were analyzed using GraphPad Prism (Version 7.0) or SPSS (Version 23.0 for windows). The Shapiro-Wilk test for normality was applied to assess data distribution. All the data were presented as the mean ± standard error (SEM). The statistical analysis of escape latency in the learning phase of MWM and DI in the NOR test, etc., by using a 2ANOVA with repeated measures over time followed by Tukey test as shown in the "Results" section. Two-tailed Student's unpaired t-test was applied to identify significant groups as shown in figure legends.
p < 0.05 was considered to indicate statistical significance.

| 40 Hz light flicker attenuated neurological deficits of TBI mice
To evaluate the learning and memory abilities of TBI mice, the MWM, SDT, and novel object recognition test (NOR) were performed as depicted in Figure S2A-D. The MWM showed 40 Hz light flicker significantly improved the spatial learning (2ANOVA, F(3, 132) = 120.7, p < 0.0001, Tukey test p TBI vs. TBI + 40 HZ = 0.0064 for day 3, Figure 1A) and memory ability in TBI mice ( Figure 1B-E).

| 40 Hz light flicker restored low gamma reduction of CA1 in hippocampus after TBI
Previous study has shown that 40 Hz light flicker entrained with hippocampal CA1 low gamma. 7 Thus, we aim to explore the effects of 40 Hz light flicker on low gamma of CA1 after TBI. Our results have shown that LFP power of CA1 in TBI mice (Figure 2A  In the NOR, the DI was obtained to measure working memory of mice in all four groups. n = 12 mice for each group. Data were mean ± SEM. Error bars indicated SEM. *p < 0.05, **p < 0.01, and ***p < 0.001 were compared with the control group. # p < 0.05, ## p < 0.01, and ### p < 0.001 were compared with the TBI group. 2ANOVA with Tukey test was applied in this section. DI, discrimination index; MWM, Morris water maze test; NOR, novel object recognition test; SDT, step-down test; TBI, traumatic brain injury.

| Effects of 40 Hz light flicker on intrinsic excitability of PCs in TBI mice
To for Figure 5G). These data may illustrate that 40 Hz light flicker decrease intrinsic excitability of PCs in TBI mice.

| Effects of 40 Hz light flicker on the expression of PSD95 in hippocampus of TBI mice
To explore the role of 40 Hz light flicker on the expression of PSD95 in hippocampus of TBI mice, immunofluorescence staining and western blotting were applied. Our results revealed that TBI . Data were mean ± SEM. Error bars indicated SEM. ***p < 0.001 and ****p < 0.0001 were compared with the control group, # p < 0.05 and #### p < 0.0001 compared with the TBI group. 2ANOVA with Tukey test was applied in this section. PCs, pyramidal cells; sEPSCs, spontaneous excitatory postsynaptic currents; TBI, traumatic brain injury. Data were mean ± SEM. Error bars indicated SEM. ***p < 0.001 was compared with the control group, ## p < 0.01 and ### p < 0.001 were compared with the TBI group. 2ANOVA with Tukey test was applied in this section. RMP, resting membrane potential; PCs, pyramidal cells; TBI, traumatic brain injury.  . Immunofluorescence staining showed the levels of NeuN and PSD95 in CA1 of hippocampus in all four groups (C-E) (n = 9 sections from three mice for each group). Scale = 100 μm. Data were mean ± SEM. Error bars indicated SEM. ****p < 0.0001 was compared with the control group, # p < 0.05, ### p < 0.001 and #### p < 0.0001 were compared with the TBI group. 2ANOVA with Tukey test was applied in this section. TBI, traumatic brain injury. Previous study has found that increasing sEPSC amplitude without changes in sEPSC frequency has been related with postsynaptic mechanisms. 4 Moreover, gamma frequency entrainment increased rheobase and had no effects on resting membrane potential (RMP)

Control
of PCs in TBI mice. Moreover, 40 Hz light flicker decreased the expression of PSD95 in hippocampus of TBI mice. Thus, we suspected that postsynaptic mechanisms may be response for cognitive enhancement after gamma frequency entrainment in TBI mice.
This study only made limited progress of gamma frequency entrainment on treatment of cognitive impairment after TBI in mice.
Future studies of synaptic mechanisms for plasticity in hippocampus after light flicker are desired to be applied. Although one limitation of this study is that all the mice were applied to electrophysiological experiments after the MWM, SDT, and NOR, principle of the three R's (replacement, reduction, and refinement) was applied in this study.

| CON CLUS ION
40 Hz light flicker rescues cognitive impairment by enhancing postsynaptic transmission in CA1 PCs after TBI treatment in mice, which may become a promising therapeutic strategy in clinic.

CO N FLI C T O F I NTER E S T S TATEM ENT
All authors declare no conflict of interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.