Sleep deprivation (SD) is widely recognized as a contributing factor to various brain disorders, such as memory dysfunction and psychosis [1-3]. The influence of SD on cognition is significant, as adequate sleep time enables self-repair of nerve injury, removal of toxic metabolites produced by the body, and a reduction in cellular stress [4]. Both animal and human studies have demonstrated that sleep deprivation leads to memory impairment [5-7].

Cognitive impairments, including working memory and behavioral flexibility, have traditionally been linked to prefrontal cortex (PFC) dysfunction. Memory deficits due to sleep loss are observed in tasks that depend on the hippocampus, but also critically involve the medial PFC (mPFC) for consolidation. Sleep loss alters activation patterns in the PFC and its connections to the parietal cortex [8, 9]. Recent research has identified a PFC area in mice, the anterior cingulate cortex (ACC), which is functionally and anatomically positioned to control visually guided behaviors. Causal manipulation studies using chemical genetics or optogenetics have demonstrated the importance of ACC activity in visually guided tasks [10-12]. However, the exact role of ACC in cognitive impairment following SD remains to be elucidated.

Previous research has shown that SD promotes the accumulation of reactive oxygen species (ROS) [13], which can impair neural function and development leading to detrimental effects on cognitive functions, such as attention and working memory [14]. Neurons in the PFC are particularly susceptible to oxidative stress [15]. SD exacerbates oxidative stress in the PFC and reduces gamma activity, a prominent feature of the awake brain [16]. Therefore, it is hypothesized that oxidative stress induced by SD affects ACC neuronal activity, resulting in impaired cognitive memory.

Plant-derived natural products have been identified as effective molecules with neuroprotective activities for treating neurological diseases. Gallic Acid (GA, 3,4,5-trihydroxy benzoic acid), a low molecular weight phenolic compound, possesses antioxidant, anti-inflammatory, and neuroprotective properties [17, 18]. GA is readily absorbed and can cross the blood-brain barrier [19]. Furthermore, this molecule can be used to treat various neurological diseases that cause neuronal damage. GA has also been shown to prevent oxidative and neuronal damage by reducing glial activation, neuroinflammation, and protein aggregation [20]. However, it remains unknown whether the antioxidant GA can alleviate memory deficits caused by SD in mice.

In this study, continuous 72-hour sleep deprivation was combined with two-photon calcium imaging to observe cortical activity in the ACC layer II/III cortex in awake mice. Sleep deprivation was found to decrease the frequency and amplitude of Ca²⁺ transients and increase the proportions of inactive neurons. Furthermore, whole-cell patch-clamp recordings were employed in acute brain cortex slices to record glutamate receptor-mediated sEPSCs and gamma amino butyric acid A (GABAA) receptor-mediated spontaneous inhibitory postsynaptic currents (sIPSCs). These SD-induced neuronal and synaptic alterations suggest that ACC neurons may exhibit lower activity levels following sleep deprivation. In addition, the level of several oxidative stress parameters was assessed, and it was found that GA administration dramatically limited cortical inactivity and reversed the level of oxidative stress parameters. These findings suggest that GA exerts neuroprotective effects in sleep deprivation through the alleviation of oxidative stress.

Animals

Male C57BL/6J mice (8 weeks old, 20~25 g) were obtained from the Institute of Zoology, Chinese Academy of Sciences. All mice had free access to food and water and were maintained in a temperature (22°C±2°C) and humidity (40%~60%) controlled room with a reverse light: dark cycle (12:12). Gallic acid powder (Sigma7384) was dissolved in saline (0.9%). Mice in the GA group received the same concentration (400 mg/kg) through gavage once a day at 5:30 p.m. Mice in the Control and SD groups were administered saline. Control: Saline. SD: SD+Saline. GA: SD+GA. All experiments were approved by the Animal Care and Use Committee of Shandong University of Traditional Chinese Medicine, in compliance with National Institutes of Health guidelines. Behavioral Assays.

Sleep deprivation procedure

The sleep deprivation procedure was performed based on a previously published method, with slight modifications [21, 22]. Mice were transferred to cylindrical sleep recording cages (25 cm in diameter and 40 cm tall, XINRUAN Technologies, China). During the sleep disruption period, the bar’s rotation speed was set at 3 rotations per minute, with reversals of rotation direction (i.e., clockwise vs. counterclockwise) occurring at semi-random intervals of 10±10 seconds. The bar was programmed to rotate for 72 hours. Control animals were placed in identical cages with bars that remained stationary throughout the experiment. Cages were equipped with corncob bedding, and food/water was available ad libitum throughout the protocol.

Y-maze test

The Y-maze test was conducted according to a previously published protocol, with slight modifications [23]. Briefly, the Y-maze apparatus (XINRUAN Technologies, China) consists of three arms with identical dimensions (120°, 30 cm long×5 cm wide×15 cm high). The arms of the maze were labelled start, novel, and other. The camera was placed on the retort stand above the Y-maze and connected to a computer via an extended USB cable. The maze was placed inside a room with dim illumination (15~20 lux). The floor of the maze consists of sawdust to eliminate olfactory stimuli. Testing was always at the same time and performed in the same room to ensure environmental coincide. Place mice in behavioral testing room for 1 h before the test so they can acclimatize to the conditions.

After acclimatization of the mice, close off one of the arms labelled novel with the divider. Place the test animal into the start arm, facing towards the center of the maze, holding the mice gently by the base of the tail. Let the mouse explore the maze undisturbed for 10 min. After the recording, return the test mouse to its home cage.

Clean all walls and the floor of the Y-maze before proceeding to the next animal. Wait 1 h and Remove divider. Place the same test mouse into the start arm, facing towards the center of the maze. Leave the animal to explore the maze undisturbed for 5 min. A mouse entry into an arm was considered valid if its body and tail completely entered the arm. The total number of arm entries and sequences were recorded using video and analyzed later on a computer. Remove mouse from the maze and place back into home cage. Spatial reference memory was assessed by identifying the start, novel, and other arms of the maze, recording the number of entries into each arm as well as the time spent in each arm [24].

Stereotaxic virus injection

Mice were anesthetized with an isoflurane-oxygen mixture (1.5% vol isoflurane/vol O₂) and administered the analgesic buprenorphine (SC, 0.3 mg/kg). Body temperature was maintained at 37±0.5°C with an electric heated pad during surgery. A 1 cm incision was made on scalp with an eye scissor. For calcium imaging with GCaMP6f, 150 nl of AAV2/1-syn-GCaMP6f-WPRE-hGH-containing solution (~2×10¹³ infectious units/ml) was slowly injected (50 nl/min) into the ACC (anterior/posterior: -0.27 mm, medial/lateral: +0.26 mm, and dorsal/ventral: -0.50 mm) using a glass microsyringe. The injection site refers to the previous research and has been slightly modified [25]. To prevent backflow during withdrawal, the pipette remained in the brain for over 15 minutes after completing the injection.

Implantation of custom-made head clamps

A round craniotomy (~4 mm diameter) in ACC of living mice was performed carefully and intermittently over the ACC. To improve imaging results, the skull and dura mater were carefully removed. A round glass coverslip (5 mm in diameter) was adhered over the opening with biological glue. Following surgery, animals were returned to their home cages and treated with carprofen (5 ug/g, bodyweight, i.p.) for a week.

Two-photon imaging in vivo

All imaging was performed using a two-photon microscope and a Ti: Sapphire laser (model “Mai-Tai Deep See” Spectra Physics). ZEISS water-immersion objectives (10X, 40X) were employed. For calcium imaging experiments, the excitation wavelength was set to 910 nm. The average power reaching the cortical surface ranged from 30 to 40 mW, depending on the expression efficiency of the virus and the depth of imaging. For somatic imaging, the field-of-view (FOV) dimensions were set to 200 µm×200 µm. Images of 512×512 pixels were acquired at a 4 Hz frame rate. Within an imaging time window of ~3 minutes (~1 minute per time, three times in total) for each imaging session, no signs of photodamage were observed. Imaging was performed at 6:30 p.m. each day. Prior to the experiment, mice were frequently fixed under the microscope objective with the chamber to acclimate to the imaging state for real calcium activity. When we observed neurons with a two-photon microscope for the first time, we first determined the approximate location of the ACC with a low magnification of the microscope (10X), and then decided the region of interest (ROI) based on the imaging quality and quantity of neurons in the field of vision with a high magnification of the microscope (40X). After the ROI was determined, the same ROI was observed each time by comparing the image characteristics of the blood vessels in the images as following steps. First, after the ROI was determined, the depth was determined (e.g. 350 um) and imaging was performed to determine the morphological characteristics of neurons or microvascular with a high magnification of the microscope (40X). Second, keeping X-axis and Y-axis unchanged, we changed the Z-axis to reduce the depth (eg. 50 um) and recorded the above features by imaging again. Third, we imaged ACC blood vessels with the low magnification of the microscope (10X) to record relative positions and characteristics of the vascular. On the next time of observation, after fixing the animal, we followed step third, second and first to find markers and the final ROI. In that way, the same ROI and the same neurons can be observed many times and multiple days.

Electrophysiology

The mice that were under deep anesthetization by breathed isoflurane were decapitated. The brains were rapidly dissected in precooled, oxygenated standard artificial cerebrospinal fluid (ACSF) (120 mM NaCl, 2.5 mM KCl, 2.4, 1.25 mM NaH₂PO₄, 26 mM NaHCO₃, 2 mM MgSO₄, 2 mM CaCl₂, 10 mM d-glucose; pH: 7.4). Both hemispheres were sliced coronally (350 µm) in cold ACSF. The slices were quickly transferred to 32°C ACSF for 30 min and then incubated at room temperature for at least 1 h. The slice was fixed in a recording chamber that was set on the fixed-stage of an upright Olympus BX50WI microscope (Olympus). The oxygenated ACSF was perfused through recording chamber continuously. Neurons in layer 2/3 were optically identified and whole-cell patched for recording.

Recording sEPSCs: brain slices were held between nylon netting in an interface chamber and fully submerged in flowing ACSF (3 ml/min) at room temperature (20~25°C). Cells in the ACC were visualized using a DIC-infrared upright microscope and recorded using whole-cell patch-clamp procedures at -70 mV. The pipette resistance ranged from 4 to 6 MΩ. The intracellular solution contained (in mM): 140 K-gluconate, 2 MgCl₂, 10 HEPES, 10 BAPTA, 2 Mg-ATP, 0.5 CaCl₂-H₂O, 0.5 Li-GTP, pH 7.2~7.4, 280~290 mOsm. 5 mM QX-314 was added to block voltage-gated Na⁺ channels and GABAB receptors. The sEPSCs were recorded in the presence of 100 μM picrotoxin (PTX).

Recording sIPSCs: Pipette electrodes were filled with an internal solution containing (in mM): 140 CsCl, 10 HEPES, 2 MgCl₂, 0.5 EGTA, 2 Mg-ATP, 0.5 Na₃GTP, 12 phosphocreatine, and 30 NMG, pH 7.2~7.4, 280~290 mOsm. To block voltage-gated Na⁺ channels and GABA_BRs, 5 mM QX-314 was added. After voltage-clamping the neuron at -70 mV, sIPSCs were recorded in the presence of 1 mM kynurenic acid.

Recordings were conducted using a 2 kHz 4-pole Bessel filter at a 10 kHz sampling frequency with an Axopatch 700B amplifier and pClamp 10.3 software. Electrode capacitance and series resistance were monitored and compensated; access resistance was monitored throughout the experiment, and cells were discarded if access resistance increased more than 10% during the experiment.

Determination of antioxidant enzyme activities

Mice were anesthetized and euthanized, after which the cerebral cortex was quickly removed, homogenized, and centrifuged (12,000×g, 10 minutes, 4°C). Supernatants were collected for subsequent experiments. SOD activity and MDA level were measured using commercial assay kits (A003, A001, Nanjing Jiancheng Bioengineering Institute, Nanjing, China), with final results expressed as U/mg protein for SOD activity.

Nrf2 ELISA

Cortical tissues were rapidly dissected from the brain, flash-frozen in liquid nitrogen, and homogenized. The lysates were centrifuged at 1,000×g for 30 minutes. Supernatants were used to assess Nrf2 concentration. Nrf2 expression were quantified using ELISA kits (CSB-E16188m, CUSABIO, China) following the manufacturer’s protocol.

Data analysis

Calcium imaging data preprocessing and cell segmentation were performed by using established methods [26-28]. Calcium imaging data were calibrated for motion artifacts and analyzed using custom-written software in MATLAB (version 2014a; MathWorks; MA). Movies focal plane displacements were corrected using Mosaic Package in MATLAB. After the ROI is determined, the same ROI is observed each time by comparing the image characteristics of the blood vessels in the images. To quantify calcium signal, somatic regions of interest (ROIs) fluorescence intensity were determined with a semi-automated algorithm that correlated the fluorescence intensity between adjacent pixels. The calcium signals of each ROI were averaged, and ROIs were confirmed again by visual inspection. Ca²⁺ signals were represented by relative fluorescence changes calculated as Δf/f=(f−f0)/f0, where f was estimated by averaging fluorescence of all pixels within each specified ROI, and f0 represented the baseline fluorescence of the ROI estimated as the 25th percentile of the fluorescence within a sliding time window. Cortical functional connectivity is defined as a strong temporal correspondence of events between two neurons [28]. Drastic motion imaging data were excluded from analyses. Electrophysiology data were filtered at 2 kHz and sampled at 10 kHz using an Axopatch 200B amplifier and pClamp 10.3 software. The amplitude and frequency of sEPSCs and sIPSCs were analyzed using MiniAnalysis (Synaptosoft, Inc). When possible, the Kolmogorov-Smirnov test was used. Otherwise, data points were assumed to have a normal distribution and were analyzed using a t-test or ANOVA followed by a post-hoc Tukey’s test or Bonferroni's test. All data were two-sided and expressed as means±SEM. Statistical significance was considered at p<0.05.

Gallic acid ameliorates spatial learning and memory deficits in sleep-deprived mice

Sleep deprivation (SD) has been considered a contributing factor in various brain disorders, such as depression, memory dysfunction, and psychosis [29]. To evaluate the behavioral impact of SD regulation in mice, Y-maze behaviors were examined in mice subjected to a continuous 3-day sleep deprivation procedure and Gallic Acid (GA) exposure (400 mg/kg, once a day for 12 days; Fig. 1A). The 3-day sleep deprivation procedure resulted in a significant decrease in distance and time spent in the novel arm, which was reversed by GA exposure (Fig. 1B, 1C). No changes were observed in alternation ratio (Fig. 1D) and mean velocity (Fig. 1E) in both SD and GA-exposed mice. These findings suggest that GA treatment mitigates SD-induced spatial learning and memory deficits.

GA mitigated the decline in Ca²⁺ transients frequency and proportion of inactive neurons induced by SD

Previous research has demonstrated the significance of ACC activity in visually guided tasks [10-12]. To examine the changes in ACC cortical activity, the ultra-sensitive protein calcium sensor GCaMP6f was injected into the ACC (Fig. 2A~2C). A chronic cranial window on the cortex was established to observe neuronal activity in layer II/III of the ACC in awake mice in vivo (Fig. 2D). Approximately three weeks later, two-photon calcium imaging was used to characterize the activity levels of neuron populations (350~450 µm) (Fig. 2E, 2F) with stably expressed GCaMP6f during the experiment.

To investigate the effects of SD and GA on neuron activity in the ACC, continuous sleep deprivation for 72 hours was performed in SD mice, and GA (400 mg/kg) was administered to GA mice for 12 days, starting seven days before SD. Representative calcium images of the same region of neurons at five time points during different periods of the experiment are shown (Fig. 3A; baseline, SD days 1~3, and recovery day 1). Neuron dysfunction was then observed to worsen. The average frequency of Ca²⁺ transients significantly decreased at 24, 48, and 72 hours of SD. A marked decrease was also observed on the first recovery day, but not the second day (Fig. 3B, 3D). Following GA exposure, the average frequency of Ca²⁺ transients significantly recovered, particularly at SD 72 hours (Fig. 3B, 3E, 3F). Based on the calcium transients, neurons were further classified into two categories: inactive cells (<3 Ca²⁺ transients/min) and normal cells (3~6 transients/min) to scrutinize the details of cortical activity in the ACC. The distribution of inactive cells in SD mice revealed a higher number of inactive neurons during SD. The proportion of inactive neurons increased on the first and second days of SD, but no significant differences were observed compared to the baseline (SD day 1: 33.3%, SD day 2: 38.1%, baseline: 18.5%) (Fig. 3G). However, continuous sleep deprivation for 72 hours led to a significant increase in the proportion of inactive neurons compared to the baseline (52.4% versus 18.5%). The changes in the proportion of inactive neurons were reversed by GA exposure (Fig. 3C, 3G). Collectively, these results demonstrate that GA effectively restricts the increase in the number of cortical inactive neurons and the decline in Ca²⁺ transients frequency induced by SD.

GA ameliorates the reduction in Ca²⁺ transients amplitude in ACC neurons

To characterize the changes in intracellular calcium concentration during SD and GA treatment, the mean amplitude of Ca²⁺ transients was analyzed. It was discovered that the mean amplitude of Ca²⁺ transients was significantly reduced at SD 24 hours (Fig. 4A, 4B). The mean amplitudes at SD 48 hours and SD 72 hours decreased, but no significant differences were observed compared to the baseline amplitude of Ca²⁺ transients (Fig. 4A, 4C, 4D). Additionally, a reflective enhancement but no difference in amplitude was observed after SD (Fig. 4N). However, GA treatment completely prevented the decrease in amplitude during SD and the enhancement of amplitude following SD (Fig. 4A, 4O). The amplitude in GA mice at SD 24 hours and SD 48 hours exhibited a significant increase compared to SD mice (Fig. 4F, 4G). No differences were observed at other time points between SD and GA mice (Fig. 4E, 4H, 4I, 4O). Subsequently, the rise time of Ca²⁺ transients was analyzed. No changes were detected during either SD or recovery time (Fig. 4J, 4L, 4O). Furthermore, both SD and GA exposure did not affect the fall time of Ca²⁺ transients (Fig. 4K, 4M, 4O). Collectively, these results suggest that GA exposure mitigates the decrease of Ca²⁺ transients amplitude induced by SD without altering the properties of Ca²⁺ transients.

Continuous sleep deprivation for 72 hours and GA exposure do not affect the synchronization and functional connectivity of cortical network

Synchrony refers to the temporal consistency of distributed neuronal activity and serves as an indicator of information transfer capacity among neurons. High synchrony implies that more calcium events occur simultaneously. Functional connectivity is defined as a key measure of temporal correspondence of calcium transients between individual neurons. It represents the temporal causality of calcium events among neurons, which signifies functional communication. These two parameters were used to estimate temporal network activity patterns by calculating synchrony and correlation matrices quantifying network synchronization (Fig. 5A) and functional connectivity (Fig. 5B) among neurons in imaged regions, respectively. No significant differences were detected in network synchronization (Fig. 5C, 5E), suggesting no gross change in the ability to transfer information among neurons during both SD and GA exposure. Subsequently, mean cell-cell correlation was assessed, and no changes were found in functional connectivity during SD and GA exposure (Fig. 5D, 5F). These results indicate that continuous sleep deprivation for 72 hours and GA exposure do not affect the synchronization and functional connectivity of the cortical network.

GA enhances frequency of glutamate receptor-mediated sEPSCs in SD

To investigate the functional consequences of changes in Ca²⁺ transients in neurons, whole-cell patch-clamp recordings were conducted on ACC neurons. Glutamate receptor-mediated spontaneous excitatory postsynaptic currents (sEPSCs) were measured (Fig. 6A), revealing that continuous sleep deprivation for 72 h and GA treatment did not affect the amplitude of sEPSCs (Fig. 6B). However, SD mice displayed a significant decrease in the frequency of sEPSCs, which was rescued by GA exposure (Fig. 6C). Neither the amplitude nor the frequency of GABAR-mediated spontaneous inhibitory postsynaptic currents (sIPSCs) were influenced by SD 72 h and GA exposure (Fig. 6D~6F). Data demonstrated a significant reduction in the frequency of sEPSCs following SD, but these changes were reversed to control levels after GA administration.

GA treatment may alleviate oxidative stress levels induced by SD

Several studies have indicated that SD triggers the accumulation of reactive oxygen species [13], which can impair neural function and development and have detrimental effects on cognition, including declines in attention and working memory [14], associated with alterations in the PFC [30]. The level of MDA, the activity of SOD, and expression of Nrf2 in different groups was subsequently examined. MDA levels were found to be significantly increased in SD mice compared to control mice, while GA administration notably decreased MDA levels (Fig. 6G). Additionally, it was discovered that SD weakened the activity of SOD and the expression Nrf2, while GA treatment counteracted the effects of SD. These results suggest that GA treatment can mitigate oxidative stress levels caused by SD.

The present study highlights the exacerbation of neuronal dysfunction in the ACC following continuous SD for 72 h, encompassing the average frequency and amplitude of Ca²⁺ transients and the proportion of inactive neurons (<3 Ca²⁺ transients/min). Furthermore, whole-cell patch-clamp recordings revealed a significant decrease in sEPSCs frequency, while the amplitude remained unchanged. These findings align with a report demonstrating a similar decrease in the excitability of PFC pyramidal cells following SD [31]. Additionally, the impact of SD on ACC neurons in layer II/III resembles the observations in subcortical neurons, such as rat hippocampal CA1 neurons, where SD induces considerable downregulation of excitability after 72 h of SD or 24 h of sleep fragmentation [23-34]. However, an increase in intrinsic excitability was observed in dorsal hippocampal CA1 neurons due to 24 h SD, alongside an increase in apical dendritic spines [35]. In layer V/VI pyramidal cells of the medial prefrontal cortex, intrinsic membrane excitability increased after 8 h of SD [36]. Despite differences in SD duration and methodology, these observations contrast with the current study’s findings of decreased cortical excitability, indicating that various brain regions respond differently to sleep loss.

The significant decrease in the frequency of glutamate receptors observed suggests that the synaptic efficacy of ACC neurons declines following SD. This contrasts with a report where an increase in both the amplitude and frequency of mEPSCs was observed in neurons within layer II/III of the frontal association cortex from rats and mice after 4 h SD by gentle handling [37]. Miniature excitatory postsynaptic currents amplitude was slightly reduced, and miniature inhibitory postsynaptic currents amplitude remained unaffected in layer V/VI pyramidal cells of the medial prefrontal cortex [36]. These seemingly inconsistent results may reflect that SD differentially modulates distinct brain regions, rather than merely exerting global effects on the brain [38]. Additionally, it was shown that the concentration of extracellular glutamate increased during the first 3 h of sleep deprivation in rats [39], but decreased rapidly when sleep deprivation continued and high sleep pressure developed. In a 6 h SD, glutamate level decreased by approximately 40% compared to the initial value [39]. Moreover, sleep loss altered the relative expression levels of the AMPA and NMDA type receptors [40]. As the glutamate receptor subunits play a crucial role in regulating Ca²⁺ entry into cells, they may be essential in determining excitability and synaptic plasticity [1, 41].

In addition to SD may have different effects on different brain regions, the duration of SD has an important impact on cognitive function, the function and morphological structure of neurons, and synaptic plasticity [42]. Our previous research found that the effects SD with different periods (24 h, 48 h, 72 h) on the cognition of mice and the morphological structure and synaptic plasticity of hippocampal neurons were significantly different. 24 h SD improved the object location memory (OLM) of mice, accompanied by a significantly increased hippocampal synaptic plasticity, including gain of dendritic spines, increased expression of hippocampal GluA1, and enhanced long-term potentiation (LTP). 48 h SD showed no effect on OLM or the hippocampal synaptic plasticity mentioned above. However, 72 h SD impaired novelty-related OLM and weakened hippocampal synaptic plasticity, including serious loss of dendritic spines, decreased expression of hippocampal GluA1, and significantly attenuated LTP [43]. In this study, we found that the spatial learning and memory deficits of mice were decreased, the frequency and amplitude of Ca²⁺ transients were reduced while the proportions of inactive neurons in layer 2/3 of ACC was increased, and the frequency of sEPSC in layer 2/3 of ACC was significantly decreased with 72 h SD. These results are similar to those in our previous study for 72 h SD. But different from our previous study, the two-photon imaging results of this study showed that the neural excitability in 2/3 layer of ACC was not increased but decreased with 24 h SD, which may be related to different brain areas, the hippocampus in our previous study and ACC in this study. Our results also found that the administration of GA notably mitigated the decline of calcium transients in ACC neurons. The effects of GA may be related to its antioxidation.

A balance exists between the antioxidant capacity and reactive oxygen species (ROS) within the body. Disruption of this balance results in oxidative stress, leading to irreversible damage to biological macromolecules such as DNA, lipids, and proteins [5]. Both preclinical and clinical studies have established that memory impairments are associated with an oxidant-antioxidant imbalance in numerous situations [43, 44]. Binding of glutamate and postsynaptic depolarization results in NMDAR activation and an influx of Ca²⁺ to induce LTP. The increase in intracellular Ca²⁺ initiates a modest and reversible increase in ROS, and in turn, ROS inhibits NMDAR activity. As oxidative stress increases, ROS has effects on NMDARs that result in a further dysregulation of Ca²⁺ homeostasis. In that way, NMDARs remain hyporesponsive and reduce the influx of Ca²⁺ and activity of CaMKII, contributing to impaired LTP induction and neural excitability [45]. Oxidative damage does not influence AMPA receptor-mediated synaptic transmission. Rather, the decrease in AMPA receptor-mediated synaptic transmission during oxidative stress is ancillary to NMDA receptor hypofunction and impaired synaptic transmission and plasticity [45, 46]. There is no consensus concerning the influence of oxidative stress on GABA release, which may depend on the brain region examined [45, 47].

In line with previous reports [48], our study showed that SD significantly elevated oxidative stress levels by reducing the expression of Nrf2 and antioxidant enzyme SOD activities and increasing MDA content in the PFC. Concurrently, GA administration notably counteracted oxidative stress damage with increased neural excitability and excitatory synaptic transmission from SD, potentially explaining the beneficial effects on cognitive performance in SD-treated mice. Earlier reports demonstrated that Ellagic acid, a dimer of gallic acid [49], improved recognition memory by increasing SOD activity while reducing ROS levels in the hippocampus in SD-induced mice through the Nrf2 pathway [50].

Additionally, Nrf2, which is widely expressed in eukaryotic cells, has been considered a critical activator of its downstream antioxidant enzymes, including SOD, and is regarded as one of the most important cellular mechanisms for protection against oxidative stress [51, 52]. The current study revealed that the expressions of Nrf2 and its downstream targets in the PFC of mice were significantly inhibited after SD, while GA treatment reversed this phenomenon [53]. As such, the detrimental effects of increased oxidative stress on memory functions during SD may be mediated by the suppression of signaling molecules essential for memory processing, such as Nrf2. GA, with its antioxidant properties, could protect cognitive performance by preventing this sequence of events during SD.

In conclusion, this study confirmed that the protective effect of GA on 72 h sleep-deprived mice primarily involves aspects of neuronal activity and excitatory synaptic transmission function. It was demonstrated that GA can alleviate neuronal hyperactivity and abnormal calcium influx by regulating excessive glutamate synaptic transmission induced by SD, potentially through mitigating oxidative stress levels in the central nervous system (CNS). These findings suggest that, GA, a small molecule polyphenol widely found in plants [54], may serve as a potential candidate agent for preventing and treating cognitive impairments induced by sleep deprivation.

We thank Dr. Zhijuan Zheng, Dr. Chuanguo Liu and Xuehuan Liu from Experiment Center, Shandong University of Traditional Chinese Medicine for their technical support.

The authors declare that they have no competing interests.