Activation of brain-derived neurotrophic factor signaling in the basal forebrain reverses acute sleep deprivation-induced fear memory impairments.

Abstract Introduction The mechanisms underlying sleep deprivation‐induced memory impairments and relevant compensatory signaling pathways remain elusive. We tested the hypothesis that increased brain‐derived neurotrophic factor (BDNF) expression in the basal forebrain following acute sleep deprivation was a compensatory mechanism to maintain fear memory performance. Methods Adult male Wistar rats were deprived of 6‐hr total sleep from the beginning of the light cycle. The effects of sleep deprivation on BDNF protein expression and activation of downstream tropomyosin receptor kinase B (TrkB)/phospholipase C‐γ1 (PLCγ1) signaling in the basal forebrain and fear memory consolidation were examined. BDNF or selective downstream TrkB receptor antagonist ANA‐12 was further injected into the basal forebrain bilaterally to observe the changes in fear memory consolidation in response to modulation of the BDNF/TrkB signaling. Results Six hours of sleep deprivation‐induced both short‐ and long‐term fear memory impairments. Increased BDNF protein expression and TrkB and PLCγ1 phosphorylation in the basal forebrain were observed after sleep deprivation. Microinjection of BDNF into the basal forebrain partly reversed fear memory deficits caused by sleep deprivation, which were accompanied by increased BDNF protein levels and TrkB/PLCγ1 activation. After ANA‐12 microinjection, sleep deprivation‐induced activation of the BDNF/TrkB pathway was inhibited and impairments of fear memory consolidation were further aggravated. Conclusions Acute sleep deprivation induces compensatory increase of BDNF expression in the basal forebrain. Microinjection of BDNF into the basal forebrain mitigates the fear memory impairments caused by sleep deprivation by activating TrkB/PLCγ1 signaling.

Within the basal forebrain, a region implicated in sleep-wake control, there are three major types of neurons: cholinergic, glutamatergic, and GABAergic neurons. Research has found that cholinergic neurons are active during wakefulness and rapid eye movement (REM) sleep but remain silent during nonrapid eye movement (NREM) sleep (Knox, 2016;Lee, Hassani, Alonso, & Jones, 2005). Furthermore, the activation of cholinergic neurons can enhance arousal, attention, and memory (Eggermann, Kremer, Crochet, & Petersen, 2014;Everitt & Robbins, 1997;Fu et al., 2014;Jones, 1993;Sarter, Hasselmo, Bruno, & Givens, 2005). Impairment of spatial memory was found to be related to injury of cholinergic neurons and hypofunction of GABAergic neurons in the basal forebrain (Jeong, Chang, Hwang, Lee, & Chang, 2011). The glutamatergic and GABAergic neurons were also found to regulate the transition of cortical activity and sleep-wake states (Fuller, Sherman, Pedersen, Saper, & Lu, 2011;Kroeger et al., 2017). As regards the neural circuit, the basal forebrain is involved in nearly all types of memories by interconnecting the hippocampal and septo-hippocampal circuits, such as fear memory and spatial reference memory (Blake & Boccia, 2018;Givens, Williams, & Gill, 2000;Hall, Gomez-Pinilla, & Savage, 2018). Nevertheless, it is still unclear whether the basal forebrain is involved in memory impairment caused by acute sleep deprivation.
However, whether basal forebrain BDNF/TrkB pathway is involved in memory impairment followed by acute sleep deprivation is not clear (Watson, 2015).
We proposed that increased BDNF expression in the basal forebrain following acute sleep deprivation acted as a compensatory mechanism to maintain fear memory performance and conducted this study.

| Animals
Adult male Wistar rats weighing 280 to 320 grams (n = 120; purchased from 301 Experimental Animal Center of PLA Medical College) were used in this experiment. The animals were housed in pairs under a 12-hr/12-hr light-dark cycle (lights on at 08:00 AM) in a temperature-controlled (23 ± 1°C) animal colony room with free access to food and water.

| Ethical statement
All procedures involving animals were approved by the Ethics Committee of the Chinese PLA Medical School and were performed in accordance with the Guide for the Care and Use of Laboratory Animals.

| Acute sleep deprivation
The gentle handling-induced sleep deprivation protocol (Oonk, Krueger, & Davis, 2016) was followed. This procedure of sleep deprivation was found to have little effect on stress response. Sleep deprivation started at 8:00 a.m. and lasted for 6 hr. During the procedure, the rats could behave naturally and had free access to food and water. When the rats closed their eyes or stopped whisking, perturbations were used to keep them awake. The perturbation strategies included a soft writing brush to stir the bedding and a finger snap to make noise and slightly tapping or rotating the cage. If the methods above were not effective, rats were touched with the writing brush. No novel objects were used.

| Experiment 1
To observe the impact of sleep deprivation on fear memory and BDNF/ tropomyosin receptor kinase B (TrkB) pathway activity in K E Y W O R D S basal forebrain, brain-derived neurotrophic factor, fear memory, sleep deprivation the basal forebrain (see Figure 1, Experiment 1). Thirty-three rats were randomly divided into two groups (n = 18 for each group) depending on whether subjected to sleep deprivation (SD) or not (RC). Immediately after sleep deprivation or control, fear training in the step-down inhibitory avoidance test was initiated. Short-term memory (STM) and long-term memory (LTM) were then assessed 1 hr and 24 hr after training, respectively (n = 10 per group). To determine the expression of brain-derived neurotrophic factor (BDNF) in the basal forebrain, the other 16 rats were euthanized and decapitated to collect the basal forebrain for Western blot (n = 4 per group) and immunohistochemical staining (n = 4 per group) one hour after behavior training just before the STM test.

| Experiment 2
To observe whether modulation of the BDNF/TrkB pathway in the basal forebrain could rescue the deficit of fear memory caused by sleep deprivation (seeFigure 1, Experiment 2). BDNF (#3897, CST) or ANA-12 (S7745, Sellect.cn) was used to activate or inhibit basal forebrain TrkB signaling, respectively, before inhibitory avoidance training. Artificial cerebrospinal fluid (aCSF) was used as vehicle control for BDNF and ANA-12. Eighty-four rats were randomly divided into the six groups (n = 14 per group) according to sleep deprivation/ control and/or BDNF/ANA-12 treatment. The six groups were rest control followed by drug vehicle microinjection (RC + aCSF), rest control followed by BDNF microinjection (RC + BDNF), rest control followed by ANA-12 microinjection (RC + ANA-12), sleep deprivation followed by drug vehicle microinjection (SD + aCSF), sleep deprivation followed by BDNF microinjection (SD + BDNF), and sleep deprivation followed by ANA-12 microinjection (SD + ANA-12). Animals underwent bilateral cannula implantation with the tip 1mm above the basal forebrain. The rats were allowed 10 days of recovery after surgery. Then, animals were subjected to sleep deprivation or not, which was followed by fear training. One hour after fear training, 24 rats were euthanized to collect basal forebrain samples for Western blot examination of BDNF/TrkB signaling activity. The left 60 rats were subjected to both STM and LTM examination. After LTM evaluation, the rats were sacrificed. Hematoxylin and Eosin staining of brain was performed to verify the injection.

| Step-down inhibitory avoidance test
The step-down apparatus was a 50 × 25 × 25 cm plastic cubic box fitted with a front transparent wall. The floor of the equipment was made up of steel bars, and a rubber platform (2.5 × 7 × 25 cm) was fixed on the left side of the floor. During the training session, the rat was placed on the platform, facing toward the steel bars. The rats received 0.4 mA/2.0 s electric shock (Izquierdo et al., 1995;Roesler et al., 2000) when they stepped down and put all four paws onto the bars. After training, the animals were placed again on the platform.
Latency of step-down to the steel bars was considered as a measure of memory retention for the aversive stimulus. Short-and long-term memories were evaluated 1 hr and 24 hr after training, respectively. The maximum observation time was 300 s.

F I G U R E 1
Experimental design schematic. The study consisted of two experiments. Experiment 1 involved assessment of acute 6-hr sleep deprivation-induced short-term fear memory (STM, 1 hr after training) and long-term fear memory (LTM, 24 hr after training) impairments. The rats were divided into the rest control (RC) and sleep deprived (SD) groups. The effects of sleep deprivation on basal forebrain BDNF protein expression, and TrkB/PLCγ1 phosphorylation were also evaluated 1 hr after behavior training. In Experiment 2, exogenous BDNF, TrkB receptor antagonist ANA-12, or the vehicle control was infused before fear memory training to test the modulating effects of BDNF/TrkB signaling on sleep deprivation-induced fear memory impairments. BDNF (250 ng/1 μl/side) was infused 10 min before fear memory training, and ANA-12 (0.5 μg/1 μl/side) was injected 1 hr before fear memory training. After LTM evaluation was finished, the rat brain was collected and subjected to hematoxylin and eosin (HE) staining to verify the injection site. Only data from rats with correction injection site were used

| Stereotaxic surgery
Rats were secured in a stereotaxic frame (RWD, China) under 2% isoflurane anesthesia, and surgical procedures were performed with further local infiltration of ropivacaine (0.5%, 1ml, AstraZeneca). The core body temperature of the rats was kept at around 37℃ with a thermostatic heating pad during surgery. The surgical area was disinfected, then 26-gauge (OD = 0.46mm) guide cannulas (RWD, China) were inserted bilaterally at the area 1mm above the basal forebrain in 84 rats (n = 14 each group) (relative to Bregma: Anterior-Posterior (AP) −1.06 mm, Medial-Lateral (ML) ±2.57 mm, and Dorsal-Ventral (DV) −9 mm) and each cannula with a dummy cannula (30 gauge, OD = 0.31 mm) inserted was fixed on the skull with dental cement.
Rats were single-housed for 10 postoperative days. Schematic drawings of coronal brain sections from the 5th The Rat Brain In Stereotaxic Coordinates (Paxinos & Watson, 2005), and a digital photograph of a representative brain section with the probe track are shown in Figure 2.

| Tissue collection
The basal forebrain tissue was freshly obtained on ice according to reports from a previous study (Basheer, Porkka-Heiskanen, Stenberg, & McCarley, 1999). Briefly, a 1-mm-thick coronal slice of brain was obtained by cutting in front of and behind the optic chiasm. Then, a horizontal cut was made on the coronal slice through the middle between the anterior commissure and the bottom of slice. Finally, the slice was cut vertically 1 mm lateral to the third ventricle and the middle of the olfactory tubercle on each side. The 1 mm × 1 mm × 2 mm basal forebrain tissue from each side obtained contains the magnocellular preoptic area (MCPO), the nucleus of the horizontal limb of the diagonal band (HDB), the substantia innominata (SI), and the basal nucleus (Basheer et al., 1999). Tissues were stored in a freezer at −80°C for further tests.

| Immunohistochemistry
Immunohistochemistry was performed to investigate the BDNF expression changes in the basal forebrain after 6-hr sleep deprivation.
The brains of the rats were collected and fixed in 4% paraformaldehyde in PBS, and then, 5 μm of coronal sections were obtained from the paraffin block by cutting through the optic chiasm. Sections were blocked in Tris-buffered saline with 0.1% Tween 20 (TBST) containing 5% goat serum for 1 hr at room temperature and then F I G U R E 2 Microinjection probe tip location in the basal forebrain. A representative microinjection probe track of the coronal brain section is shown on the left, which is consistent with a schematic drawing of the coronal brain section of the 5th The Rat Brain In Stereotaxic Coordinates (Paxinos & Watson, 2005)  and four visual fields were randomly selected and observed for each tissue section. The cells were counted, and the mean rates of positive cells from each rat were compared between the rest control and sleep deprivation groups.

| Western blot
The proteins of the basal forebrain were extracted with a protein extraction kit (C500007, Sangon Biotech). The protein con- to TrkB or PLCγ1, respectively, and relative protein expressions were compared among groups.

| Statistical analysis
The SPSS 18.0 software (https://www.ibm.com/produ cts/spss-stati stics SPSS, RRID:SCR_002865) was used for statistical analysis. As the behavioral test data did not follow a normal distribution, the data were expressed as median and interquartile combined with scatter dot diagrams. The behavioral data were further ranked, and the ranks were analyzed using one-way analysis of variance (ANOVA) with post hoc Bonferroni pairwise comparison. Western blot and immunohistochemistry data were expressed as mean＋standard deviation (SD).
One-way ANOVA with post hoc Bonferroni's test was used for pairwise comparison. Statistical significance was set at p < .05 (two-sided). Figure 3a showed the short-term memory evaluated as the latency.

| Acute sleep deprivation impairs both shortand long-term fear memory
The latency was significantly decreased after sleep deprivation (F 1,18 = 8.019, p = .011) compared with rest control. The latency measured 24 hr after the training was also decreased by sleep deprivation as compared with control (F 1,18 = 7.007, p = .016, Figure 3b).

| Microinjection of BDNF into the basal forebrain rescued fear memory deficit by sleep deprivation
BDNF contents in the basal forebrain were significantly increased after BDNF microinjection (F 3,12 = 90.893, Figure 4a  For the rest control rats, exogenous BDNF caused no differences in the short-and long-term memory performance.

| TrkB receptor antagonist ANA-12 aggravated sleep deprivation-induced fear memory impairments
The Western blot results showed that TrkB phosphorylation in the basal forebrain (Figure 6a,c-e) was significantly decreased

| D ISCUSS I ON
We found that 6-hr total sleep deprivation-induced impairments of fear memory were accompanied with increased brain-derived neurotrophic factor (BDNF) protein levels in the basal forebrain.
Subsequently, we injected exogenous BDNF or tropomyosin receptor kinase B (TrkB) antagonist (ANA-12) into the basal forebrain of rats before inhibitory avoidance training. The results showed that bilateral microinjection of BDNF activated the BDNF/TrkB pathway and partly reversed short-and long-term impairments in fear memory induced by acute sleep deprivation ( Figure 5). Moreover, the selective TrkB receptor antagonists ANA-12 inhibited the downstream phosphorylation of TrkB (Y816) and phospholipase C-γ (PLCγ) (Y783) and aggravated sleep deprivation-induced fear memory deficits. Taken together, these results supported our hypothesis that enhanced basal forebrain BDNF/TrkB signaling acted as a compensation to counteract fear memory impairments caused by sleep deprivation.
Expression of BDNF has been shown to be related to the sleep debt of rats. For example, the levels of BDNF in the cerebral cortex rise with extended wakefulness (Cirelli & Tononi, 2000). The increase of BDNF after sleep deprivation in this research (Figure 3c) were consistent with human studies and rat studies following sleep deprivation protocols (Giacobbo et al., 2016;Wallingford, Deurveilher, Currie, Fawcett, & Semba, 2014). Increase in BDNF protein level could be caused by enhanced expression and secretion, or reduced degradation of BDNF (Lessmann & Brigadski, 2009). After acute sleep deprivation, an exon-specific increase in the expression of BDNF transcripts 1, 4, and 9a was found in the basal forebrain, while the 5-methylcytosine DNA modification, which is important for the daily regulation of BDNF in the basal forebrain, was absent (Ventskovska, Porkka-Heiskanen, & Karpova, 2015). Based on these results, the increased expression of BDNF might be attributed to increased transcription and translation.
It is tempting to relate the preservation of fear memory with the increased BDNF levels seen in the sleep deprived rats. Indeed, brain adaptive response after acute sleep deprivation has been confirmed in gene expression (Cirelli, Gutierrez, & Tononi, 2004), neurotransmitter release (Dash, Douglas, Vyazovskiy, Cirelli, & Tononi, 2009), and compensatory recruitment of different brain structures, such as hippocampus (Yan et al., 2019). In this context, and considering the pivotal functions of BDNF in memory consolidation, elevation of transcription of BDNF during acute sleep deprivation could play a modulating role of the fear memory process.
BDNF could activate three main signaling pathway through binding to the TrkB receptor. The Ras/mitogen-activated protein kinase (MAPK) pathway and the phosphoinositide 3-kinase (PI3K) pathway are activated primarily through Shc/FRS-2 binding to Y515, whereas the PLCγ pathway is activated through Y816 phosphorylation (Minichiello et al., 2002). Among them, BDNF/TrkB (Y816)/ PLCγ (Y783) activation was necessary and sufficient to mediate synaptic plasticity and was involved in short-and long-term fear memory formation Minichiello et al., 2002). Our results found that increased basal forebrain BDNF expression was accompanied with enhanced activation of TrkB (Y816)/ PLCγ (Y783), implicating a role of this signaling pathway in sleep deprivation-induced fear memory impairments. We further used ANA-12, a highly potent and selective TrkB inhibitor to testify this. We found that injection of ANA-12 abolished the increase in p-TrkB and p-PLCγ1 levels induced by sleep deprivation which were consistent with results from recent study (Contreras-Zárate et al., 2019). Furthermore, ANA-12 aggravated sleep deprivation-induced memory impairments ( Figure 5). These results suggested BDNF/TrkB (Y816)/ PLCγ (Y783) activation as a compensatory way to reduce the adverse effects of sleep deprivation on memory.
The 75-kDa neurotrophin receptor (p75 NTR ) is another transmembrane receptor for BDNF. P75 NTR is a low-affinity receptor for BDNF but a high-affinity for precursor of BDNF (proBDNF) (Matsumoto et al., 2008;Woo et al., 2005;Yang et al., 2009). Interestingly, BDNF and proBDNF play distinct roles in fear memory processing . The administration of cleavage-resistant proBDNF or its antibody into the medial prefrontal cortex (mPFC) could facilitate or block fear extinction (Sun, Li, & An, 2018). Further studies are needed to investigate the role of basal forebrain P75 NTR signaling in sleep deprivation-induced memory deficits.

| CON CLUS ION
Acute sleep deprivation induces compensatory increase of BDNF expression and activation of TrkB/PLCγ1 signaling in the basal forebrain. Microinjection of BDNF into the basal forebrain mitigates the fear memory impairments caused by sleep deprivation by activating TrkB/PLCγ1 signaling.

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interest to declare. The authors are willing to meet costs of color reproduction if needed.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used during the study are available from the corresponding author by request.