Cholecystokinin octapeptide improves hippocampal glutamatergic synaptogenesis and postoperative cognition by inhibiting induction of A1 reactive astrocytes in aged mice

Abstract Aims Delayed neurocognitive recovery (dNCR) is a common postoperative complication in geriatric surgical patients for which there is no efficacious therapy. Cholecystokinin octapeptide (CCK‐8), an immunomodulatory peptide, regulates memory and learning. Here, we explored the effects and mechanism of action of CCK‐8 on dNCR. Methods We applied laparotomy to establish a model of dNCR in aged mice. Morris water maze and fear conditioning tests were used to evaluate cognition. Immunofluorescence was used to detect the density of CCK‐8, A1 reactive astrocytes, glutamatergic synapses, and activation of microglia in the hippocampus. Quantitative PCR was performed to determine mRNA levels of synapse‐associated factors. A1 reactive astrocytes, activated microglia, and glutamatergic synapse‐associated protein levels in the hippocampus were assessed by western blotting. Results Administration of CCK‐8 suppressed the activation of microglia, the induction of A1 reactive astrocytes, and the expression of tumor necrosis factor alpha, complement 1q, and interleukin 1 alpha in the hippocampus. Furthermore, it promoted glutamatergic synaptogenesis and neurocognitive recovery in aged dNCR model mice. Conclusion Our findings indicated that CCK‐8 alleviated cognitive impairment and promoted glutamatergic synaptogenesis by inhibiting the induction of A1 reactive astrocytes and the activation of microglia. CCK‐8 is, therefore, a potential therapeutic target for dNCR.


| INTRODUC TI ON
Delayed neurocognitive recovery (dNCR) is a common postoperative complication in elderly surgical patients, 1 which is defined as cognitive decline up to 30 days postoperatively and characterized by impaired memory, learning, and attention. 2 dNCR is associated with longer hospital stays, higher mortality and heavier social burden. 3 However, the neuropathogenesis of dNCR is unclear and lack of effective treatment.
Astrocytes, as the most abundant glial cell in the central nervous system (CNS), play a key role in health and disease. 4 Extensively branched processes of astrocytes are in close proximity to synapses; therefore, their role in regulating the functions of synapses has been widely discussed. [5][6][7] Different properties of activated astrocytes have been demonstrated in age-associated CNS diseases, such as Alzheimer's (AD), Parkinson's (PD), and Huntington's (HD) diseases. 8,9 Reactive astrocytes are classified as neurotoxic type A1 and neuroprotective type A2. Numbers of A1 reactive astrocytes are increased in the brain of AD, PD, HD, and multiple sclerosis patients. 10 Several studies show that A1 reactive astrocytes are induced by activated microglia-released factors, including tumor necrosis factor alpha (TNFα), complement component 1 q subcomponent (C1q) and interleukin 1 alpha (IL-1α). 10,11 Fei et al. 12 demonstrated that etomidate (a common clinical general anesthetic)induced long-term synaptic inhibition and cognitive dysfunction via A1 reactive astrocytes in aged mice. However, it is still unknown that whether the A1 reactive astrocytes participates in the dNCR.
Therefore, in the present study, we investigated the neurotoxicity of A1 reactive astrocytes on hippocampal glutamatergic synaptogenesis and the changes of synaptogenesis-associated factors in dNCR.
Cholecystokinin (CCK) is a 33-amino acid peptide hormone found in the gastrointestinal tract and it is the most abundant peptide neurotransmitter in the brain, 15 with especially high levels in the hippocampus, amygdala, hypothalamus, and ventral tegmental area. 16 CCK regulates feeding, learning, memory, and nociception by binding to two G-protein receptors: CCK1 and CCK2. [17][18][19] CCK has numerous isoforms, including CCK-58, CCK-22, CCK-8, CCK-5, and CCK-4. 20 CCK-8 is one of the most abundant isoforms and mediates the biological functions of CCK in the CNS. 21 Although several studies have confirmed the involvement of CCK in aging and neurodegenerative disease-induced memory impairment, 22-24 the functions of CCK in perioperative cognitive impairment have not yet been reported.
In the present study, we found that surgery/anesthesia decreased the levels of CCK-8 in the hippocampus, especially in CA1 and dentate gyrus (DG) regions. Asrican et al. showed that reduced CCK abundance induced reactive astrocytes and increased the expression of genes involved in neuroinflammation in the DG. 25 Therefore, we speculated that induction of A1 reactive astrocytes by reduced levels of CCK-8 participates in the neuropathogenesis of dNCR.
Furthermore, we explored the cognitive protective effect and possible mechanism of action of CCK-8.
The mice were group housed, five per cage, with a 12 h light/dark cycle in a temperature-controlled (24 ± 1℃) room with free access to food and water in accordance with the standards established by the Experimental Animal Laboratory. Mice were acclimatized to these housing conditions for a week before commencement of the experiment. Every effort was made to minimize the pain and discomfort of animals. The study protocol was approved by the Animal Ethics Committee of Peking University Health Science Center (LA2021423) and conducted in accordance with the Guiding Principles for the Care and Use of Animals in Research, the ARRIVE 2.0 guidelines. 26 The diagram of the experiment is shown in Figure 1A.
After the surgical site was shaved and sterilized, a 1.5 cm vertical incision was made 0.5 cm below the right costal margin through the skin and muscle wall. After exploring the abdominal organs (liver, spleen, kidneys, and bowel), a 10 cm section of the small intestine was removed from the abdominal cavity and rubbed vigorously between the index finger and thumb for 30 s. The small intestine was placed back into the abdominal cavity and the incision closed with 4-0 sutures. Sevoflurane anesthesia was stopped and lidocaine cream was applied to reduce incision pain every eight hours for three days after surgery. During this procedure, which lasted ~30 min, mice were placed on a heating pad to maintain body temperature. Mice in the sevoflurane group received the same sevoflurane exposure without laparotomy, while mice in the control group received no treatment.

| Morris water maze
The Morris water maze (MWM) is used for hippocampus-dependent tests that reflect the spatial navigation and reference memory of rodents. 31 The MWM (Sunny Instruments Co. Ltd., Beijing, China) consists of a circular white tank (120 cm in diameter and 50 cm high) containing water (23 ± 1℃) that is divided into four quadrants and a platform (10 cm in diameter) located 1 cm below the water in the target quadrant. In the place navigation test, the mice were placed in one quadrant facing the wall of the maze and allowed to explore for the hidden platform for 90 s in each trial (four trials per day with an intertrial interval of 5 min). The time to locate the submerged platform was recorded (defined as the escape latency). If the platform was not found within 90 s, the mice were guided to the platform, where they stayed for 15 s. Mice underwent daily testing in the water maze from day 1 to day 5 after surgery. On postoperative day 6, the submerged platform was removed from the water maze and a spatial probe test was performed for 60 s. The swimming speed, escape latency, number of platform crossings, and the time spent in the target quadrant were recorded by a video camera.

| Fear conditioning
Fear conditioning consists of a training phase to establish long-term memory and a testing phase. [32][33][34] Training was performed 1 day prior to surgery. Mice were allowed to adapt to the conditioning chamber for 2 min, and then a 70 dB tone was played (conditional stimulus, 20 s) followed by a trace interval of 25 s. Next, a 2 s 0.7 mA electric foot-shock was administered (unconditional stimulus). Six pairs of conditionalunconditional stimuli with 60 s intervals between each pair were administered. The mice were then allowed to stay in the chamber for 60 s.
The testing phase consisted of a context test and a tone test. In the context test, mice were put back in the chamber for 5 min without tone or electric foot-shock. The tone test was performed 2 h after the context test. Mice were placed in a chamber that was different from the training phase for 5 min. During this period, mice were given sound stimuli (70 dB 3 min) without electric foot-shock. The total distance and freezing time were recorded using tracking system software (Macroambition S&T Development Co. Ltd., Beijing, China). After each test session, the chamber was wiped with 75% alcohol to avoid an odor effect.

| Western blotting
Hippocampus samples were homogenized using radioimmunoprecipitation lysis buffer containing protease and phosphatase inhibitors. The lysate was centrifuged at 15,000 g for 10 min at 4℃ to remove debris. Supernatants were collected, and protein concentration was determined using a bicinchoninic acid protein assay kit peroxidase-conjugated secondary antibody for 1 hour at room temperature. The membranes were then exposed to a chemiluminescence reagent (TianGen, Beijing, China, PA112-02) and then analyzed using Image J (National Institutes of Health, USA).

| Immunofluorescence
Mice were deeply anesthetized with an overdose of sevoflurane and transcardially perfused with PBS, followed by 4% paraformaldehyde in PBS. Brains were removed and post-fixed in 4% paraformaldehyde at 4℃ for 24 h. Brains were then coronally sectioned at 30 µm

| Quantitative real-time PCR (qPCR)
Total RNA was extracted from the hippocampus using TRIzol (TianGen, Beijing, China, DP424) in accordance with the manufacturer's protocol. Total RNA concentration and purity were determined using a NanoDrop spectrophotometer (NanoDrop Technologies).
RNA samples were reverse transcribed into cDNA using the Fastking 1st Strand cDNA Synthesis Kit (TianGen, KR116) and qPCR was performed using SYBR Green Talent qPCR PreMix (TianGen, FP209) according to the manufacturer's instructions. β-actin was used as an internal control. Relative gene expression was calculated using the 2 −ΔΔCt method and the primers used are listed in Table 1.

| Enzyme-linked immunosorbent assay (ELISA)
The concentration of CCK-8 in the hippocampus was quantified using an ELISA kit in accordance with the manufacturer's instructions (Reddot Biotech, British Columbia, Canada, RD-CCK-8-Mu).

| Statistical analysis
GraphPad Prism 6 (GraphPad, New York, USA) was used for statistical analyses. The Shapiro-Wilk test was used to analyze the normality of the data, and we found that the data were normally distributed. Quantitative data are expressed as the mean ± standard error of the mean (SEM). Statistical significance was determined using analysis of variance (ANOVA), followed by Bonferroni's post hoc test. A p-value <0.05 was considered statistically significant.
Statistical power analysis was used to verify significant differences with respect to sample size (GPower 3, ≥0.8 for sufficient power validation). 35

| Surgery/anesthesia decreases the levels of CCK-8 in the hippocampus of aged mice
We previously showed that 30 min of inhalation anesthesia alone did not induce cognitive impairment in mice. 32,36 Here, we examined the effect of sevoflurane anesthesia with or without surgical trauma on the levels of CCK-8 in the hippocampus of aged mice. We found no significant differences in CCK-8 levels between the control and sevoflurane groups ( Figure 1B-F

| CCK-8 alleviates hippocampus-dependent memory impairment in aged dNCR model mice
On the account of the postoperative reduction in CCK-8 levels, we explored the effect of CCK-8 supplementation on cognitive function in aged dNCR model mice.
We used the MWM to evaluate spatial learning and memory. 31 There was no significant difference in swimming speed between groups of mice ( Figure 2A). However, we found that surgery/anesthesia significantly increased the escape latency on postoperative days 2-5 ( Figure 2B

| CCK-8 upregulates the levels of PSD95 and vGLUT1 in the hippocampus of aged dNCR model mice
The density of hippocampal glutamatergic synapses is associated with cognition in rodents and co-localization of PSD95 and vGLUT1 is commonly used to label the synapses of glutamatergic neurons. 14 Therefore, we explored the relationship between improve-   Synapses are basic information processing units in the brain. 44,45 The correct number and type of synaptic connections are essential for normal CNS functions. Specifically, glutamatergic synapses play a pivotal role in memory formation, learning, and emotion. 46 Xiong and colleagues found that a decreased number of hippocampal glutamatergic synapses induced cognitive impairment in rats with neuropathic pain. 14 Here, we found that CCK-8 upreg- In the past 20 years, numerous studies have reported key roles of astrocytes in regulating synapse formation and function. 47 Inflammation or trauma have been demonstrated to induce A1 reactive astrocytes, which are involved in CNS diseases. 48 Taylor et al.

| CCK-8 reduces the density of A1 astrocytes in the hippocampus of aged dNCR model mice
showed that disruption to immune response networks can induce A1 reactive astrocytes, which participate in the pathogenesis of ADrelated dementia, such as early-stage cerebral amyloid angiopathy. 49 Furthermore, etomidate induces synaptic inhibition and cognitive decline in aged mice by inducing A1 reactive astrocytes. 12 Here,   55 ; therefore, we investigated the effect of CCK-8 on microglia. We showed that CCK-8 significantly inhibits activation of microglia and the release of inflammatory factors, including IL-1α, TNFα and C1q. To the best of our knowledge, this is the first report of CCK-8 to inhibit the release of IL-1α, TNFα, and C1q, which complements the neuroprotective mechanism of CCK-8.
Li and colleagues found that microglia depletion before etomidate administration could inhibit etomidate-induced A1 reactive astrocyte activation. 12 Furthermore, a rat model of chronic postsurgical pain confirmed that microglia induce A1 reactive astrocytes via the CXCR7/PI3K/Akt pathway. 11 Thus, we speculated that CCK-8 reduced A1 reactive astrocyte induction by inhibiting microglia activation. However, these data must be interpreted with caution because CCK2 receptors are also located on astrocytes. 25 CCK released by CCK interneurons increases neurogenic proliferation of glia-like neural stem cells by inhibiting A1 reactive astrocytes. 25 We did not use astrocyte CCK2 receptor conditional knockout mice to verify the effect of CCK on neurogenic proliferation; therefore, the direct effect of CCK on astrocytes via CCK2 receptors cannot be excluded.
The present study had several limitations. First, we only used female mice, although a recent study demonstrated no difference in behavior and physiological state between female and male mice. 56 Second, the lidocaine cream applied relieves incisional pain, but not visceralgia in aged mice. Therefore, we cannot exclude the effect of postoperative visceralgia on the cognition of mice. Third, we did not assess surgery/anesthesia-induced peripheral inflammatory factors entering the CNS via a damaged blood brain barrier, which is one of the most important pathogenic mechanisms of dNCR. 57 Data are expressed as the mean ± SEM (one-way ANOVA followed by Bonferroni's post hoc test, n = 6 per group). *p < 0.05; **p < 0.01, ***p < 0.001 versus the control group; #p < 0.05, ##p < 0.01 versus the surgery group plays a pivotal role in synaptic transmission plasticity, which is often used in studies of learning and memory. 59 However, we did not observed the effect of CCK-8 on long-term potentiation.
Finally, we focused only on short-term postoperative cognitive impairment; however, the effects of CCK-8 on long-term cognition in aged mice warrant further study.

| CON CLUS IONS
CCK-8 alleviates postoperative cognitive impairment and promotes glutamatergic synaptogenesis in the hippocampus via inhibition of A1 reactive astrocytes, which may be mediated by inhibition of microglia activation. These findings indicate CCK-8 to be a potential therapeutic target for dNCR.

ACK N OWLED G M ENTS
This study was supported by the grants from the National Natural

CO N FLI C T O F I NTE R E S T
The authors declare that there is no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets analyzed in the present study are available from the corresponding author on reasonable request.