Aging-Related Neural Disruption Might Predispose to Postoperative Cognitive Impairment Following Surgical Trauma

Abstract

Background:

Accumulating evidence has demonstrated that aging is associated with an exaggerated response to surgical trauma together with cognitive impairments. This has significant implications for the development of clinical phenotype such as perioperative neurocognitive disorders (PND), which is a common complication following surgery, especially for the elderly. However, the mechanism by which aging brain is vulnerable to surgical trauma remains to be elucidated.

Objective:

To test whether age-related alterations in hippocampal network activities contribute to increased risk of PND following surgery.

Methods:

Thirty-two adult and seventy-two aged male C57BL/6 mice undergone sevoflurane anesthesia and exploratory laparotomy were used to mimic human abdominal surgery. For the interventional study, mice were treated with minocycline. Behavioral tests were performed post-surgery with open field, novel object recognition and fear conditioning tests, respectively. The brain tissues were then harvested and subjected to biochemistry studies. Local field potential (LFP) recording was performed in another separate experiment.

Results:

Aged mice displayed signs of neuroinflammation, as reflected by significantly increased proinflammatory mediators in the hippocampus. Also, aged mice displayed persistently decreased oscillation activities under different conditions, both before and after surgery. Further correlation analysis suggested that theta power was positively associated with time with novel object, while γ oscillation activity was positively associated with freezing time to context. Of note, downregulation of neuroinflammation by microglia inhibitor minocycline reversed some of these abnormities.

Conclusion:

Our study highlights that age-related hippocampal oscillation dysregulation increases the risk of PND incidence, which might provide diagnostic/prognostic biomarkers for PND and possible other neurodegenerative diseases.

Keywords

Aging cognitive function inflammation microglia oscillation

INTRODUCTION

Aging is a major risk factor for inflammation-induced mild cognitive impairments and other neurodegenerative diseases [1 –3]. Accumulating evidence has demonstrated that aging is associated with an exaggerated response to inflammatory challenges or surgical trauma together with behavioral and cognitive impairments [4, 5]. This has significant implications for the development of clinical phenotype such as perioperative neurocognitive disorders (PND), which is a common complication following surgery. PND includes cognitive decline diagnosed before operation, postoperative delirium, delayed neurocognitive recovery, which can result in poor patient outcomes such as increased hospital stay, reduced quality of life, loss of social dependence, and increased mortality [6]. It is well established that advanced age is an independent risk factor for the development of PND, as demonstrated by both animal and clinical studies [5 –8]. Current studies regarding the mechanisms underlying PND highlight the role of inflammation and immune activation [7]. However, inflammation occurs in every surgical patient, but only certain individuals develop PND. This can be explained by a decrease in individual’s brain reserve capacity in response to subsequent insult [9]. Consistently, earlier studies have suggested reduced brain reserve capacity, as reflected by preexisting cognitive impairment, is the key risk factor for PND development [10, 11]. Thus, understanding the mechanisms by which aging brain is vulnerable to surgical trauma is an important approach to develop novel treatment strategies for PND.

Many aspects of cognition, including memory and learning, are thought to involve dynamic interactions among ensembles of neurons that are coordinated by different classes of patterned network oscillations [12]. There are multiple different types of network oscillations, which span a wide range of frequencies (∼0.05 to 1000 Hz) and are implicated in different cognitive and behavioral states [13]. For example, theta oscillation (4–12 Hz) modulates long-term potentiation and excitability in the hippocampus and facilitates neural processing and cognitive functions [14], while gamma oscillation (30–80 Hz) is involved in a variety of brain functions including attention [15], sensory processing [16], and memory [17]. Importantly, a specific class of GABAergic interneurons such as PV interneurons plays a key role in organizing these highly coordinated patterns of activity [18]. By contrast, abnormal function of PV interneurons and network oscillations may occur with aging, contributing to cognitive and behavioral impairments [19, 20], while increased working memory evoked gamma oscillatory activity provides a neurophysiological marker in the healthy aging brain [21]. These findings raised a fundamental question as whether abnormal network oscillations in the hippocampus contribute to increased risk of PND induced by surgical trauma.

As the population ages, there is an ongoing public concern regarding the cognitive decline as a result of surgery. The present study thus tested whether age-related alterations in hippocampal network activities contribute to increased risk of PND following surgery.

MATERIALS AND METHODS

Animals

Thirty-two adult (3-4 months, 24–28 g) and seventy-two aged male C57BL/6 mice (16–18 months, 26–35 g) were bred at the Animal Center of Jinling Hospital, Nanjing, China. Mice were housed group (2–4 per cage) at a temperature of 22-23°C and 12:12 h light-dark cycle, with free access to water and food. All studies performed were approved by the Animal Care and Use Committees of Nanjing Medical University, Nanjing, China. Before we started the experiments, mice were allowed to acclimate the new environment for 2 weeks. In our studies, no animal died during the experimental period.

Animal model

Exploratory laparotomy was performed under sevoflurane anesthesia as we described previously [7]. Briefly, mice were anesthetized in an anesthesia chamber prefilled with 2.1%sevoflurane in oxygen. After exposure to sevoflurane for 30 min, then the surgery was performed immediately. An abdominal median incision (approximately 1 cm) was made to allow penetrating the peritoneal cavity. Then the viscera, intestine and musculature were explored by the operator. Sterile 4-0 chromic gut sutures were used to suture the peritoneal lining and skin. For animals that served as controls, neither anesthesia nor surgery was performed.

Drug

Minocycline hydrochloride (Sigma-Aldrich, lot# 095M4083V, 40 mg/kg) is a semi-synthetic tetracycline analog that exerts neuroprotective effects in many neurodegenerative diseases [22]. It was dissolved in normal saline and administered (i.p.) daily immediately following surgery and another two consecutive days. The dose of minocycline was selected based on one previous study [22], in which they showed it can improve memory dysfunction following surgery.

Animal grouping

In the first set of experiment, we measured the time course of inflammatory mediators following surgery in aged mice. Animals were divided randomly and equally into the following groups (n = 6): control + aged group, surgery + aged day 1 group, surgery + aged day 3 group, and surgery + aged day 7 group.

In the second set of experiment, adult mice were randomly divided into the following two groups (n =16): control + young group and surgery + young group, while aged mice were randomly divided into the following three groups (n = 16): control +aged group, surgery + aged group, and surgery +aged + minocycline group. Among them, 11 animals of each group underwent behavioral tests. For local field potential (LFP) recording, 5 animals per group were used. After then, the hippocampi were removed and subjected to immunofluorescence (n = 4), enzyme-linked immunosorbent assay (ELISA, n = 6), and western blot (n = 4), respectively.

Open field test (OFT)

On postoperative day 3, OFT was performed in an open arena (40 long×40 wide×40 cm tall) to evaluate the exploratory behavior and anxiety behavior. Mice were placed in the center of the arena and their activity was assessed for 5 min. Total distance traveled and time exploring the center area were analyzed (Shanghai Softmaze Information Technology Co. Ltd., Shanghai, China). The arena was cleaned with 75%ethanol between each trial.

Novel object recognition test (NOR)

Two hours after OF, the NOR test was performed in an open field (40 cm long×60 cm wide×50 cm tall) with three objects, two of which were almost the same, the other was different. For habituation, animals were allowed to explore the open field for 5 min. Twenty-two hour later, the animal was exposed to two familiar objects for 10 min. To avoid a preference for one side of the open field, two familiar objects were counterbalanced between each mouse. On postoperative day 5, one of the objects was changed into a novel object with different color and shape. The exploration of the new object and familiar object was recorded by a video-tracking system for 10 min. Exploration of an object was defined as the animal’s nose being in the zone at a distance of ≦ 2 cm. The discrimination score for novel object exploration ratio was calculated with the following formula: time exploring novel object/(time exploring novel object + time exploring familiar object)×100%. Equipment and apparatus were cleaned using 70%ethanol between trials.

Fear conditioning tests

On postoperative day 6, mice were subjected to fear conditioning training to evaluate the associative memory. Briefly, mice were placed individually into the conditioning chamber. The mice were allowed to explore the chamber for 180 s and then received a 30 s tone (70 dB, 3 kHz) followed by 2 s foot shock (0.75 mA). Afterward, the mice were left to stay in the training chamber for another 30 s and then placed back into the home cage. Twenty-four hours after training, the mice were exposed to the contextual fear conditioning test. The animals were returned to the same chamber where they were trained to monitor the freezing behavior for 5 min without tone or foot shock. The cued fear memory was tested two hours later in a total new chamber. The mice were allowed to explore the novel chamber and scored for the freezing behavior with a continuous 3 min training tone deliver. Freezing behavior was defined as the absence of any movements except for respiration. Between each testing session, the chamber was thoroughly cleaned with 75%ethanol to avoid the interference of olfactory cues.

ELISA

Mice were killed by an i.p. injection of 2 %sodium pentobarbitone (60 mg/kg) and then the hippocampi were collected, then separated, and placed in a homogenizer. The tissues were homogenized with 1 ml extraction buffer per 100 mg brain tissue. Hypothermal centrifugation was performed at 10,000×g for 10 min and the supernatant was obtained. The levels of tumor necrosis factor (TNF)-α (ab208348), interleukin (IL)-1β (ab197742), IL-6 (ab100712), and IL-10 (ab108870) were detected by ELISA according to the manufacturer’s instructions (Abcam, China). All samples were assayed in duplicate, and the readings were normalized to the amount of standard protein.

Western blot

After behavioral test at postoperative day 7, hippo-campi were dissected from freshly perfused mouse brains and homogenized in ice-cold lysis buffer (1%Nonidet P-40, 0.1%sodium deoxycholate, 0.1%SDS, 66 mM EDTA, and 10 mM Tris-HCl, pH 7.4) supplemented with a protease inhibitor cocktail. The samples were centrifuged at 15,000 g for 10 min at 4°C. The protein concentration was determined by Bradford assay and equal sample (40μg) were loaded per lane and electrophoresed on SDS-PAGE gels. The separated proteins were then transferred to polyvinylidene fluoride membranes. The membranes were blocked with 5%skim milk in Tris-buffered saline with Tween (TBST) and then were incubated with rabbit anti-PV (1:1,000; Abcam, Cambridge, MA, United Kingdom) and mouse anti-GAPDH (1:5,000, Bioworld, St. Louis Park, MN, USA) overnight at 4°C. After the membranes were washed three times in TBST, they were incubated with horseradish peroxidase-conjugated secondary antibodies (goat anti-rabbit and goat anti-mouse, Bioworld Technology, St. Louis Park, MN, USA), diluted to 1:5,000 for 1 h at room temperature. The protein bands were detected by enhanced chemiluminescence, exposed onto X-ray film, and quantitated with Image J software (National Institutes of Health, Bethesda, MD, USA).

Immunofluorescence

After behavioral test on postoperative day 7, the mice were deeply anesthetized and perfused transcardially with saline, followed by 4%paraformaldehyde in phosphate buffered saline. The brains were harvested and postfixed in 4%paraformaldehyde, and then dehydrated in 30%sucrose overnight at 4°C. Samples were embedded in Optimal Cutting Temperature compound, cut into 30-μm-thick sections using a cryostat, and mounted on slides. Slices were blocked with 3%bovine serum albumin for 1 h at room temperature. The sections were then incubated with rabbit anti-PV (1:500; Abcam, ab11427) overnight at 4°C, followed by 1 h incubation with the secondary antibodies (Cy3-conjugated donkey anti-rabbit IgG (1:300; Santa Cruz Biotechnology, TX, USA) at room temperature. For mounting and counterstaining, sections were incubated with 4’, 6-diamidino-2-pheny-lindole (DAPI). The images of each section were acquired by a confocal laser scanning microscope (Leica, TCS SP2, Germany). Image J was used to measure the mean value of the immunofluorescence in each section.

LFP recording

LFP recording was performed as we previously described [23]. Briefly, mice were anesthetized by pentobarbital sodium (40 mg/kg, i.p.) and fixed in a stereotaxic apparatus with left and right ear rods. A longitudinal incision was cut along the middle line of the cranial crest to expose bregma and posterior fontanelle. After craniotomy and removal of dura, an 8-channel linear silicon probes were used to record right CA1 region of the hippocampus. The coordinates were determined according to the mouse brain atlas in stereotaxic coordinates (posterior, 2.1 mm; lateral, 1.5–1.7 mm; depth, 1.7–2.1 mm). LFPs were recorded while the mice were stayed in the home cage or underwent different behavioral tests at the indicated time points. The signals were filtered with a pass-band of 0.3–300 Hz and were further amplified and digitized at 2 kHz. The recorded LFPs were filtered by a 50 Hz notching filter to remove the powerline artifact. For LFP analysis, the wideband recordings were down-sampled at 1000 Hz. All data analyses were performed by Neuroexplorer (Plexon Inc., Dallas, TX) software.

Statistical analysis

Data are presented as Mean±SEM. All statistical analysis was performed by SPSS 19 software (IBM). All data were first tested for normality (Shapiro–Wilk test) and homoscedasticity (Levene’s test). Since all the data in this study was normally distributed and the variance was equal, unpaired t-tests were used for comparison of two groups. Differences among multiple groups were compared by one-way ANOVA followed by Tukey multiple comparison tests. Bivariate relationships were evaluated using Pearson correlation coefficients. A p value < 0.05 was considered as statistically significant.

RESULTS

Aging was associated with enhanced inflammatory response to surgical trauma

To investigate the time course of inflammatory mediators following surgery in aged mice, we measured TNF-α, IL-1β, IL-6, and IL-10 expressions at 1, 3, and 7 days after surgery. There were significantly increased TNF-α (F_3,20 = 6.037, p = 0.0042, Fig. 1B), IL-1β (F_3,20 = 10.99, p = 0.0002, Fig. 1C), and IL-6 (F_3,20 = 13.36, p < 0.0001, Fig. 1D) expressions in the hippocampus at day 1 after surgery compared with control + aged group. At 3 days after surgery, IL-1β and IL-6 expressions in the hippocampus were higher in surgery + aged group compared with control + aged group. In addition, increased IL-1β level persisted at 7 days after surgery. However, there was no difference in IL-10 level among groups (F_3,20 = 2.737, p = 0.0706, Fig. 1E).

Fig. 1

Aging was associated with enhanced inflammatory response to surgical trauma. A) Schematic timeline of the experimental procedure. B-E) Time course expressions of TNF-α, IL-1β, IL-6, and IL-10 in the hippocampus after surgery. F-I) Expressions of TNF-α, IL-1β, IL-6, and IL-10 in the hippocampus 7 days after surgery among groups. Data are shown as mean±SEM (n = 6), ^*p < 0.05 versus control + aged group, ^δp < 0.05 versus surgery + aged group. Min, minocycline.

At 7 days after surgery, there were significantly increased IL-1β (F_3,20 = 9.132, p = 0.0001, Fig. 1G) and IL-6 (F_3,20 = 4.15, p = 0.0103, Fig. 1H) levels in the hippocampus in surgery + aged group compared with control + aged group. Microglia inhibitor minocycline was able to decrease IL-1β but not IL-6 level. There was no difference in TNF-α (F_3,20 = 0.1611, p = 0.956, Fig. 1F) and IL-10 (F_3,20 = 0.1975, p = 0.9373, Fig. 1I) levels among groups.

Aging was associated with PV deficit in the hippocampus of mice, which was aggravated by surgery

As shown in Fig. 2B-C, the expression of PV was decreased in the hippocampus in control +aged and surgery + aged groups compared with control + young group (p = 0.0131 and p = 0.0038, respectively). Compared with control + aged group, the expression of PV was significantly decreased in the hippocampus in surgery + aged group (p =0.0165). However, minocycline treatment was able to reverse the decreased PV expression in surgery +aged + minocycline group as compared with sur-gery + aged group (p = 0.0038).

Fig. 2

Aging induced PV deficit in the hippocampus, which was aggravated by surgery. A) Schematic timeline of the experimental procedure. B) Representative image of PV expression in the hippocampus. C) Quantification of PV level in the hippocampus among groups. D) Representative images of PV interneurons in all subregions of the hippocampus. E) Quantification of mean PV immunofluorescence in the hippocampus. Data are shown as mean±SEM (n = 4), ^*p < 0.05 versus control + young group, ^#p < 0.05 versus control + aged group, ^δp < 0.05 versus surgery + aged group. Scale bar = 100μm. IF, immunofluorescence, Min, minocycline.

As shown in Fig. 2D-E, the intensity of PV was significantly decreased in the CA1 (F_4,15 = 8.726, p = 0.0008) and CA3 (F_4,15 = 8.053, p = 0.0011) regions of the hippocampus in control + aged group compared with the control + young group. Compared with surgery + young group, the intensity of PV was significantly decreased in the CA1 (p = 0.005) and CA3 (p = 0.0125) regions of the hippocampus in surgery + aged group. Minocycline treatment reversed the decreased PV expression in the CA1 (p = 0.0179) and CA3 (p = 0.0086) regions of the hippocampus in surgery + aged + minocycline group compared with surgery + aged group. However, there was no difference in PV expression in the DG among groups (F_4,15 = 0.555, p = 0.6986).

Aging was associated with exaggerated cognitive impairments to surgical trauma

There was no difference in total distance traveled (F_4,50 = 0.07106, p = 0.9905, Fig. 3B) or time spent in the center (F_4,50 = 0.954, p = 0.441, Fig. 3C) among these groups. In the novel object recognition test, there was a significantly decreased time with novel object in surgery + aged group compared with control + aged group (F_4,50 = 5.952, p = 0.0005, Fig. 3D). However, minocycline treatment reversed the decreased time with novel object. In addition, there was a significantly decreased novel object exploration ratio in surgery + aged group compared with control + young group (F_4,50 = 6.361, p =0.0003, Fig. 3E), which was not reversed by minocycline treatment. In the fear conditioning tests, there was a significantly decreased freezing time to context in surgery + aged group compared with control + aged group (F_4,50 = 4.59, p = 0.0031, Fig. 3F), which was not prevented by minocycline treatment. In addition, there was no difference in freezing time to cue among these groups (F_4,50 = 0.7159, p = 0.585, Fig. 3G).

Fig. 3

Aging was associated with exaggerated cognitive impairments to surgical trauma. A) Schematic timeline of the experimental procedure. B,C) There was no difference in total distance in the open arena or time spent in the center among groups. D) Animals in the surgery + aged group had significantly decreased time with novel object compared with control + aged group, which was reversed by minocycline treatment. E) Surgery decreased novel object recognition ratio in surgery + aged group compared with control + young group. F) Surgery decreased freezing time to context in surgery + aged group compared with control + young or control + aged group. G) There was no difference in freezing time to cue among groups. Data are shown as mean±SEM (n = 11), ^*p < 0.05 versus control + young group, ^#p < 0.05 versus control + aged group, ^δp < 0.05 versus surgery + aged group. Min, minocycline.

Aging was associated with hippocampal oscillation disturbance

To evaluate the association between oscillation activities and cognition performance, we recorded LFP recordings in mice under different conditions. As shown in Fig. 4, there was no difference in different bands of oscillation activities at home cage between control + young and control + aged groups before surgery. However, when the mice explored the novel object, the control + aged group displayed significantly lower γ oscillation power (t = 2.958, p = 0.0182) compared with control + young group. These results suggested that γ oscillation activities in the hippocampus varied with the changing cognitive and movement demands.

Fig. 4

Aging was associated with abnormal hippocampal oscillation before surgery in the novel object recognition test. A) Schematic timeline of the experimental procedure. B) Representative images of local field potential in the CA1 of the hippocampus at home cage. C) Representative images of local field potential in the CA1 of the hippocampus in the novel object recognition test. D) Quantification of local field potential in the CA1 of the hippocampus at home cage. E) Quantification of local field potential in the CA1 of the hippocampus in the novel object recognition test. F) Quantification of average theta, alpha, beta, and gamma power in the CA1 of the hippocampus at home cage. G) Quantification of average theta, alpha, beta, and gamma power in the CA1 of the hippocampus in the novel object recognition test. Data are shown as mean±SEM (n = 5), ^*p < 0.05 versus control + young group.

After surgery, there was also no significant difference in different bands of oscillation activities at home cage among groups (Fig. 5). However, mice in surgery + aged group displayed significantly lower θ (F_2,12 = 4.156, p = 0.0425), β (F_2,12 = 5.431, p =0.0209), and γ (F_2,12 = 6.054, p = 0.0152) power compared with surgery + young group when the animals explored the novel object. As expected, minocycline treatment was able to reverse the decreased β (p = 0.0429) and γ (p = 0.0298) but not the θ power. These data suggested that surgery further disrupted θ, β, and γ activities during a cognitive task, which can be partly prevented by minocycline treatment. Further correlation analysis showed that theta power was positively associated with time with novel object (r = 0.5307, p = 0.0418, Fig. 6A), but there was no such an association for other oscillation activities (Fig. 6B-D).

Fig. 5

Aging was associated with abnormal hippocampal oscillation after surgery in the novel object recognition test. A) Schematic timeline of the experimental procedure. B) Representative images of local field potential in the CA1 of the hippocampus at home cage. C) Representative images of local field potential in the CA1 of the hippocampus in the novel object recognition test. D) Quantification of local field potential in the CA1 of the hippocampus at home cage. E) Quantification of local field potential in the CA1 of the hippocampus in the novel object recognition test. F) Quantification of average theta, alpha, beta, and gamma power in the CA1 of the hippocampus at home cage. G) Quantification of average theta, alpha, beta, and gamma power in the CA1 of the hippocampus in the novel object recognition test. Data are shown as mean±SEM (n = 5), ^*p < 0.05 versus surgery + young group, ^#p < 0.05 versus surgery + aged group. Min, minocycline.

Fig. 6

Theta power was positively associated with time with novel object. A) Correlation between theta oscillation and time with novel object. B) Correlation between alpha oscillation and time with novel object. C) Correlation between beta oscillation and time with novel object. D) Correlation between gamma oscillation and time with novel object.

As shown in Fig. 7, the surgery + aged mice showed significantly lower power in θ (F_2,12 = 5.113, p =0.0248), β (F_2,12 = 7.478, p = 0.0078), and γ (F_2,12 = 5.223, p = 0.0236) power compared with surgery + young group. However, there was no difference in α power among groups (F_2,12 = 1.939, p = 0.1863). However, minocycline treatment was not able to reverse these abnormities. The correlation analysis suggested that only γ power was significantly associated with freezing time to context (r = 0.6179, p = 0.0141, Fig. 7H).

Fig. 7

Aging was associated with abnormal hippocampal oscillation after surgery in the fear conditioning test. A) Schematic timeline of the experimental procedure. B) Representative images of local field potential in the CA1 of the hippocampus in the fear conditioning. C) Quantification of local field potential in the CA1 of the hippocampus in the fear conditioning. D) Quantification of average theta, alpha, beta, and gamma power in the CA1 of the hippocampus in the fear conditioning. E) Correlation between theta oscillation and freezing time to context. F) Correlation between alpha oscillation and freezing time to context. G) Correlation between beta oscillation and freezing time to context. H) Correlation between gamma oscillation and freezing time to context. Data are shown as mean±SEM (n = 5), ^*p < 0.05 versus surgery + young group. Min, minocycline.

DISCUSSION

Aging is a major risk factor for inflammation-induced cognitive impairments [1 –3]. In the present study, we confirmed previous findings that aging is associated with an exaggerated response to surgical trauma. Importantly, we showed that impaired oscillation activities were persistent in aged mice during preoperative as well as postoperative period. This existing vulnerable brain can explain why aged patients are more likely to develop PND following surgery. Intriguingly, minocycline treatment was able to reverse some of these abnormities, suggesting a key role of neuroinflammation in PND. Together, our results could be relevant for the treatment and also may contribute to future studies on diagnostic/prognostic biomarkers for PND.

The impacts of inflammation on cognitive function have been widely established in the literature. Notably, inflammation can be activated not only by inflammatory challenges but also by other stimuli such as aseptic trauma [5]. Thus, inflammation is emerging as a common pathway contributing to cognitive decline in the aged and vulnerable brain. Indeed, excessive circulating proinflammatory cytokines in older organisms have been linked to learning and memory impairments [24]. In humans, older adults with high blood concentrations of proinflammatory cytokines perform worse on certain cognitive assessments in comparison to those with low concentrations [25], while non-steroidal anti-inflammatory drugs has protective effects against subsequent development of Alzheimer’s disease (AD) [26]. In the present study, we have demonstrated that aged mice displayed signs of enhanced inflammatory response in particular for IL-1β expression, as well as greater cognitive deficits in response of surgical trauma. Indeed, the hippocampus is vulnerable to increased IL-1β, with elevated levels implicated in the etiology of neurodegenerative diseases in including surgery induced cognitive impairment [22 , 26]. Thus, the enhanced inflammatory response can partly explain the more robust cognitive impairment during aging following surgery.

In normal aging or neurodegenerative disorders, microglia cells lose their supportive role in neuroplasticity and undertake a primed over-reactive phenotype [27]. In case of a peripheral or central inflammatory challenge, the sensitized microglia become hyperactivated and produce pathologic levels of proinflammatory cytokines, thereby interfering with synaptic plasticity processes and consequently resulting in memory decline [22, 28]. In addition, aged subjects are often accompanied by an exaggerated and long- lasting increase of pro- inflammatory cytokines in the hippocampus, resulting in impaired synaptic plasticity and hippocampal-dependent memory deficits [29]. However, this effect was not observed in the hippocampus of aged mice after surgery. The 16–18-month-old mouse used in the current study is actually equivalent to late middle age for human. Despite this discrepancy, our study suggests that such superimposed inflammatory activation can exacerbate the progression of neurodegenerative disease. However, the mechanism by which inflammation predisposes aged subjects to greater deficits in cognitive function remains unclear.

Hippocampal network activity is generated by a complex interplay between excitatory pyramidal cells and inhibitory interneurons, which plays critical roles in the information processing underlying learning and memory [14]. Aberrant network activity caused by impairment of PV interneurons is widely thought to underlie the disturbances in mood, cognition and memory associated with mental illnesses [30, 31]. In particular, PV interneurons are susceptible to stress due to the property of high-energy demands [32]. Disruption of PV expression is observed in various neuropsychiatric diseases as well as during aging [30 , 33]. Our previous study has suggested that neuroinflammation-induced hippocampal PV interneuron phenotype loss and consequent cognitive impairment in aging mice following surgery [34]. Our current study also suggested that minocycline treatment was able to reverse decreased PV expression, suggesting neuroinflammation might be an upstream of hippocampal PV interneuron loss and aberrant network activity. Moreover, PV interneuron progressively decreases during ageing in animal studies [35 –37]. In support, we confirmed decreased PV expression in the aged brain, which was not further decreased by surgery. This neurodegenerative brain has reduced brain reserve, which can explain why aged patients are more likely to develop PND following surgery. Indeed, it has been suggested reduced brain reserve capacity is the key risk factor for PND incidence in clinical studies [10, 11].

Optimal neural circuit function is critical for effective information processing that mediates cognitive processes [12, 13]. Notably, our study showed that aged mice displayed network dysregulation, especially for theta and gamma oscillations. In particular, oscillations in theta and gamma frequency range are associated with working memory and other tasks of cognition [30, 38]. On the other hand, reduced gamma oscillation is found in the elderly population that may be related to general cognitive decline [39]. In an animal of AD, it has been shown APP/PS1 mice showed an age-dependent decrease in hippocampal theta power [38]. These findings suggest that the existing neurodegenerative brain, as reflected by hippocampal oscillation dysregulation, might explain why aged patients are more likely to develop PND following surgery. While acute elevation of inflammatory mediators can produce cognitive impairment, the symptoms of PND are relatively long-lasting. It is possible that inflammatory mediator elevations following surgical trauma in the early phase further disrupts of hippocampal oscillation dysregulation and contributes to long-term cognitive dysfunction. In our study, this disrupted oscillation activities actually precedes enhanced inflammation, suggesting existing neurodegenerative alternation. Further correlation analysis suggested that decreased theta power is associated with working memory dysfunction, while γ oscillation activity deficit is related to fear learning impairment. These results suggested that different oscillation activities participate in different cognitive tasks. Thus, these persistent abnormal oscillation activities in aged mice, at least in part, contribute to the vulnerability to subsequent surgical trauma.

There are few limitations in our study. Firstly, surgical trauma may affect the feeding/drinking/sleeping of the mice, which might influence behavioral interpretation of our study. Thus, our further studies should exclude these confounding factors. Secondly, to exclude the direct effect of increased inflammatory mediators on cognitive impairment, longer time neurobehavioral tests should also be performed when the inflammatory response has normalized following surgery. Finally, to exclude the influence of estrogens on cognitive performance, we only used male mice in our study. Therefore, female mice should also be included in our future studies.

In conclusion, our study suggests that aged related hippocampal oscillation dysregulation might play a key role in the development of PND. This knowledge could be relevant for the treatment of PND, and also may contribute to future studies on diagnostic/prognostic biomarkers for PND and other neurodegenerative diseases. Future studies, presumably using specific tools such as optogenetic manipulation, will be important to understand the causal role of oscillation dysregulation in PND.

Footnotes

ACKNOWLEDGMENTS

This study was supported by the grants from the National Natural Science Foundation of China (Nos., 81771156, 81772126, 81971020, 81971892), Jiangsu Province’s Key Provincial Talents Program (QNRC2016822), and Six Talent Peaks Project of Jiangsu Province (WSW-002).

Authors’ disclosures available online ().

References

Singhal

, Morgan

, Jawahar

, Corrigan

, Jaehne

, Toben

, Breen

, Pederson

, Manavis

, Hannan

(2020) Effects of aging on the motor, cognitive and affective behaviors, neuroimmune responses and hippocampal gene expression. Behav Brain Res 383, 112501.

Van Skike

, Lin

, Roberts Burbank

, Halloran

, Hernandez

, Cuvillier

, Soto

, Hussong

, Jahrling

, Javors

, Hart

, Fischer

, Austad

, Galvan

(2020) mTOR drives cerebrovascular, synaptic, and cognitive dysfunction in normative aging. Aging Cell 19, e13057.

Harrison

, Maass

, Adams

, Du

, Baker

, Jagust

(2019) Tau deposition is associated with functional isolation of the hippocampus in aging. Nat Commun 10, 4900.

Schreuder

, Eggen

, Biber

, Schoemaker

, Laman

, de Rooij

. Pathophysiological and behavioral effects of systemic inflammation in aged and diseased rodents with relevance to delirium: A systematic review (2017), Brain Behav Immun 62, 362–381.

Miller-Rhodes

, Kong

, Baht

, Saminathan

, Rodriguiz

, Wetsel

, Gelbard

, Terrando

(2019) The broad spectrum mixed-lineage kinase 3 inhibitor URMC-099 prevents acute microgliosis and cognitive decline in a mouse model of perioperative neurocognitive disorders. J Neuroinflammation 16, 193.

VanDusen

, Li

, Cai

, Hall

, Hiles

, Will Thompson

, Arthur Moseley

, Cooter

, Acker

, Levy

, Ghadimi

, Quiñones

, Devinney

, Chung

, Terrando

, Moretti

, Browndyke

, Mathew

, Berger

; MADCO-PC Investigators (2021) Cerebrospinal fluid proteome changes in older non-cardiac surgical patients with postoperative cognitive dysfunction. J Alzheimers Dis 80, 1281–1297.

Qiu

, Pan

, Luo

, Zhang

, Zhou

, Sun

, Yang

, Ji

(2020) Dysregulation of BDNF/TrkB signaling mediated by NMDAR/Ca2+/calpain might contribute to postoperative cognitive dysfunction in aging mice. J Neuroinflammation 17, 23.

Evered

, Silbert

, Knopman

, Scott

, DeKosky

, Rasmussen

, Oh

, Crosby

, Berger

, Eckenhoff

; Nomenclature Consensus Working Group (2018) Recommendations for the nomenclature of cognitive change associated with anaesthesia and surgery-2018. Anesthesiology 129, 872–879.

Fischer

, Wolf

, Scheurich

, Fellgiebel

(2014) Association of structural global brain network properties with intelligence in normal aging. PLoS One 9, e86258.

10.

Fritz

, King

, Ben Abdallah

, Lin

, Mickle

, Budelier

, Oberhaus

, Park

, Maybrier

, Wildes

(2020) Preoperative cognitive abnormality, intraoperative electroencephalogram suppression, and postoperative delirium: A mediation analysis. Anesthesiology 132, 1458–1468.

11.

Silbert

, Evered

, Scott

, McMahon

, Choong

, Ames

, Maruff

, Jamrozik

(2015) Preexisting cognitive impairment is associated with postoperative cognitive dysfunction after hip joint replacement surgery. Anesthesiology 122, 1224–1234.

12.

Hahn

, Ponce-Alvarez

, Deco

, Aertsen

, Kumar

(2019) Portraits of communication in neuronal networks. Nat Rev Neurosci 20, 117–127.

13.

Voytek

, Knight

(2015) Dynamic network communication as a unifying neural basis for cognition, development, aging, and disease. Biol Psychiatry 77, 1089–1097.

14.

Law

CSH

, Leung

(2018) Long-term potentiation and excitability in the hippocampus are modulated differently by θ rhythm. eNeuro 5, ENEURO.0236-182018.

15.

Rohenkohl

, Bosman

, Fries

(2018) Gamma synchronization between v1 and v4 improves behavioral performance. Neuron 100, 953–963.

16.

Onorato

, Neuenschwander

, Hoy

, Lima

, Rocha

, Broggini

, Uran

, Spyropoulos

, Klon-Lipok

, Womelsdorf

, Fries

, Niell

, Singer

, Vinck

(2020) A distinct class of bursting neurons with strong gamma synchronization and stimulus selectivity in monkey V1. Neuron 105, 180–197.

17.

Makino

, Polygalov

, Bolaños

, Benucci

, McHugh

(2019) Physiological signature of memory age in the prefrontal-hippocampal circuit. Cell Rep 29, 3835–3846.

18.

, Gan

, Jonas

(2014) Interneurons. fast-spiking, parvalbumin GABAergic interneurons: From cellular design to microcircuit function. Science 345, 1255263.

19.

Gillespie

, Jones

, Lin

, Karlsson

, Kay

, Yoon

, Tong

, Nova

, Carr

, Frank

, Huang

(2016) Apolipoprotein E4 causes age-dependent disruption of slow gamma oscillations during hippocampal sharp-wave ripples. Neuron 90, 740–751.

20.

Fernandez

, Salgueiro-Pereira

, Gandin

, Raymond

, Barik

, Goutagny

, Bethus

, Lopes

, Migliore

, Marie

, Pousinha

, Mouska

, Bianchi

, Temido-Ferreira

, Rajão-Saraiva

, Gomes

(2019) The amyloid precursor protein C-terminal domain alters ca1 neuron firing, modifying hippocampus oscillations and impairing spatial memory encoding. Cell Rep 29, 317–331.

21.

Barr

, Radhu

, Guglietti

, Zomorrodi

, Rajji

, Ritvo

, Daskalakis

(2014) Age-related differences in working memory evoked gamma oscillations. Brain Res 1576, 43–51.

22.

Cibelli

, Fidalgo

, Terrando

, Ma

, Monaco

, Feldmann

, Takata

, Lever

, Nanchahal

, Fanselow

(2010) Role of interleukin-1beta in postoperative cognitive dysfunction. Ann Neurol 68, 360–368.

23.

, Li

, Zhang

, Gao

, Zeng

, Mao

, Yang

(2020) Sepsis induced cognitive impairments by disrupting hippocampal parvalbumin interneuron-mediated inhibitory network via a D4-receptor mechanism. Aging (Albany NY) 12, 2471–2484.

24.

Villeda

, Luo

, Mosher

, Zou

, Britschgi

, Bieri

, Stan

, Fainberg

, Ding

, Eggel

, Lucin

, Czirr

, Park

, Couillard-Després

, Aigner

, Li

, Peskind

, Kaye

, Quinn

, Galasko

, Xie

, Rando

, Wyss-Coray

(2011) The ageing systemic milieu negatively regulates neurogenesis and cognitive function. Nature 477, 90–94.

25.

Trollor

, Smith

, Agars

, Kuan

, Baune

, Campbell

, Samaras

, Crawford

, Lux

, Kochan

(2012) The association between systemic inflammation and cognitive performance in the elderly: The Sydney Memory and Ageing Study. Age (Dordr) 34, 1295–1308.

26.

Etminan

, Gill

, Samii

(2003) Effect of non-steroidal anti-inflammatory drugs on risk of Alzheimer’s disease: Systematic review and meta-analysis of observational studies. BMJ 327, 128.

27.

Linnartz-Gerlach

, Bodea

, Klaus

, Ginolhac

, Halder

, Sinkkonen

, Walter

, Colonna

, Neumann

(2019) TREM2 triggers microglial density and age-related neuronal loss. Glia 67, 539–550.

28.

Tanaka

, Cortese

, Barrientos

, Maier

, Patterson

(2018) Aging and an immune challenge interact to produce prolonged, but not permanent, reductions in hippocampal L-LTP and mBDNF in a rodent model with features of delirium. eNeuro 5, ENEURO.0009-18.2018.

29.

Barrientos

, Hein

, Frank

, Watkins

, Maier

(2012) Intracisternal interleukin-1 receptor antagonist prevents postoperative cognitive decline and neuroinflammatory response in aged rats. J Neurosci 32, 14641–14648.

30.

Gonzalez-Burgos

, Cho

, Lewis

(2015) Alterations in cortical network oscillations and parvalbumin neurons in schizophrenia. Biol Psychiatry 77, 1031–1040.

31.

, Wang

, Tan

, Wang

, Lu

, Chen

, Zhang

, Gao

, Liu

, Yin

, Liu

, Yang

, Hu

, Xu

, Wang

, Liu

, Wang

(2017) Correcting miR92a-vGAT-mediated GABAergic dysfunctions rescues human Tau-induced anxiety in mice. Mol Ther 25, 140–152.

32.

Kann

, Papageorgiou

, Draguhn

(2014) Highly energized inhibitory interneurons are a central element for information processing in cortical networks. J Cereb Blood Flow Metab 34, 1270–1282.

33.

Cattaud

, Bezzina

, Rey

, Lejards

, Dahan

, Verret

(2018) Early disruption of parvalbumin expression and perineuronal nets in the hippocampus of the Tg2576 mouse model of Alzheimer’s disease can be rescued by enriched environment. Neurobiol Aging 72, 147–158.

34.

Qiu

, Luo

, Zhang

, Shi

, Li

, Wu

, Chen

, Ji

, Yang

(2016) Nox-2-mediated phenotype loss of hippocampal parvalbumin interneurons might contribute to postoperative cognitive decline in aging mice. Front Aging Neurosci 8, 234.

35.

Stanley

, Fadel

, Mott

(2012) Interneuron loss reduces dendritic inhibition and GABA release in hippocampus of aged rats. Neurobiol Aging 33, 431.e1–13.

36.

Dugan

, Ali

, Shekhtman

, Roberts

, Lucero

, Quick

, Behrens

(2009) IL-6 mediated degeneration of forebrain GABAergic interneurons and cognitive impairment in aged mice through activation of neuronal NADPH oxidase. PLoS One 4, e5518.

37.

Lee

, Hwang

, Yoo

, Choi

, Park

, Lee

, Won

(2008) Parvalbumin immunoreactivity and protein level are altered in the gerbil hippocampus during normal aging. Neurochem Res 33, 2222–2228.

38.

Scott

, Feng

, Kiss

, Needle

, Atchison

, Kawabe

, Milici

, Hajós-Korcsok

, Riddell

, Hajós

(2012) Age-dependent disruption in hippocampal θ oscillation in amyloid-β overproducing transgenic mice. Neurobiol Aging 33, 1481.

39.

Insel

, Patron

, Hoang

, Nematollahi

, Schimanski

, Lipa

, Barnes

(2012) Reduced gamma frequency in the medial frontal cortex of aged rats during behavior and rest: Implications for age-related behavioral slowing. J Neurosci 32, 16331–16344.