Exp Neurobiol 2018; 27(6): 564-573
Published online December 12, 2018
© The Korean Society for Brain and Neural Sciences
John Man Tak Chu2,†, Wei Xiong1,2,†, Ke Gang Linghu1, Yan Liu2, Yan Zhang2, Guan Ding Zhao1, Michael G. Irwin2, Gordon Tin Chun Wong2*, and Hua Yu1,3,4*
1Institute of Chinese Medical Sciences, State Key Laboratory of Quality Research in Chinese Medicine, University of Macau, Macao SAR 999078, China.
2Department of Anaesthesiology, LKS Faculty of Medicine, The University of Hong Kong, Pokfulam 999077, Hong Kong, China.
3HKBU Shenzhen Research Center, Shenzhen 518000, Guangdong, China.
4School of Chinese Medicine, Hong Kong Baptist University, Kowloon Tong 999077, Hong Kong, China.
Correspondence to: *To whom correspondence should be addressed.
Hua Yu, TEL: 853-8822 8540, FAX: 853-28841358
Gordon Tin Chun Wong, TEL: 852-22554527, FAX: 852-28551654
†These authors share equal contributions in this manuscript.
A proportion of patients experience acute or even prolonged cognitive impairment after surgery, a condition known as postoperative cognitive dysfunction (POCD). It is characterized by impairment in different cognitive domains and neuroinflammation has been implicated as one of the inciting factors as strategies targeting inflammation tend to improve cognitive performance.
Postoperative cognitive dysfunction (POCD) is a clinical condition that is characterized by impairment in multiple cognitive domains including memory, concentration, language comprehension and learning difficulties . Although acute and mild cognitive decline commonly occurs after surgery, some patients develop more severe and enduring forms of cognitive dysfunction that maybe associated with increased mortality . The underlying mechanism of POCD is still unclear but postoperative neuroinflammation has been shown to be involved with the development of this condition . Previous studies have demonstrated that both surgical trauma and anesthesia could initiate systemic and neuroinflammation in postoperative models . For instance, the general anesthetic agent isoflurane up-regulates pro-inflammatory cytokines brain of mice . Macrophages penetrate the brain and trigger glia activation and up-regulates inflammatory cytokines in the hippocampus after surgery [5,6]. This was associated with cognitive impairment but the degree of dysfunction was reduced by inhibiting macrophages activation and neuroinflammation in the brain . Neuroinflammation also plays a critical role in synaptic dysfunction and tauopathy, in which tau hyperphosphorylation is observed under neuroinflammatory states . Abnormal tau phosphorylation and accumulation are shown to interfere with synaptic function and results in cognitive impairment . Collectively, these data highlight the pivotal role of neuroinflammation in the pathogenesis of POCD and the therapeutic potential of anti-inflammatory strategies to minimize POCD by inhibiting inflammation. These strategies include pharmacological pretreatment that renders an organism more resistant to a subsequent significant insult, a concept that is similar to the practice of Traditional Chinese Medicine (TCM) that emphasizes the “preventive measure”.
Siegesbeckiae Herba (SH, also called Xixiancao in Chinese) is a traditional Chinese medicine (TCM) commonly used for eliminating symptoms including wind-damp, limbs weakness and detoxification as first recorded in the
3-month-old C57BL6/N mice were obtained from The University of Hong Kong. The handling of animal and all procedures were conducted in accordance with National Institutes of Health guide for the care and use of Laboratory animals and Animals (Control of Experiments) Ordinance, Hong Kong, China. The use of animals was approved by the Department of Health, Hong Kong and Committee on the Use of Live Animals in Teaching and Research, The University of Hong Kong. All efforts were made to minimize animal numbers and suffering. In the current report, mice were divided into 6 groups: Sham control (Ctrl), surgery (Lap), low dose drug control (0.75 g/kg/day), high dose drug control (1.5 g/kg/day), low dose drug with surgery (0.75 g/kg/day plus Lap) and high dose drug with surgery (1.5 g/kg/day plus Lap).
The herb of
For use in the experiments, a powdered SO (100 g) was refluxextracted twice with 10-fold volume of 50% ethanol (v/w) for 1 hr for each. The combined extract was cooled, filtered, and then concentrated under reduced pressure to remove the ethanol. Subsequently, the concentrated extract was lyophilized with a Virtis Freeze Dryer (The Virtis Company, New York, USA). The powdered
Quantifications of two major active compounds (kirenol and darutoside) in the
To examine if
Hippocampal dependent object recognition memory was determined by the novel object recognition (NOR) test as described before but with modifications . Briefly, on postoperative day (POD) 1, mice were put in an open field box (50×50 cm) with no objects for 10 mins for habituation. On POD 2, mice were placed in the same box with 2 identical objects for 10 mins for familiarizing the object. Twenty-four hours after familiarization on POD 3, one of the two objects was replaced with a novel object. The mice were allowed to stay in the box for 10 mins and the behavior of the animals was video recorded. Schematic diagram is shown in Fig. 2B. Object exploration was scored by the amount of time with the nose pointed towards and located within 2 cm of the object. The discrimination index was calculated by the formula: Tn/Tt, where Tn is the time of exploring novel object while Tt is the total time of exploring novel and familiarized object. After completing the novel object recognition test, the mice were sacrificed by CO2 asphyxiation. 1 ml of blood was extracted by cardiac puncture and serum was isolated after 1,300 g centrifugation for 10 mins. Liver and hippocampal tissues were then dissected out from animals for subsequent real time PCR and Western Blot experiments.
Hepatic and hippocampal mRNA were isolated by RNAiso plus (Takara, Japan). Tissues were first homogenized in RNAiso plus and vigorously mixed with chloroform. After 10000 g centrifugation at 4℃, the upper aqueous layer was isolated. mRNA pellets were precipitated by isopropanol and washed with 75% ethanol. It was then dissolved in diethyl pyrocarbonate (DEPC) treated water. 1µg of mRNA was converted to complementary DNA sequence by reversed transcription using cDNA synthesis kit according to manufactures' protocol (Takara, Japan). Levels of cDNA of different inflammatory cytokines were assessed by real time PCR with respective primers: 1) IL-1β, forward: CCTCCTTGCCTCTGATGG, reverse: AGTGCTGCCTAATGTCCC; 2) IL-6, forward: TTCACAAGTCCGGAGAGGAG, reverse: TCCACGATTTCCCAGAGAAC; 3) IL-8, forward: TGCCGTGACCTCAAGATGTGCC, reverse: CATCCACAAGCGTGCTGTAGGTG; 4) TNF-α, forward: CCCCAGTCTGTATCCTTCT, reverse: ACTGTCCCAGCATCTTGT; 5) GAPDH, forward: ATTCAACGGCACAGTCAA, reverse: CTCGCTCCTGGAAGATGG.
Inflammatory cytokine expression in the serum was evaluated by MILLIPLEX MAP mouse cytokine/chemokine magnetic bead panels (EMD Millipore Corp., Billerica, MA). All primary data points were collect on a Luminex MAGPIX system. Protein samples and detection substrates were incubated on the plate with specific antibodies-conjugated magnetic beads coated on each well. Fluorescence signal detection and analysis were performed according to the manufacturer's protocol. Concentrations of IL-1β, IL-6, MIP-2 (homologues of IL-8) and TNF-α in serum were determined.
Hippocampal tissue was dissected out and homogenized as mentioned above. Proteins were extracted with RIPA lysis buffer (Cellsignal, Danvers, MA) supplemented with protease and phosphatase inhibitors (Roche, Berlin, German). Protein samples were quantified by BCA protein assay. Proteins were resolved by SDS-PAGE gel and transferred to PVDF membrane. After blocking with 2% non-fat milk, membranes were probed overnight at 4℃ with different primary antibodies (JNK, p-JNK, p65 and p-p65 were from Cellsignal, MA; P-tau S396 and β-actin were from Thermo Fisher Scientific, MA; Pan-tau was from DAKO, Japan), followed by respective HRP-conjugated secondary antibodies for 1 hr. Protein bands were visualized by enhanced chemiluminescence (ECL) reagents and signals were captured by X-ray film. The intensities of protein bands were quantified by Image J analysis software.
Comparison between different groups of treatment was analyzed by one-way ANOVA with Turkey post-hoc test. Significant differences were considered between groups when p<0.05.
Kerinol (Fig. 1C) and darutoside (Fig. 1D) are the two most important compounds in
Memory deficits on POD 3 were observed in the surgical group as manifested by a significant reduction of novel object exploration time (Fig. 2). Drug treatment alone did not exert any adverse effects on memory function. In the surgery plus drug groups, although no significant improvement of cognitive deficit was found with the low dose (0.75 g/kg/day) treatment, the high dose (1.5 g/kg/day) treatment significantly improved memory function as reflected by the increase in the discrimination index (Fig. 2, p=0.0265 between Ctrl and Lap; p=0.0376 between Lap and 1.5g/kg/day plus Lap). No significant differences were found between other groups; p>0.05). These data demonstrated that pretreatment with the extract improved postoperative memory function in dose dependent manner.
To investigate if extract pretreatment could attenuate inflammatory responses in postoperative animals, mRNA of pro-inflammatory cytokines in the liver were examined by real time PCR. Significant up-regulation of IL-6 was observed in the laparotomy group while high dose extract pretreatment ameliorated the increased mRNA expression of these pro-inflammatory cytokines (Fig. 3A) (p=0.0117 between Ctrl and Lap; p=0.0127 between Lap and 1.5 g/kg/day plus Lap, non-significance was found between other groups; p>0.05). In addition, cytokine levels in the serum were examined by Milliplex assay. In line with the results from real time PCR, serum concentration of IL-6 (p=0.015 between Ctrl and Lap; p=0.044 between Lap and 1.5 g/kg/day plus Lap, non-significance was found between other groups; p>0.05) and MIP-2, the homologue of IL-8, were significantly increased in surgery group while reduced after high dose extract treatment (Fig. 3B. p=0.0137 between Ctrl and Lap; p=0.0236 between Lap and 1.5 g/kg/day plus Lap, No significant difference was found between other groups; p>0.05) Finally, to examine the neuroinflammatory response, mRNA expression of inflammatory cytokines were examined by real-time PCR. Increase in IL-1β (p=0.0337 between Ctrl and Lap; p=0.0416 between Lap and 1.5 g/kg/day plus Lap, no significant difference between other groups; p>0.05) and IL-6 (p=0.0339 between Ctrl and Lap; p=0.0482 between Lap and 1.5 g/kg/day plus Lap, no significant difference between other groups; p>0.05) were observed in the hippocampi of the surgical group. These increases were attenuated by high dose extract (Fig. 4). Overall, these results demonstrated that the extract reduced inflammation in both peripherally and centrally.
After observing the inflammatory response in the hippocampus, we further examined if extract pretreatment inhibited neuroinflammation through modulating pro-inflammatory pathways (JNK and NF-κB) by measuring phosphorylation of JNK and p65 subunit using Western blot. In postsurgical animals, significant up-regulation of phosphorylated JNK (p=0.0226 between Ctrl and Lap; p=0.0403 between Lap and 1.5 g/kg/day plus Lap, non-significance was found between other groups; p>0.05) and p65 (p=0.0182 between Ctrl and Lap; p=0.021 between Lap and 1.5 g/kg/day plus Lap, non-significance was found between other groups; p>0.05) was observed in the hippocampus, whereas extract pretreatment reduced the phosphorylation of both molecules in a dose dependent manner (Fig. 5). These results imply that extract pretreatment may attenuate neuroinflammatory responses through suppressing JNK and p65 activities and subsequent nucleus translocation in the hippocampus.
Neuroinflammatory response is closely related to tau phosphorylation, which leads to the destabilization of tau from microtubules and subsequent memory deficits . To examine if surgery induces tau phosphorylation in the hippocampus and whether extract pretreatment impedes this process, protein samples were subjected to Western blotting and examined by phosphorylated tau antibody at Serine 396 epitome. Significant up-regulation of tau phosphorylation was found in the hippocampus after surgery, while extract pretreatment reduced tau phosphorylation (Fig. 6) (p=0.0159 between Ctrl and Lap; p=0.0268 between Lap and 1.5 g/kg/day plus Lap, non-significance was found between other groups; p>0.05). These results indicated that extract may provide a neuroprotective effect to postoperative animals through reducing neuroinflammation and tau phosphorylation.
Substantial evidence suggest that surgery induces cognitive dysfunction in patients in whom a continuous inflammatory response is present . Apart from various physical barriers, perioperative inflammation was once considered as the second line of defense mechanism to protect human from pathogens after surgery. Nevertheless, it is increasingly recognized that several postoperative complications are related to inflammation including atrial fibrillation , hemorrhage  and cognitive dysfunction . Various therapeutic strategies targeting inflammation throughout the surgical period have been suggested to combat these complications. In agreement with previous reports, our results suggest that surgery induces systemic and neuroinflammatory response in postoperative animals . Furthermore, activation of pro-inflammatory pathways was observed, which was associated with an increase in tau phosphorylation and cognitive impairment, thus showing that treatment using
It is well established that peripheral inflammation could trigger a neuroinflammatory response in the brain . Chronic neuroinflammation has been shown to accompany deterioration in cognitive function. For instance, inflammation cytokines induce the hyperactivation of astrcotyes , which may trigger the release of other cytokines or gliotransmitter such as GABA that affect neuronal activity and plasticity in the AD model . Inhibiting neuroinflammation in turn resulted in amelioration of astrogliosis and improved subsequent cognitive impairment in AD . On the other hand, therapeutic strategies targeting systemic inflammation may improve neuroinflammation and ameliorate neuronal damage, synaptic dysfunction and cognitive performance in different models, including POCD . Traditional Chinese medicine has been used over many centuries and mounting evidence has demonstrated anti-inflammatory properties.
In view of these anti-inflammatory properties, we examined whether extract from
To further elucidate how the extract could reduce the production of pro-inflammatory cytokines in brain, we investigated if the activation of pro-inflammatory intracellular signaling was modulated. JNK and NF-κB are major pathways involve in inflammation [24,25] and they respond to external stimulus such as reactive oxygen species, bacteria or virus. In neurodegeneration, activation of JNK and NF-κB are observed in the brains of Alzheimer's Disease models [26,27]. Both molecules are phosphorylated and translocated to the nucleus, binding to the promoter and enhancing the transcriptional activities of the genes, thus increasing the synthesis of pro-inflammatory cytokine [24,25]. Over-activation of JNK and NF-κB in the brain are shown to worsen inflammatory responses while inhibition of these molecules ameliorates neuroinflammation and subsequent cognitive dysfunction. Our results demonstrated that surgery leads to the phosphorylation of both JNK and NF-κB in the brain, which was in line with the increase in pro-inflammatory cytokine gene expression. In contrast, extract pretreatment significantly reduced the phosphorylation of both JNK and NF-κB (Fig. 4), thus further confirming that the extract reduces the neuroinflammatory response by impeding the activation of pro-inflammatory pathways in the hippocampus.
One of the possible explanations for the anti-inflammation effect of
We further asked how the attenuation of neuroinflammation by the extract may modulate tau phosphorylation in the hippocampus. Tau is an important microtubule associated protein in which abnormal phosphorylation and aggregation of tau protein is a pathological hallmark in cognitive dysfunction model . Upon phosphorylation, tau was shown to be dissociated from microtubule which undergo aggregation and oligomerization and exert neurotoxicity or inhibit the function of synapse, leading to synaptic and neuronal dysfunction. It has been shown that inflammatory cytokines such as IL-6 could induce tau phosphorylation which deteriorate neuronal degeneration . In our model, tau phosphorylation was up-regulated in the hippocampus after surgery. With extract pretreatment , reduction of phosphorylated tau was observed compared with surgery alone group (Fig. 5). These results imply that
Despite the promising effect of
To conclude, this study presented