Neuroprotective effect of ketamine against TNF‐α‐induced necroptosis in hippocampal neurons

Abstract Tumour necrosis factor‐α (TNF‐α), a crucial cytokine, has various homeostatic and pathogenic bioactivities. The aim of this study was to assess the neuroprotective effect of ketamine against TNF‐α‐induced motor dysfunction and neuronal necroptosis in male C57BL/6J mice in vivo and HT‐22 cell lines in vitro. The behavioural testing results of the present study indicate that ketamine ameliorated TNF‐α‐induced neurological dysfunction. Moreover, immunohistochemical staining results showed that TNF‐α‐induced brain dysfunction was caused by necroptosis and microglial activation, which could be attenuated by ketamine pre‐treatment inhibiting reactive oxygen species production and mixed lineage kinase domain‐like phosphorylation in hippocampal neurons. Therefore, we concluded that ketamine may have neuroprotective effects as a potent inhibitor of necroptosis, which provides a new theoretical and experimental basis for the application of ketamine in TNF‐α‐induced necroptosis‐associated diseases.


| INTRODUC TI ON
Systemic inflammatory response syndrome (SIRS) is caused by the activation of the innate immune system and results in the stimulation of excessive inflammatory responses, and the production and secretion of pro-inflammatory cytokines, such as tumour necrosis factorα (TNFα), and reactive oxygen intermediates. 1,2 In addition, the disbalanced and dysregulated inflammatory response also causes sepsis, which may lead to life-threatening organ dysfunction. 3 Statistically, sepsis developing from SIRS affects ~18 million people worldwide; it is a common cause of acute and severe diseases, and sepsis-associated mortality rates range from 35% to 55%. Survivors often have a poor prognosis, which seriously affects the quality of life. 4 Evidence from previous studies indicates that SIRS can lead to cognitive impairment, physical disability or even sepsis-associated encephalopathy 5 ; however, the underlying pathogenic mechanisms remain to be fully elucidated.
TNFα is as a pleiotropic factor that plays both homeostatic and pathophysiological roles in the central nervous system (CNS), 6 and is mainly generated by activated microglia and astrocytes in response to various stimuli related to infection or injury. For example, it has been reported that necroptosis could be activated in the mouse hippocampus by intracerebroventricular injection of TNFα. 7 The TNF receptor family-mediated necroptosis signalling pathway requires the activation of receptor-interacting protein kinase 1 (RIPK1), which subsequently recruits and activates the kinase receptor interacting serine/threonine kinase 3 (RIPK3). 8,9 Activated RIPK3 phosphorylates its substrate, mixed lineage kinase domainlike protein (MLKL), which can oligomerize and translocate from the cytosol to the membranes to lead to membrane disintegration, resulting in necrosis. 10,11 The aim of this study was to examine whether necroptosis of hippocampal neurons is induced in an in vivo experimental model of SIRS, [12][13][14] created by injecting TNFα intravenously.
As a non-competitive glutamatergic N-methyl-D-aspartate receptor antagonist, ketamine has been extensively used as a clinical anaesthetic and analgesic. 15 Recent studies have also demonstrated that ketamine exerts rapid antidepressant, 16 anti-inflammatory and immunomodulatory effects. 17,18 In addition, evidence shows that ketamine may also reverse synaptic deficits and induce synaptogenesis, exhibiting neuroprotective effects. 19 In the present study, the effects of ketamine on TNFα-induced necroptosis were examined using cultured cells and a mouse model of TNFα-induced SIRS. We also evaluated the effects of ketamine on the long-term physical functions of post-TNFα mice, hippocampal damage, neuronal loss, and oxidative stress using the open field test, Nissl staining, immunofluorescence, flow cytometry and western blotting. Furthermore, the present study addressed the possible contribution of the clinical transformation of ketamine to the pathological mechanism in the SIRS model.

| Animals
Adult male C57BL/6J wild-type mice (18-25 g; Shanghai SLAC Laboratory Animal Co, Ltd.) were bred in specific pathogen-free conditions, and housed in air-conditioned, temperature-controlled rooms with a 12 hours/12h light/dark cycle (lights on, 08:00 am), 22-25°C ambient temperature, and ad libitum access to food and water in the Laboratory Animal Center of Xiamen University. Prior to experimentation, animals were allowed to habituate to the new housing environment for 7 days. All procedures and animal use were approved by the Animal Ethics Committee of Xiamen University (Approval No. XMULAC20190054). Every effort was made to minimize stress to the animals.

| Open field test
The open field apparatus consisted of a square arena (50 × 50 cm) and 50 cm high walls made of grey polyvinyl chloride plastic. On the day of the test, mice were transported to the testing room and left in their home cages for 1 hour prior to testing. At the initiation of each session, a mouse was placed in a particular corner of the arena and allowed to explore for 5 minutes. The apparatus was cleaned with 70% ethanol prior to testing each animal. Time was subsequently recorded, and each mouse was allowed to explore the testing area for 10 minutes. The statistical data were recorded and analysed using the Noldus EthoVision XT system (Ugo Basile SLE).

| Nissl staining
The mice were sacrificed under isoflurane anaesthesia 48 hours after tail-vein injection of TNFα, and then perfused through the ascending aorta with 100 mL of normal saline followed by 100 mL of 4% (w/v) paraformaldehyde in 0.1 M PBS (pH 7.4). The brain of each mouse was dissected and removed from the skull, and resected-brain tissue was fixed in 4% paraformaldehyde for 24 hours at 4°C. Tissues were subsequently stored overnight in 30% sucrose phosphate buffer until the tissue sank to the bottom of the solution. Tissue sections (8 μm) were cut in the coronal plane using a freezing microtome (CM19500; Leica Microsystems, Inc) and mounted on gelatine-coated slides. The sections were then stained with 0.1% cresyl violet solution (Sigma-Aldrich; Merck KGaA) at 37°C for 30 minutes. The sections were subsequently rinsed in distilled water, rehydrated using a descending alcohol series, and checked microscopically for optimal results. Tissues were dehydrated in 100% ethanol, washed in xylene and finally scanned using an Olympus BX53 Scanner (Olympus Corporation) at a magnification of ×20/×4.

| Immunofluorescence
Brain tissue containing the hippocampus was embedded in optimal cutting temperature compound and then cut into thick coronal sec-
Recombinant TNFα was purified as described previously. 8

| Reactive oxygen species (ROS) detection
Reactive oxygen species levels were measured using a fluorometric intracellular ROS kit (cat. no., MAK143; Sigma-Aldrich; Merck KGaA), according to the manufacturer's protocols. Stained cells were viewed using a fluorescence microscope. Cells were treated with TNFα/z-VAD or TNFα/z-VAD/ketamine for 3 hours, and fluorescence was measured using a CytoFLEXS flow cytometer (Beckman Coulter, Inc). no., ab133411; Abcam). The membranes were blocked using 5% skim milk for 40 minutes to reduce non-specific binding and then incubated with primary antibodies (dilution, 1:1,000) at 4°C overnight. Subsequent to being incubated with mice (cat. no., S0100;

| Statistical analysis
Independent experiments were performed in duplicate or triplicate, and data were analysed using GraphPad Prism 7.0 (GraphPad Software, Inc). Data are presented as mean ± standard error of the mean unless indicated otherwise. One-way analysis of variance and the Bonferroni multiple comparison post hoc test were used. P <.05 was considered to indicate a statistically significant difference.

| Result 1. TNFα-induced motor dysfunction is attenuated by pre-treatment with ketamine
Tumour necrosis factorα is a crucial mediator of neuroinflammation, and elevated levels of TNFα are associated with various neurodegenerative conditions and contribute to neurotoxicity. 20 As TNFα-induced SIRS represents an acute model, mimicking a cytokine storm and inducing tissue damage only a few hours after TNFα injection, tail-vein injection with a non-lethal dose of TNFα was performed ( Figure S1)

| Result 2. effect of ketamine on TNFα-induced necroptosis in hippocampal neurons in vivo
To further explore whether TNFα-induced motor dysfunction was associated with brain injury, Nissl staining of neurons was performed to evaluate neurological dysfunction and TNFα-induced neuronal damage. It was observed that administration of TNFα caused a reduction in neuronal density in the hippocampus, particularly in the carbonic anhydrase 3 region. At the same time, pre-treatment with ketamine prevented the TNFα-induced loss of hippocampal neuron number and density in vivo ( Figure 3A). We also demonstrated that this loss was caused by necroptosis, and immunofluorescence revealed that ketamine pre-treatment could significantly increase the number of NeuN-positive neurons in the hippocampal region ( Figure 3B,C,D). Meanwhile, p-MLKL and p-RIP3 levels were markedly decreased in the hippocampus after ketamine pre-treatment ( Figure 3E,F). These results clearly indicate that ketamine may have a novel function in SIRS as a potent inhibitor of necroptosis.

| Result 3. TNFα-induced microglial activation in the hippocampus is attenuated by ketamine pretreatment
The effect of a single systemic challenge with TNFα on C57BL/6J mice was examined using the microglial activation markers Iba-1 and CD68. Confocal microscopy results showed an increase in Iba-1-CD68 double-positive cells after TNFα administration, which was attenuated by ketamine pre-treatment ( Figure 4A,B). Overall, ketamine may inhibit the activation of microglia in TNFα-treated mice, which would contribute to its anti-inflammatory potential.

| Result 4. Ketamine improved the survival of HT-22 hippocampal neuronal cells
The subsequent experiments were designed to elucidate the modulatory effects of ketamine on the necroptosis signalling cascade. HT-22 hippocampal cells have been reported to be sensitive to TNFα only upon caspase blockage and subsequently undergo necroptosis. 7 In this study, we first evaluated the effects of ketamine against TNFα (10 ng/mL)/z-VAD (20 μmol/L) administration on cell viability by measuring ATP levels in HT-22 cells. The results showed that TNFα/z-VAD administration significantly reduced cell viability, and ketamine markedly reduced TNFα/z-VAD-induced cell toxicity in HT-22 cells in a dose-and time-dependent manner ( Figure 5A,B).
Thereafter, 500 μg/mL of ketamine was selected as a major dose to explore its effects on TNFα-induced neurotoxicity in HT-22 cells in vitro. After 4 hours, analysis of propidium iodide-positive HT-22 cells treated with TNFα/z-VAD/ketamine (500 μg/mL) indicated that ketamine prevented these cells from losing membrane permeability and undergoing necroptosis ( Figure 5C).

F I G U R E 2
Ketamine alleviated motor dysfunction caused by TNF α-induced SIRS. (A) The trajectory, (B-D) the total motion distance and the average motion speed in the open field test after 3, 14 and 28 d following tale-vein TNFα injection and pre-treatment with ketamine. All data were shown as mean ± s.e.m. n = 7/ group, * significantly different from the normal group; # significantly different from the TNFα group. *P <.05 and #P <.05

| Result 5. Treatment with ketamine exerted neuroprotective effects by inhibiting ROS accumulation and suppressing TNFα-induced necroptosis of HT-22 hippocampal neuronal cells
In vitro ROS assays revealed that after 4 hours, there was enhanced expression of ROS in TNFα/z-VAD-treated HT-22 cells compared with that in the control group ( Figure 6A). In addition, flow cytometry results indicated that ROS levels were significantly downregulated in the TNFα/z-VAD/ketamine-treated cells ( Figure 6B,C).
Finally, to assess the effects of TNFα on the expression of RIP1 and p-MLKL, western blotting analysis was performed, which revealed that p-MLKL expression was significantly upregulated in TNFα/z-VAD-treated cells, and significantly downregulated in TNF-α/z-VAD/ketamine-treated cells ( Figure 6D,E). Overall, these results support the hypothesis that ketamine suppressed TNFα-induced necroptosis of HT-22 hippocampal neuronal cells by inhibiting ROS accumulation and MLKL phosphorylation.

| D ISCUSS I ON
The present study revealed that surviving mice with TNFα-induced SIRS had motor function decline, and the open field test data indicated that these mice had problems related to anxiety and exploration. However, based solely on the aforementioned observations, it could not be discerned whether this phenomenon was a result of decreased activity levels caused by dyskinesia. Furthermore, TNFα-induced motor dysfunction was attenuated by pre-treatment with ketamine. We also demonstrated that TNFα-induced brain injury led to neuronal necroptosis and microglial activation in C57BL/6J mice. Collectively, these data indicate that the neuroprotective Systemic inflammatory response syndrome, an over-reactive immuno-inflammatory response, represents a significant disease burden and is associated with long-term physical, cognitive and psychosocial morbidity. 21 Previous studies suggested that a high intravenous dose of TNFα (>10 μg) promoted mouse death within 24-36 hours, 22,23 while our data suggested that a sub-lethal dose (5-10 μg) dramatically induced long-lasting sterile pathological SIRSlike effects, with motion-related dysfunction and susceptible tissue damage, even in the brain. With the impaired structure and function of the blood-brain barrier induced by intravenous TNFα administration, 24 peripheral immuno-inflammatory dysfunction would ignite intense central neuroinflammation in a TNFα-dependent and/or independent manner. Necroptosis, or caspase-independent programmed cell death, is known to be involved in various pathological conditions, including TNFα-induced peripheral and central inflammatory processes, both in vivo and in vitro. 25 Recent studies identified that the activation of RIPK1, RIPK3 and MLKL is involved in necroptosis, 8,26 and provided evidence of the signalling events of TNFα-initiated neurotoxicity being mediated by RIPK1-RIPK3-MLKL both in the mouse hippocampus after intracerebroventricular injection of TNFα and in HT-22 hippocampal neuronal cells with TNFα incubation. 7 Considering that some studies have reported that RIPK3 has a pro-inflammatory effect independent of its role in necroptosis, the pseudokinase MLKL is currently regarded as the sole and prime effector of necroptosis, which terminates in the rupture of the plasma membrane and the leakage of intracellular contents from apoptotic cells. During necroptosis, MLKL is a functional substrate of RIPK3. Upon phosphorylation by RIPK3, MLKL forms oligomers and translocates to the plasma membrane. 27 Our data from the present study strongly support the notion that the TNFα- Classically, necroptotic cell death is known to be characterized by disrupted plasma membrane; however, the downstream events leading to membrane collapse are far from being clarified. All data were shown as mean ± s.e.m. n = 7/group, *significantly different from the normal group; # significantly different from the TNFα group. ***P <.001 and ###P <.001 ROS production and accumulation have been suggested to be required for necroptosis in cells such as L929, the human 5-8F NPC cell line and HT-29 human colon cancer cells. [28][29][30] In the present study, we observed increased ROS production and accumulation in TNFα-induced necroptotic HT-22 mouse hippocampal neurons; however, further studies are required to identify the relationship between ROS and TNFα-induced necroptosis as well as to clarify crucial downstream events required for causing this necroptosis of hippocampal neurons.
Ketamine is a traditional narcotic analgesic and psychotomimetic drug with abuse potential in medical practice. It was considered a notable and attractive "drug of the year" when a rapid and sustained anti-depressant profile, with selective rescue of eliminated spines and restoration of coordinated activity in multicellular ensembles, was revealed. 31,32 However, a growing body of evidence indicates that a sub-anaesthetic dose of ketamine (5-30 mg/kg) exerts immuno-inflammatory modulation in sepsis, ischaemia-reperfusion, and a burn injury rodent models. 33,34 Interestingly, our data challenge the generally held view of the role of ketamine as an anti-inflammatory agent when used both clinically 35-37 and experimentally. [38][39][40] Our data showed that ketamine alleviated TNFα-induced motor dysfunction in a dose-dependent manner that a dose of 20 mg/kg could significantly improve the activity and speed of mice ( Figure S2). Meanwhile, we also demonstrated that the protection provided by ketamine was by inhibiting the expression of p-MLKL and p-RIPK3. It is speculated that mitochondrial ROS generation can result in necroptosis, but is bypassed, activating the necroptotic pathway downstream at the RIPK3 or MLKL expression level. 41 In this study, it was identified that ketamine alleviated TNFα-induced necroptosis of hippocampal neurons both in vitro and in vivo, which were indicated to be associated with the inhibition of MLKL phosphorylation and ROS levels. However, the underlying mechanism by which activated MLKL kills cells and the role of ketamine remains unclear, and further studies are required to clarify the process of necroptosis in hippocampal neurons. In conclusion, the results presented in this All data were shown as mean ± s.e.m. *Significantly different from the normal group; #significantly different from the TNFα/z-VAD group. ***P <.001, #P <.05 and ###P <.001. All experiments were repeated three times with similar results the injured brain. 43,44 Innate immune responses and phagocytosis represent a portion of the microglial functional repertoire in terms of the expression of numerous receptors, cell surface molecules, and proteins that enable bidirectional interactions with other cell types in the brain. 28,45 Iba-1, which is expressed exclusively in microglia/ macrophages, 46 and the CD68 marker are most commonly used for discerning macrophages immunohistochemically. Using Iba-1 and CD68 double-labelling, this study was able to confirm the physiological state of microglia based on morphology and immunoreactivity. 47 Immunohistochemical analysis of Iba-1 and CD68 revealed that microglia were activated in the in vivo TNFα-induced SIRS model; moreover, exacerbated neuronal dysfunction was verified and found to be attenuated by ketamine pre-treatment. Further investigation is required regarding microglial involvement in the regulation of neurons and the role of ketamine in this process.
Overall, this study provided experimental evidence that necroptosis of hippocampal neurons may be induced by TNFα both in vivo and in vitro, and could be attenuated by ketamine via inhibition of ROS production and MLKL phosphorylation.

CO N FLI C T O F I NTE R E S T
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
All datasets generated for this study are included in the manuscript.