Propofol can Protect Against the Impairment of Learning-memory Induced by Electroconvulsive Shock via Tau Protein Hyperphosphorylation in Depressed Rats△

Wan-fu Liuand Chao Liu*

1Department of Anesthesiology, Fuzhou General Hospital of the People's Liberation Army Nanjing Military Region, Fujian 350025, China

2Department of Anesthesiology, Tianjin Chest Hospital, Tianjin 300222, China

ORIGINAL ARTICLE

Propofol can Protect Against the Impairment of Learning-memory Induced by Electroconvulsive Shock via Tau Protein Hyperphosphorylation in Depressed Rats△

Wan-fu Liu1and Chao Liu2*

1Department of Anesthesiology, Fuzhou General Hospital of the People's Liberation Army Nanjing Military Region, Fujian 350025, China

2Department of Anesthesiology, Tianjin Chest Hospital, Tianjin 300222, China

propofol; Tau protein; learning-memory abilities; glutamate; electroconvulsive therapy

Objective To explore the possible neurophysiologic mechanisms of propofol and N-methyl-D-aspartate (NMDA) receptor antagonist against learning-memory impairment of depressed rats without olfactory bulbs.

Methods Models of depressed rats without olfactory bulbs were established. For the factorial design in analysis of variance, two intervention factors were included： electroconvulsive shock groups (with and without a course of electroconvulsive shock) and drug intervention groups [intraperotoneal (ip) injection of saline, NMDA receptor antagonist MK-801 and propofol. A total of 60 adult depressed rats without olfactory bulbs were randomly divided into 6 experimental groups (n=10 per group)： ip injection of 5 ml saline; ip injection of 5 ml of 10 mg/kg MK-801; ip injection of 5 ml of 10 mg/kg MK-801 and a course of electroconvulsive shock; ip injection of 5 ml of 200 mg/kg propofol; ip injection of 5 ml of 200 mg/kg propofol and a course of electroconvulsive shock; and ip injection of 5 ml saline and a course of electroconvulsive shock. The learning-memory abilities of the rats was evaluated by the Morris water maze test. The content of glutamic acid in the hippocampus was detected by high-performance liquid chromatography. The expressions of p-AT8Ser202in the hippocampus were determined by Western blot analysis.

Results Propofol, MK-801 or electroconvulsive shock alone induced learning-memory impairment in depressed rats, as proven by extended evasive latency time and shortened space probe time. Glutamic acid content in the hippocampus of depressed rats was significantly up-regulated by electroconvulsive shock and down-regulated by propofol, but MK-801 had no significant effect on glutamic acid content. Levels ofphosphorylated Tau protein p-AT8Ser202in the hippocampus was up-regulated by electroconvulsive shock but was reduced by propofol and MK-801 alone. Propofol prevented learning-memory impairment and reduced glutamic acid content and p-AT8Ser202levels induced by electroconvulsive shock.

Conclusion Electroconvulsive shock might reduce learning-memory impairment caused by protein Tau hyperphosphorylation in depressed rats by down-regulating glutamate content.

Chin Med Sci J 2015; 30(2)：100-107

ELECTROCONVULSIVE seizure over a period of 120 to 180 seconds can lead to cognitive disorders1 because of pathological dysfunction of the glutamic acid (Glu) signal system.2-4 Glu is the main excitatory neurotransmitter that transmits approximately 40% of synapses. It plays a two-way role in learning-memory abilities. Glu receptors (GluR), such as N-methyl-D-aspartic acid (NMDA) and quisqualic acid, participate in the excitatory synaptic transmission of neurons.5 However, if intracerebral Glu is released in large quantities, its rapid excitatory function can lead to depolarisation, causing influxes of Cl-, Na+ and water as well as permeable dissolution of cells. Depolarisation activates membrane potential-dependent GluR, resulting in excessive influx of Ca2+.6, 7 This phenomenon leads to the activation of various enzymes sensitive to Ca2+ and damaged oxygen-derived free radicals and mitochondria, further causing the activation of phosphoinositide loop and damage of cell ultrastructures.8 This series of events lead to neuronal apoptosis or death9 and then learning-memory impairment.

Tau proteins are low-molecular-weight microtubuleassociated proteins.10 They are highly asymmetric phosphorproteins distributed at the frontal and temporal lobes of the brain, axons and dendrons of the hippocampus and neurons in the entorhinal area. Their primary structures have the most prominent feature of amino acid residue repeats in a ‘tandem repeat’ form with the major structure of Pro-Gly-Gly-Gly at carboxyl termini. These repeats participate in the microtubule-associated region of Tau proteins, promote assembly and stabilisation of axon microtubules,11 maintain space between microtubules,12 affect axonal material transport of nerve cells, promote neuronal growth and development, inhibit lipid peroxidation and tubulin aggregation and facilitate learning and memory. Tau protein phosphorylation is one of the major mechanisms that regulate neuronal functions.13 Abnormal phosphorylation of Tau proteins induces misfolding and molecular aggregation,14 which consequently weaken their capability to stabilize microtubules and decrease axonal transfer efficiency, resulting in transmitter transport, storage and release disorders and synapse degeneration or change in the distribution and activity of prion proteins.15These phenomena can also cause neuronal apoptosis or death and lead to learning-memory impairment. GluR antagonists can alleviate stress-induced Tau protein hyperphosphorylation in the hippocampus by inhibiting the excitatory neurotransmission system.16,17

Propofol is an ideal intravenous anaesthetic, and its mechanism of action on the central nervous system is associated with inhibition of Glu and iontropic GluR. It can alleviate electroconvulsive therapy (ECT) induced learningmemory impairment in depressed rats by decreasing the excitotoxicity of the central nervous system. Relevant presumed mechanisms of propofol are as follows: neuronal inhibition to synthesize Glu;18recovery of Glu transfer rate of oxidative stress-damaged neuroglial cells by protecting Na+/H+exchanger activity;19and reduction of membrane potential depolarisation induced by elevated extracellular K+concentration in stress or inhibition of nitric oxide synthase activity by inhibiting the Na+channel and inhibition of Glu release.20Moreover, given that its benzene ring groups act on the phencyclidine binding site of NMDA, propofol has a non-competitive antagonistic action on NMDA receptor (NMDA-R). It shortens the duration of the excitatory postsynaptic potential and reduces neuronal injury caused by the overexcitation of NMDA-R.21Kingston et al22found that propofol exerts its functions directly as an NMDA-R antagonist and partially reverses post-ECT learningmemory impairment.

This study aims to explore the possible neurophysiologic mechanisms of propofol and NMDA-R antagonist against learning-memory impairment of depressed rats without olfactory bulbs.

MATERIALS AND METHODS

Principles for animal experimentation

The experiments in this study were approved by the Hospital Ethics Committees of Tianjin Chest Hospital. All animal experiments were carried out according to the Principle for Treatment of Laboratory Animals issued by the American Medical Association and the Guide for the Care and Use of Laboratory Animals issued by the AmericanSociety of Animal Science and National Institutes of Health. All tests and inspections followed the double-blind principle.

Establishment of depressed rat models

In this study, 24-week-old healthy male Sprague Dawley rats weighing 250 to 300 g were provided by the Department of Laboratory Animal Science, Tianjin Medical University. The rats were kept in a well-ventilated place, free access to food and water, subjected to alternating light/dark cycle of 12 hours and touched for 2 minutes daily for acclimatisation. All rats were raised in separate cages throughout the experiment.

After 1 week of adaptive breeding, models of depressed rats without olfactory bulbs were established as follows.23The rats were anaesthetised with intraperotoneal (ip) injection of 2.75% sodium pentobarbital (55 mg/kg, ml/kg). The skin at the midpoint of two ears was incised to expose the skull, and 2-mm intersections were cut 7 to 8 mm in front of the anterior fontanels and both sides of the median raphes. After drilling two holes 2 mm in diameter using an electric grinder (Shanghai Huien Medical Instrument and Device Co., Ltd., China), the olfactory bulb tissues were removed. The incisions were washed with penicillin solution (200 000 U/ml), and the skin was sutured. The rats were injected with 40 000 U penicillin sodium for three consecutive days.

The rats were touched and weighed daily after the operation. The rats were subjected to the open field test for 2 weeks. A total of 60 rats with total horizontal and vertical scores ranging from 30 to 120 in the open field test were included in the experimental group.

Grouping and intervention methods

For the factorial design in analysis of variance (ANOVA), two intervention factors were included: electroconvulsive shock composed of two levels (with and without a course of ECT) and drugs (ip injection of saline, propofol and MK-801). NMDA-R antagonist, pure 5-methyl-dihydropropylcyclohepten-imine maleate (dizocilpine or MK-801) was purchased from Sigma (USA), and propofol was purchased from AstraZeneca (USA).

A total of 60 adult depressed rats without olfactory bulbs were randomly divided into 6 experimental groups (n=10 per group): (I) ip injection of 5 ml saline; (Ⅱ) ip injection of 5 ml of 10 mg/kg MK-801;20(Ⅲ) ip injection of 5 ml of 10 mg/kg MK-80124and a course of electroconvulsive shock; (Ⅳ) ip injection of 5 ml of 200 mg/kg propofol;4(Ⅴ) ip injection of 5 ml of 200 mg/kg propofol4and a course of electroconvulsive shock; and (Ⅵ) ip injection of 5 ml saline and a course of electroconvulsive shock.

Each group of corresponding drugs was administered by ip injection 15 minutes before each ECT. The electrodes were placed at the bilateral temporal areas of the rats, and a Harvard sine-wave ECT apparatus (NatureGene Co., USA) was employed to provide electrical stimulation with a square wave (a single half sine wave consisted of 20 ms) at a current of 50 mA and a frequency of 50 Hz for 1 second. The occurrence of tonic-clonic convulsion seizure was considered to indicate successful treatment.25ECT was carried out once every 2 days at 9 am for seven times.

Testing of the learning-memory functions of experimental rats by the Morris water maze video analysis system

The Morris water maze test (Academy of Military Medical Sciences, USA) was carried out within 24 hours after all rats received ECT. The Morris water maze was equally divided into four quadrants: I, Ⅱ, Ⅲ and Ⅳ. Before training, the water maze was filled with tap water and added with ink to make the water cloudy. All experiments were carried out at 9 am to 3 pm in a quiet room, with consistent article placement and light state and a water temperature of 24?C ± 1?C. Morris 1.0 software was used to track and record analysis associated data.

Place navigation tests were carried out on days 1 to 6. Briefly, the rats were placed in water by facing the pool wall from quadrants I, Ⅱ, Ⅲ and quadrant Ⅳ in counterclockwise direction and then observed for 120 seconds. Before testing, the platform was placed 2 cm under the water surface in the centre of quadrant I. The escape latency, which is the time during which the rats looked for and climbed the platform, was detected using a camera system. When the rats failed to find the platform within 120 seconds, they were led back to the platform, and the escape latency was recorded as 120 seconds. After testing, the mean escape latency on days 1 to 6 was regarded as the learning result. The shorter the escape latency, the better the learning capacity of the rats.

A space probe trial was carried out on day 7. Briefly, the platform was removed, and the rats were placed in water by facing the pool wall from quadrant Ⅲ farthest from the original platform. The swimming time of the rats in each quadrant within 60 seconds was recorded using a camera system. The swimming time in quadrant I of the original platform (space probe time) was used as the memory performance. The longer the time was, the better the memory capacity of the rats.

Sample collection

As an important region closely related to learning-memory functions in the brain, the hippocampus participates in information acquisition, preservation and extraction, and is the main target area subject to injury by stress; therefore, the hippocampus was selected as the study region in the experiment. The hippocampal tissues of the rats were extracted within 24 hours after the Morris water maze test. The rats were fasted without water deprivation 8 hours before sample collection and then anaesthetised with ip injection of 20% ethyl carbamate (1.5 g/kg). They were quickly killed by decapitation to remove the brain tissues. The blood stains were soaked on ice-cold de-diethyl pyrocarbonate water to separate the bilateral hippocampal tissues.

The hippocampus tissue was frozen in liquid nitrogen overnight at -80?C in an ultra-low-temperature refrigerator for Western blot analysis. After weighing, tissue was added with 1 ml methanol-water centrifugate and then homogenized at a low temperature. A part of the homogenate (10 000 ×g) was centrifuged at 4?C for 15 minutes, and the supernatant was then collected, filtered with a filter membrane and then kept at -80?C to measure Glu content (μg/g).

Determination of hippocampal Glu content of rats by high-performance liquid chromatography (HPLC)26

Apparatus and chromatographic conditions HPLC system (Waters, USA) was equipped with a 600-series pump, a model 2475 fluorescence detector and an Empower chromatographic workstation and 18-ODS chromatographic column (Dima, USA) at 35?C. Mobile phase A was 0.1 mol/L potassium acetate; mobile phase B was methanol undergoing binary gradient elution. Gradient elution procedure was as follows: T (B%), 0, (45%), 1 (65%), 6 (75%) and 20 (45%), where T refers to time and B% refers to the proportion of mobile phase B. The mobile phase was filtered by a 0.45 m microporous filter membrane and subjected to ultrasonic degassing at a flow rate of 1.0 ml/min, with excitation wavelength of 250 nm and emission wavelength of 410 nm. Quantification was made based on the Glu peak area.

Preparation of derivatisation reagent A total of 20 mg orthophthalaldehyde was dissolved in 500 μl methanol for ultrasonic dissolution, added with 500 μl β-mercaptoethanol and 9 ml boric acid buffer solution (pH 10.0) and then stored at 0 to 4?C.

Preparation of amino acid standard solution Standard solution (100 μmol/L) prepared by the Glu standard was diluted before testing. A total of 100 μl standard solution or tissue sample solution was placed in an EP tube and then reacted with 100 μl derivatisation reagent for 2 minutes, with 20 μl of sample injected.

Establishment of the Glu standard curve Glu standard solutions with concentrations of 0.15, 0.30, 0.735, 1.47, 2.94, 3.675 and 5.88 mg/L were prepared and determined after derivatisation. Quantitative analysis was carried out by the external standard method, and the concentrations (X) and peak areas (Y) were subjected to linear regression to obtain a linear equation.

Determination of hippocampal Glu content The homogenate supernatant of hippocampal tissues was unfrozen, added with frozen formic acid (1 mol/L, 2 ml) and then homogenised manually in an ice bath. The homogenate was centrifuged at 8.0×1012×g for 30 minutes at 4?C, and the supernatant was kept at -20?C for subsequent use. The homogenate supernatant (1 ml) of the brain tissues was added with 0.75 ml 4% sodium bicarbonate solution and then centrifuged at 3.4×1012×g for 5 minutes at 4?C. The supernatant was filtered by a 0.45 μm filter membrane and then loaded. Subsequently, 24 μl loaded solution was collected, and 12 μl derivatisation reagent and 960 μl sodium tetraborate buffer solution (pH 9.18) were added to the sample vial and allowed to stand at 20?C for 3 minutes. Sampling and gradient elution were accomplished to determine Glu content.

Determination of p-AT8Ser202by Western blot analysis

The hippocampus of the rats was homogenized, and 0.2 g homogenate was placed in a cell lysis buffer for Western blot analysis. Protein concentration was determined using a BCA protein assay kit and then adjusted for consistency. Equal protein samples were extracted, diluted with 5×sodium dodecyl sulphate (SDS) sample loading buffer solution at 1:1 (V/V) and then boiled at 100?C for 5 minutes. The mixture of the pre-stained protein molecular weight markers was dissolved further in 1×SDS sample loading buffer solution and then boiled at 100?C for 3 minutes. A 15 μl aliquot of the test samples was loaded using glyceraldehyde-3-phosphate dehydrogenase for calibration and then subjected to SDS-polyacrylamide gel electrophoresis (PAGE) until the target molecular weight was achieved. The protein bands were electrically transferred to immun-blot polyvinylidine fluoride membranes by a wet process. Briefly, the membranes was sealed with 50 g/L non-fat milk powder for 3 hours, and then incubated with rabbit anti-human p-AT8Ser202monoclonal antibody (Gene Tex, USA) at 4?C overnight. The IgG (1:200) marked with the corresponding horseradish peroxidase was added and incubated with the membranes at 37?C for2 hours. A DAB kit (Shanghai Bi-yun-tian Biotechnology Co. Ltd., China) was used for colour development, and the Goldisc multimedia image processing system (Chengdu Goldisc UESTC Multimedia Technology Co., Ltd, China) was used for determining the integral absorbance value of the positive bands.

Statistical analysis

Data were expressed as mean±standard deviation (SD). Homogeneity of variances for each sample group was tested using SPSS 19.0 statistical software. Each group was subjected to factorial design and one-way ANOVA to determine the main and interaction effects of each treatment factor. Simple effect of each treatment factor was analyzed by one-way ANOVA, and multiple comparisons were determined by the least significant difference test and Student-Newman-Kuels-q test. Differences were considered statistically significant at the P＜0.05 level.

RESULTS

Learning-memory functions: escape latency and space probe time

As shown in Tables 1, 2, ECT and both drugs (propofol and MK-801) could result in learning-memory impairment ofrats, with prolonged escape latency (ECT: F=148.986, P=0.000; MK-801 and propofol: F=3.809, P=0.030) and shortened space probe time (ECT: F=4.376, P=0.043; MK-801 and propofol: F=17.863, P=0.000). However, their effects presented a negative relationship (escape latency: F=32.870, P=0.000; space probe time: F=98.938, P=0.000). Propofol or MK-801 could alleviate learningmemory impairment induced by electroconvulsive shock.

Table 1. Effects of intervention factors on escape latency of depressed rats without olfactory bulbs§(s, n=10)

Table 2. Effects of intervention factors on space probe time of depressed rats without olfactory bulbs§(s, n=10)

Glu content in rat hippocampus

ECT (F=277.841, P=0.000) significantly elevated the concentration of Glu in the hippocampus and propofol significantly decreased Glu content (F=21.320, P=0.000). MK-801 had no clear effect on the concentration of Glu in the hippocampus. Propofol could decrease elevated Glu content induced by electroconvulsive shock (F=11.091, P=0.000). (Table 3)

Expression levels of phosphorylated Tau protein p-AT8Ser202in the rat hippocampus

ECT increased the expression levels of p-AT8Ser202(F= 350.725, P=0.000), whereas propofol and MK-801 reduced their expression levels (F=73.129, P=0.000). Propofol and MK-801 slowed down the ECT-induced enhancement of p-AT8Ser202expression (F=4.580, P=0.016) (Fig. 1, Table 4).

Table 3. Effects of intervention factors on glutamic acid content in the hippocampus of depressed rats without olfactory bulbs§(μmol/g prot, n=10)

Figure 1. Western blot analysis of p-AT8Ser202expression levels in the hippocampus of depressed rats without olfactory bulbs.I. Intraperotoneal (ip) injection of 5 ml saline; Ⅱ. ip injection of 5 ml of 10 mg/kg MK-801; Ⅲ. ip injection of 5 ml of 10 mg/kg MK-801 and a course of electroconvulsive shock; Ⅳ. ip injection of 5 ml of 200 mg/kg propofol; Ⅴ. ip injection of 5 ml of 200 mg/kg propofol and a course of electroconvulsive shock; Ⅵ. ip injection of 5 ml saline and a course of electroconvulsive shock.

Table 4. Results of quantitative analysis of p-AT8Ser202expression levels in the hippocampus of depressed rats without olfactory bulbs§(n=10)

DISCUSSION

In this experiment, significant post-ECT elevated hippocampal Glu concentration was accompanied with declined learning-memory abilities, as indicated by prolonged escape latency and shortened space probe time. The former represents impaired learning ability, whereas the latter shows declined explicit memory. These findings are in agreement with those of previous studies, in which ECT was found to induce Glu-associated excitotoxicity.

The hippocampus of the limbic system plays an important role in memory. The episodic memory in the explicit memory absolutely depends on the hippocampus.27The hippocampus is not only closely associated with the short-term memory but also with the long-term spatial memory of rats.28The water maze test was employed to determine the spatial memory in the episodic memory. The spatial memory of humans or animals is summarized in the cognitive map stored in the hippocampus.29Hippocampal cells can receive and process spatial information from different sources, enable cognitive map formation orenhance synaptic contact of cell assemblies in the association cortex to form the permanent memory of spatial positions. The results of our study revealed the elevated hippocampal Glu concentration and the enhanced Tau protein hyperphosphorylation might cause obvious impairment of the spatial memory ability of rats. By contrast, propofol partially blocked the excitotoxicity of Glu and further alleviated Tau protein hyperphosphorylation. These results also proved that the hippocampal tissues might have functions in the spatial memory and explicit memory of rats.

Palmio et al30indicated that ECT could not impair the neurons because the post-ECT neuron-specific enolase and S-100B proteins in the serum do not obviously increase. Their findings are inconsistent with the present results. Whether the result is associated with very few cases (10 cases) still needs further verification.

Previous studies indicated that propofol can lower the Glu content of the brain and inhibit neuronal apoptosis by improving Akt activity.31In this study, propofol decreased the post-ECT hippocampal Glu content and improved the post-ECT learning-memory abilities of the rats. This finding is consistent with the study of Xu et al.32In addition, propofol can decrease Tau protein hyperphosphorylation to improve post-ECT learning-memory abilities. Therefore, we presume that propofol can break the signaling pathway of ECT-induced Tau protein phosphorylation enhancement. The NMDA-R antagonist showed no obvious effect on hippocampal Glu content. Therefore, the NMDA-R antagonist may function by blocking the excitability of Glu instead of decreasing the excretion of Glu in the hippocampus.

A comparative analysis of the present and previous results was conducted. In contrast to the present findings, Stover and his colleague33revealed that propofol can increase Glu concentration in the cerebrospinal fluid. This finding may be attributed to the study objects (patients undergoing neurosurgery) and the effect of the interference factor (neurosurgery operation) on the Glu concentration of the nervous system. Previous studies also showed that propofol cannot protect the hippocampal slices from injury. At high doses, propofol can aggravate the injury and even increase the release of Glu. This finding is in disagreement with our experimental result. The in vitro brain slices used in the experiment may not be appropriate to simulate the in vivo environment. In contrast to the present results, Pesi? et al34found that propofol induces cortical neuron death. This inconsistency in results may be attributed to the fact that the rats used in the previous study are 7-day-old neonates; thus, their nervous system is still immature and more sensitive to drugs than adult rats.

In conclusion, electroconvulsive shock might reduce learning-memory impairment caused by protein Tau hyperphosphorylation in depressed rats by down-regulating glutamate content.

REFERENCES

1. Conrad MS, Richard A. The Handbook for Electroconvulsive Therapy. California: Eagle Race Medical Technologies Company Press; 1999. p. 13-5.

2. Luo J, Min S, Wei K, et al. Propofol protects against impairment of learning-memory and imbalance of hippocampal Glu/GABA induced by electroconvulsive shock in depressed rats. J Anesth 2011; 25:657-65.

3. Andrade C, Singh NM, Thyagarajan S, et al. Possible glutamatergic and lipid signalling mechanisms in ECT-induced retrograde amnesia: Experimental evidence for involvement of COX-2, and review of literature. J Psychiatr Res 2008; 42:837-50.

4. Dong J, Min S, Wei K, et al. Effects of electroconvulsive therapy and propofol on spatial memory and glutamatergic system in hippocampus of depressed rats. J ECT 2010; 26:126-30.

5. Kloda A, Martinac B, Adams DJ. Polymodal regulation of NMDA-R channels. Channels (Austin) 2007; 1:334-43.

6. Dennis SH, Jaafari N, Cimarosti H, et al. Oxygen/glucose deprivation induces a reduction in synaptic NMDA-Rs on hippocNMDAl CA3 neurons mediated by mGluR1 and adenosine A3 receptors. J Neurosci 2011; 31:11941-52.

7. Vieira M, Fernandes J, Burgeiro, et al. Excitotoxicity through Ca2+-permeable NMDA-Rs requires Ca2+-dependent JNK activation. Neurobiol Dis 2010; 40:645-55.

8. Bliss RM, Finckbone VL, Trice J, et al. Tumor necrosis factor-α (TNF-α) augments NMDA-induced Purkinje neuron toxicity. Brain Res 2011; 1836:1-14.

9. Shutter L, Tong KA, Holshouser BA. Proton MRS in acute traumatic brain injury: Role for glutamate/glutamine and choline for outcome prediction. J Neurotrauma 2004; 21:1693-705.

10. Weingarten MD, Lockwood AH, Hwo SY, et al. A protein factor essential for microtubule assembly. Proc Natl Acad Sci USA 1975; 72:1858-62.

11. Osiecka KM, Nieznanska H, Skowronek KJ, et al. Tau inhibits tubulin oligomerization induced by prion protein. Biochim Biophys Acta 2010; 1813:1845-53.

12. Cleveland DW, Hwo SY, Kirschner MW. Purification of Tau, a microtubule-associated protein that induces assembly of microtubules from purified tubulin. J Mol Biol 1977; 116:207-25.

13. De Calignon A, Fox LM, Pitstick R, et al. Caspase activation precedes and leads to tangles. Nature 2010; 464:1201-4. 14. Combs B, Voss K, Gamblin TC, et al. Pseudohyperphosphorylation has differential effects on polymerization and function of tau isoforms. Biochemistry 2011; 50:9446-56.

15. Canu N, Filesi I, Pristerà A, et al. Altered intracellular distribution of PrPC and impairment of proteasome activity in Tau overexpressing cortical neurons. J Alzheimers Dis 2011; 27:603-13.

16. Wu FY, Feng Q, Cheng M, et al. The activation of excitatory amino acid receptors is involved in tau phosphorylation induced by cold water stress. Prog Biochem Biophys 2010; 37:510-6.

17. Petroni D, Tsai J, Mondal D, et al. Attenuation of low dose methylmercury and glutamate induced-cytotoxicity and tau phosphorylation by an N-methyl-D-aspartate antagonist in human neuroblastoma (SHSY5Y) cells. Environ Toxicol 2013; 28:700-6.

18. Zhang H, Wang W, Gao W, et al. Effect of propofol on the levels of neurotransmitters in normal human brain: A magnetic resonance spectroscopy study. Neurosci Lett 2009; 467:247-51.

19. Daskalopoulos R, Korcok J, Farhangkhgoee P, et al. Propofol protection of sodium-hydrogen exchange activity sustains glutamate uptake during oxidative stress. Anesth Analg 2001; 93:1199-204.

20. Hans P, Bonhomme V, Collette J, et al. Propofol protects cultured rat hippocampal neurons against N-methyl-D-aspartate receptor-mediated glutamate toxicity. J Neurosurg Anesthesiol 1994; 6:249-53.

21. Albertson TE, Walby WF, Stark LG, et al. The effect of propofol on CA1 pyramidal cell excitability and GABAA-mediated inhibition in the rat hippocampal slice. Life Sci 1996; 58:2397-407.

22. Kingston S, Mao LM, Yang L, et al. Propofol inhibits phosphorylation of N-methyl-D-aspartate receptor NR1 subunits in neurons. Anesthesiology 2006; 104:736-9.

23. Hans P, Bonhomme V, Collette J, et al. propofol protects cultured rat hippocampal neurons against N-methyl-D-aspartate receptor-mediated glutamate toxicity. J Neurosurg Anesthesiol 1994; 6:249-53.

24. Tasset I, Medina FJ, Pena J, et al. Olfactory bulberctomy induced oxidative and cell damage in rat: Protective effect of melatoni. Physiol Res 2010; 59:105-12.

25. Wang D, Noda Y, Tsunekawa H, et al. Bshavioural and neurochemical feaures of olfactory bulbertomized rats resembling depression with comorbid anixiety. Behav Brain Res 2007; 178:262-73.

26. Mutlu O, Ulak G, Celikyurt IK, et al. Effects of olanzapine, sertindole and clozapine on learning and memory in the Morris water maze test in naive and MK-801-treated mice. Pharmacol Biochem Behav 2011; 98:398-404.

27. Altar CA, Laeng P, Jurata LW, et al. Electroconvulsive seizures regulate gene expression of distinct neurotrophic signaling pathways. J Neuresci 2004; 24:2667-77.

28. Kato S, Kito Y, Hemmi H, et al. Simultaneous determination of D-amino acids by the coupling method of D-amino acid oxidase with high-performance liquid chromatography. J Chromatogr B Analyt Technol Biomed Life Sci 2011; 879: 3190-5.

29. Choi BR, Kwon KJ, Park SH, et al. Alternations of septal-hippocampal system in the adult Wistar rat with spatial memory impairments induced by chronic cerebral hypoperfusion. Exp Neurobiol 2011; 20:92-9.

30. Palmio J, Huuhka M, Laine S, et al. Electroconvulsive therapy and biomarkers of neuronal injury and plasticity: Serum levels of neuron-specific enolase and S-100b protein. Psychiatry Res 2010; 177:97-100.

31. Zhao LL, Hu GC, Zhu SS, et al. Propofol pretreatment attenuates lipopolysaccharide-induced acute lung injury in rats by activating the phosphoinositide-3-kinase/Akt pathway. Braz J Med Biol Res 2014; 47:1062-7.

32. Xu JJ, Wang YL. propofol attenuation of hydrogen peroxide-mediated oxidative stress and apoptosis in cultured cardiomyocytes involves haeme oxygenase-1. Eur J Anaesthesiol 2008; 25:395-402.

33. Stover JF, Kempski OS. Anesthesia increases circulating glutamate in neurosurgical patients. Acta Neurochir (Wien) 2005; 147:847-53.

34. Pesi? V, Milanovi? D, Tani? N, et al. Potential mechanism of cell death in the developing rat brain induced by propofol anesthesia. Int J Dev Neurosci 2009; 27:279-87.

for publication January 3, 2015.

Tel: 86-18102022730, E-mail: liuchao74001@126.com

△Supported by the National Natural Science Foundation (30972831) and the China Postdoctoral Science Foundation (2013M530880).