Protective Effect of Total Flavones of Abelmoschus manihot L. Medic Against Poststroke Depression Injury in Mice and Its Action Mechanismкод для вставкиСкачать
THE ANATOMICAL RECORD 292:412–422 (2009) Protective Effect of Total Flavones of Abelmoschus manihot L. Medic Against Poststroke Depression Injury in Mice and Its Action Mechanism MEI LIU, QIU-HONG JIANG, JI-LI HAO, AND LAN-LAN ZHOU* Department of Pharmacology, Basic Medical College, Anhui Medical University, Hefei 230032, People’s Republic of China ABSTRACT Total ﬂavones of Abelmoschus manihot L. Medic (TFA) is the major active component isolated from the traditional Chinese herb Abelmoschus manihot L. Medic. We investigated the protective effect of TFA against poststroke depression (PSD) injury in mice and its action mechanism. A mouse model of PSD was induced by middle cerebral artery occlusion (MACO) 30 min/reperfusion, followed by isolation feeding and chronic unpredictable mild stress for 2 weeks. Treatment groups received TFA at three different doses (160, 80, and 40 mg/kg, p.o.) or ﬂuoxetine (Flu, 2.5 mg/kg, p.o.) daily for 24 days. Change in behavior, brain tissue malondialdehyde (MDA) levels, and the activity of superoxide dismutase (SOD) and glutathione peroxidase (GSH-Px) were measured. The expression of brain-derived neurotrophic factor (BDNF) was detected by immunohistochemistry, and mRNA expression of BDNF and cAMP response elementbinding protein (CREB) analyzed by reverse transcription-polymerase chain reaction (RT-PCR). Treatment with TFA (160, 80, and 40 mg/kg) signiﬁcantly ameliorated mice escape-directed behavioral impairment induced by PSD, markedly reduced MDA levels, and increased the activity of SOD, GSH-Px close to normal levels. TFA administration also attenuated PSD-induced neuronal death/losses, upregulated expression of BDNF both at mRNA and protein levels, as well as CREB mRNA levels. TFA had a protective effect against PSD injury in mice. Cardioprotection involves the inhibition of lipid peroxidation and upregulation of BDNF-CREB levels in the hippocampus, which may also be important mechanism of its antidepressants. This potential protection makes TFA a promising therapeutic agent for the PSD. Anat Rec, 292:412–422, 2009. Ó 2009 Wiley-Liss, Inc. Key words: poststroke depression; total ﬂavones of Abelmoschus manihot L. Medic; brain-derived neurotrophic factor; cAMP response element-binding protein; lipid peroxidation Grant sponsor: Annual Research Project of Anhui Provincial Natural Science Foundation; Grant number: 05021005; Grant sponsor: Doctoral Foundation of Anhui Medical University; Grant number: xj2004005. *Correspondence to: Lan lan Zhou, Department of Pharmacology, Basic Medical College, Anhui Medical University, Hefei Ó 2009 WILEY-LISS, INC. 230032, People’s Republic of China. Fax: 186-5515161133. E-mail: firstname.lastname@example.org Received 18 August 2008; Accepted 3 December 2008 DOI 10.1002/ar.20864 Published online in Wiley InterScience (www.interscience.wiley. com). POSTSTROKE DEPRESSION Poststroke depression (PSD) is a common mood disturbance in stroke patients that potentially affects survivors and their recovery, and its estimated frequency ranges from 25%–79% (Spalletta et al., 2005). It is associated with excess disability, increased cognitive impairment, poor rehabilitation outcomes, worse quality of life, and increased suicidal tendencies and mortality (Ramasubbu et al., 1998; Kauhanen et al., 1999; Kimura et al., 2000). The pathophysiology of PSD remains elusive and appears to be multifactorial rather than purely biological or psychosocial in origin; psychological and social stressors associated with stroke are therefore considered to be the primary cause of depression (Gainotti et al., 1999; House, 1996). Recognition and treatment of the depression in stroke survivors is very important. An effective neuroprotective therapy for PSD has not been postulated until now. Recent studies have shown that major depression is associated with a signiﬁcant excess of free radicals and/ or lipid peroxidation, The involvement of oxidative stress in neuronal loss after PSD is well established. Several components of reactive oxygen species (ROS), i.e., superoxide, hydroxyl radical, hydrogen peroxide and peroxynitrite radical have been found to be generated after ischemia reperfusion injury, and play an important part in the neuronal loss after stoke. Inhibition of formation of free radicals during lipid peroxidation prevents development of brain edema and neuronal damage, which may be an important antidepression mechanism. Several enzymes (e.g., superoxide dismutase (SOD), glutathione peroxidase (GSH-Px)) are important in the antioxidant defense system because they metabolize free radicals or reactive oxygen intermediates to nonradical products. Dysfunction of SOD and GSH-Px may result in the loss of protective activity exerted by these enzymes, and has manifested as increased infraction in many studies. An emerging hypothesis suggests that impaired neuroprotection is closely involved associated with the pathogenesis of major depression. For example, decreased neuronal survival and disordered neurogenesis in the hippocampus have been repeatedly found in patients with major depression. They and are considered to be potential pathophysiological factors and therapeutic targets for major depression. The hippocampus is an important limbic structure that modulates mood, anxiety, learning and memory, all of which are commonly disrupted in depression (Fumagalli et al., 2005). Preclinical and clinical studies have demonstrated that major depression results in a reduction in the total number of neurons in the hippocampus by inducing a prolonged reduction in the rate of cell proliferation (WarnerSchmidt and Duman, 2006). Neurotrophins are known to be critical for the survival of neurons in the hippocampus during development in vivo and in culture (Duman et al., 1997; McAllisler et al., 1996; Thoenen, 1995). Brain-derived neurotrophic factor (BDNF) is the most widespread growth factor in the brain. It has diverse functions in the adult brain as a regulator of neuronal survival, proliferation and differentiation of cells, fast synaptic transmission, and activity-dependent synaptic plasticity. The BDNF plays a central part in the development and plasticity of the brain by opposing neuronal damage, and promoting neurogenesis and cell survival (Lewin and Barde, 1996; Blum and Konnerth, 2005). The transcription factor cAMP-response element- 413 binding protein (CREB) is an important regulator of BDNF-induced gene expression, and has a central role in mediating neurotrophin responses (Finkbeiner et al., 1997). Studies have shown that long-term exposure to stress is accompanied by neuronal atrophy of the hippocampus, and a decrease in the expression of BDNF (Smith et al., 1995). Brain imaging studies have demonstrated that the volume of the hippocampus is reduced in patients with chronic depression (Sapolsky, 2001). Long-term administration of antidepressants causes an increase in the expression of BDNF and CREB, and upregulation of CREB and BDNF occurs in response to several different classes of antidepressant treatment, including norepinephrine reuptake inhibitors, serotoninselective reuptake inhibitors, and electroconvulsive seizure. This indicates that the BDNF–CREB cascade is a common postreceptor target of these therapeutic agents. At present, there are three main kinds of classical antidepressants for PSD in clinical practice, including tricyclic antidepressants, selective reuptake inhibitors and monoamine oxidase inhibitors. Most of these drugs, however, have undesirable side effects, and their mechanism of action incompletely understood. Seeking safe and effective antidepressant agents from traditional herbs may uncover novel treatments for depressive disorders and reveal unknown mechanisms by which depressive symptoms can be alleviated. Abelmoschus manihot L. Medic is a plant rich in ﬂavones such as quercetin, hyperin and rutin. Total ﬂavone of Abelmoschus manihot L. Medic (TFA) is the major active component of this plant. TFA has been reported to have a neuroprotective effect against cerebral ischemia injury in rats and rabbits (Wen and Chen, 2007; Guo and Chen, 2002). Previous studies from our research group demonstrated that TFA treatment produced a protective effect against injury due to PSD in rats by normalizing hyperactivity in the hypothalamic–pituitary–adrenal axis (Hao et al., 2007). However, the mechanisms underlying the antidepressant effect are still obscure. Therefore, in the present study, we used a PSD model in the mouse to further explore the neuroprotective effect of TFA and its mechanism. We focused mainly on the antioxidant defense system and hippocampal neurogenesis, which are deeply involved in the etiology and treatment of depression. MATERIALS AND METHODS Animals This investigation conformed to the regulations stipulated by Anhui Medical University Animal Care Committee, which follows the protocol outlined in The Guide for the Care and Use of Laboratory Animals published by the US National Institute of Heath (NIH publication number 85-23; revised 1996). Experimental protocols described in this study were approved by the Ethics Review Committee for Animal Experimentation of Anhui Medical University. Healthy male mice (25 6 3.0 g) were housed in a room at a constant temperature of 228C 6 38C with 40% humidity under a 12-hr light-dark cycle. They had free access to tap water and food. They were allowed 1 week to acclimatize before experimentation. 414 LIU ET AL. Drugs and Reagents TFA (yellow powder; content of ﬂavones >60%; batch number, 20050809) was obtained from the Institute of Medicine of Anhui Province. Fluoxetine (Flu) was provided by the Eli Lilly and Drug Company Limited (Suzhou, China). Malondialdehyde (MDA), GSH-Px and SOD test kits were purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). The antibody of BDNF (rabbit anti-mouse), Streptavidin–Biotin Complex (SABC) kit and diaminobenzidine (DAB) staining kit were purchased from Boster Biological Engineering Company (Wuhan, China). Primer DNA was synthesized by the Shanghai Bioengineering Technical Service Limited Company (Shanghai, China). Trizol reagent and reverse transcription-polymerase chain reaction (RTPCR) Test Kits were obtained from Biotechnology Company (TakaRa, Biotechnology Company, Limited, Beijing, China). Drugs were dissolved in distilled water and all other chemicals were of the highest analytical grade available. Experimental Model of PSD Treatment of animals. Mice were randomly divided into seven groups: sham, ischemia, model, TFA (160, 80, and 40 mg/kg) and Flu (2.5 mg/kg). Drugs were given via the oral route, once a day for 24 consecutive days. Sham, ischemia, and model groups were given an equal volume of vehicle. Experiments were conducted on day-3, 60 min after the designated amount of TFA, Flu or vehicle had been given. Ischemia/stroke was induced in anesthetized (4% chloral hydrate, 350 mg/kg, i.p.) mice. After the stroke, mice in model and drug treatment groups were housed individually in metabolic cages; sham and ischemia mice were assigned randomly to two groups of ﬁve: maintained at a room temperature of 238C 6 28C, on a light-dark cycle of 12 hr/12 hr, with free access to water and food. On day-8, mice in the model and drug treatment groups were exposed to stress for 14 days. On day-24, behavioral changes were evaluated, and mice were killed for biochemical study. The design of the time course of the experiment and sampling times are shown in Fig. 1. Middle cerebral artery occlusion in mice. Middle cerebral artery occlusion (MCAO) was induced in mice using the intraluminal ﬁlament method as previously described (Kilic et al., 2001) with mild modiﬁcation. Brieﬂy, anesthesia was induced by injection of 4% chloral hydrate (350 mg/kg, i.p). Under an operating microscope, a mid-line incision was made on the ventral surface of the neck, and the right common carotid arteries and external carotid artery isolated and ligated with an 8.0 silk suture. A polyamide monoﬁlament (Ethilon, Johnson and Johnson International, Belgium) coated to a round tip with silicone resin (thread thickness; Xantopren, Bayer Dental, Germany) was introduced into the intracranial internal carotid artery (ICA) through an incision in the common carotid artery. The monoﬁlament was carefully advanced approximately 11– 13 mm distal to the carotid bifurcation (which was beyond the origin of the middle cerebral artery). After occlusion was achieved, the monoﬁlament was secured Fig. 1. Outline of the design of the time course of the experiments and the sampling times. in place with ligature. The incision was sutured. Occlusion was done for a period of 30 min. After 30 min of MCAO, the ﬁlament was completely withdrawn from the ICA to allow reperfusion. Sham-operated mice underwent the same surgical procedure except that their common carotid arteries were not ligated and occluded. The rectal temperature was maintained at 378C 6 0.58C throughout surgery using a heating pad (Biomed S.L, Spain). The heart rate and electrocardiogram were monitored throughout the procedure (Surgi Vet, USA). Neurological status. Neurological function was evaluated 2 hr, 1, 3, 5, 7, 14, 21 days after MCAO 30min/reperfusion using a modiﬁed seven-point behavioral rating scale (Longa et al., 1989). Neurological scores were deﬁned as: 0 5 no neurological deﬁcit; 1 5 failure to fully extend left forepaw; 2 5 reduced grip of left forelimb while the tail was gently pulled; 3 5 circling toward the contralateral side if tail was pulled when moving spontaneously; 4 5 circling or walking to the left; 5 5 walking only if stimulated; 6 5 unresponsive to stimulation with a depressed level of consciousness; 7 5 death. Mice with neurological score of 1–6 were included in the current study for further establishment of a depression model; mice displaying subarachnoid hemorrhage and deep anesthesia signs were excluded. Chronic unpredictable mild stress model. On the day-8, poststroke mice in drug treatment and model groups were subjected to different stressors per day randomly over 14 days between 8:00 AM to noon. The procedure was a minor modiﬁcation of the method described previously (Willner et al., 1987). The order of stressors used was: high-speed agitation for 1 min; day and night reversal for 24 hr; cold stress at 48C for 5 min; overhang for 1 min; heat stress at 458C for 5 min; food deprivation for 24 hr; and water deprivation for 24 hr. Each stressor was repeated twice during the 14-day stress procedure. After 24 hr, mice underwent behavioral tests. Behavioral Tests The forced swimming Forced swimming test. test employed was similar to that described elsewhere (Porsolt et al., 1977). Brieﬂy, mice were individually forced to swimming inside a rectangular glass jar (25 cm 3 12 cm 3 25 cm containing 15 cm of water maintained at (238C–258C) for 6 min. After the initial 2–3 min of vigorous activity, they showed immobility by ﬂoating POSTSTROKE DEPRESSION 415 with minimum movements. A mouse was considered to be immobile when it ceased struggling and remained ﬂoating motionless in the water, making only small movements necessary to keep its head above water. The duration of immobility was recorded during the last 4 min of the 6-min testing period. ﬁxed for 24 hr with 10% paraformaldehyde PBS. They were embedded in parafﬁn, and coronal sections (thickness, 5 mm) cut on a parafﬁn microtome (Leica RM2135, Germany) for histological analysis. Parafﬁn sections were stained with hematoxylin and eosin for morphological evaluation. Passive avoidance task. Twenty-four hours after the last time stress, mice were trained individually in a step-through passive avoidance apparatus termed a ‘‘shuttle-box.’’ A straight wooden case (40 cm 3 12 cm 3 18 cm) had a removable cover. The alley was divided into two equal-length compartments (20 cm 3 12 cm 3 18 cm) by a partition with a rotundum. A light bulb (40 W) was positioned in the corner on the left side of the apparatus. The left compartment was bright and the right compartment was dark. The ﬂoor of the apparatus was made of copper grid (length, 12 cm; diameter, 2 cm). The method was similar to a previously described method (Simona and Claudio, 1997; Zhou et al., 2000). On the learning day, each mouse was placed in the bright compartment, facing away from the dark compartment. When the mouse turned around, the door leading to the dark compartment was opened. When the mouse had stepped with four paws into the dark side, a foot shock (40 V) was delivered and the time (latency, L) taken to move to the dark compartment was recorded. Numbers (number of errors, NR) for the mice to enter the darkroom to suffer electric shock in 5 min were also recorded. Mice were returned immediately to their cages. Retention was tested 24-hr later following the same procedure, after which mice were killed. Brain and hippocampal tissues were isolated for biochemical studies and histopathological examination. Immunohistochemistry for BDNF. Parafﬁn sections from mice brains were used for immunohistochemical analysis. In brief, 5-mm sections were deparafﬁnized in xylene and dehydrated through graded ethanol, and washed thrice with PBS (0.02 mmol/L, pH 7.4) for 5 min. Sections were incubated with 3% H2O2 for 10 min at room temperature to inactivate native peroxidase. After rinsing thrice with PBS for 5 min, they were heated in a microwave oven at a low power for 5 min in 0.01 mmol/L citrate buffer (pH 6.0). After rinsing thrice with PBS for 5 min, sections were kept in 5% BSA for 30 min, followed by incubation with a polyclonal antibody to BDNF (diluted 1:150; Boster) overnight at 48C (PBS buffer solution was used to replace antibody-BDNF in the negative control group). Sections were rinsed for 5 min three times with PBS before incubation with biotinylated goat anti-rabbit IgG antibody for BDNF for 60 min, and then incubated with streptavidin-biotin complex for 60 min at 378C after rinsing three times with PBS. Sections were rinsed for another 5 min three times with PBS before the reaction with DAB solution for 5 min. Sections were dehydrated through graded ethanol. They were enveloped with gelatin and observed under a bright-ﬁeld microscope. Analyses were carried out with an observer blinded to the experimental protocol. Six sections per animal and three ﬁelds per section of the same magniﬁcation were utilized for quantitative analysis. Expression of BDNF-positive neurons in the CA1, CA3, and DG of hippocampus was observed under the microscope. The mean optical density of BDNF-positive neurons was measured using Image J 1.34 analysis system in each section. Biochemical analysis. Six mice from each group were decapitated at the end of the behavioral test. Brains were quickly removed and the hippocampus excised on an ice-plate. It was washed with ice-cold sterile physiological (0.9%) saline, blotted dry on ﬁlter paper and weighed before 10% (wt/vol) homogenates were made in ice-cold saline. Homogenates were centrifuged at 1000g for 10 min, and the supernatant stored at 2808C for biochemical study. Measurement of MDA levels, SOD, and GSH-Px activity in brain tissue. Brain homogenates were used for estimation of MDA levels, as well as measuring the activity of SOD and GSH-Px. The MDA levels were measured at 523 nm by the thiobarbituric acid method (Wong et al., 1987). The activity of SOD and GSH-Px was detected by spectrophotometry at 550 nm (Sun et al., 1998) and 423 nm (He et al., 2008), respectively. Protein concentration was estimated using bovine serum albumin as a standard (Lowry et al., 1951). Histological examination of hippocampus specimens. At the end of behavioral test, mice were anesthetized with 4% chloral hydrate (350 mg/kg, i.p) Intracardiac perfusion with cold saline was done, followed by 4% paraformaldehyde in phosphate-buffered saline (PBS) (0.1 M; pH 7.4). Brains were removed and Analysis of BDNF mRNA and CREB mRNA expression levels by RT-PCR. The levels of BDNF mRNA and CREB mRNA were determined by RT-PCR. Mice were decapitated and whole brains quickly removed on an ice-plate. Tissues were washed with cold saline. Hippocampal extracts were carefully excised and stored at 2808C until RNA extraction. Levels of BDNF and CREB transcripts were analyzed by RT-PCR in studies in which total RNA from hippocampal tissues (30–50 mg) were isolated using an RNA isolation kit (TakaRa) according to manufacturer’s instructions. cDNA was reverse transcribed from 1 mL of total RNA with a RNA PCR Kit (AMV) Ver. 3.0. Brieﬂy, reverse transcription was done using 1 mL total RNA with 5 U/mL AMV Reverse Transcriptase 0.5 mL, 40 U/mL RNase Inhibitor 0.25 mL, 103 RT Buffer 1 mL, Oligod T-Adaptor Primer 0.5 mL, 10 mM dNTP Mixture 1 mL, 25 mM MgCl2 2 mL, RNase Free dH2O 3.75 mL in a reaction volume of 10 mL. Then incubation at 558C for 30 min, 998C for 5 min, and 58C for 5 min was done. Template cDNA were ampliﬁed according to manufacturer’s instructions: 53 PCR Buffer 10 mL, TaKaRa ExTaqHS 0.25 mL, Template cDNA 10 mL Sense Primer (10 mM) 0.5 mL, Anti-Sense Primer (10 mM) 0.5 mL then 416 LIU ET AL. TABLE 1. Effect of TFA on recovery of neurological function in PSD mice Group Sham Ischemia Model Flu TFA Postoperative Dose / mg kg21 – – – 2.5 160 80 40 2.50 1.67 1.78 1.73 1.63 1.40 2h d1 d3 0 6 1.35 6 1.32D 6 1.09 6 0.82 6 0.80 6 0.84 0 6 0.71 6 0.76D 6 0.87 6 0.71 6 0.64 6 0.67 0 6 0.64 6 0.76D 6 0.60* 6 0.43* 6 0.46 6 0.60* 1.50 1.56 1.33 1.64 1.25 1.3 0.90 1.56 0.89 0.80 1.00 0.90 d5 0.70 1.67 0.67 0.58 0.88 0.60 0 6 0.45 6 0.93D 6 0.36** 6 0.52** 6 0.49* 6 0.42** d7 0.50 1.33 0.56 0.20 0.50 0.58 0 6 0.41 6 1.00D 6 0.53* 6 0.10** 6 0.38* 6 0.42* d14 0.40 0.78 0.44 0.10 0.38 0.36 0 6 0.35 6 0.69D 6 0.26 6 0.10* 6 0.37 6 0.25 d21 0.40 0.56 0.23 0.09 0.10 0.20 0 6 0.35 6 0.44D 6 0.20* 6 0.07** 6 0.10** 6 0.20* Note: Neurological function was evaluated postoperatively using a seven-point behavioral rating scale as described in the ‘‘Materials and Methods’’ section. Data are mean 6 SD (n 5 15). D P < 0.01, versus the Sham; *P < 0.05 and **P < 0.01 versus the model. sterile water was added to a total volume of 40 mL. Primer sequences for the RT-PCR primers were: (forward: 50 -GACAAGGCAACTTGGCCTAC-30 and reverse: 50 -CTGTCACACACGCTCAGCTC-30 , BDNF, 352 bp, accession number, NM007540); (forward: 50 -TACCCAGG GAGGAGCAATAC-30 and reverse 50 -GAGGCAGCTT GAACAACAAC-30 , CREB, 183 bp accession number, NM009952); (forward: 50 -GCTGAGAGGGAAATCGT 0 0 GCGT-3 and reverse: 5 -GAAGCATTTGCGGTGCAC GATG-30 , b-actin, 516 bp, accession number, NM007393). PCR was done by MyCyclerTM Thermal cycler (BioRad, USA) under the following ampliﬁcation conditions: PCR cycles consisted of 3 min at 948C, 33 cycles 948C for 20 sec, 558C for 20 sec, 728C for 30 sec and a ﬁnal extension at 728C for 4 min (BDNF, 352bp); 45 cycles 948C for 20 sec, 558C for 20 sec, 728C for 30 sec and a ﬁnal extension at 728C for 4 min (CREB, 183 bp); 26 cycles 948C for 20 sec, 558C for 20 sec, 728C for 30 sec and a ﬁnal extension at 728C for 4 min (b-actin, 516bp). After ampliﬁcation, samples underwent electrophoresis on 2% agarose gels, stained with ethidium bromide and photographed under UV light. The intensity of the PCR products generated by the b-actin and target sequences (BDNF, CREB) was quantiﬁed by a computer-assisted, linear scanning densitometric analysis of the photograph in reﬂectance mode. A ratio of target to b-actin PCR product units was used to calculate the relative increase of BDNF mRNA and CREB mRNA expression. Statistical analysis. Statistical analysis was done by SPSS 11.5. Results were mean 6 SD signiﬁcant differences were assessed by one-way analysis of variance (ANOVA) for comparisons among groups with least signiﬁcant difference (LSD) test. Differences between groups were deemed signiﬁcant at P < 0.05. score of 0). All ischemic mice showed prominent neurological function deﬁcits 2 hr after stroke, indicated by physical signs. Over a period of 14 days after surgery, the neurological function of all ischemic mice recovered to some extent. In comparison with the sham or ischemia groups, recovery of neurological function in the model group were signiﬁcantly slower (P < 0.05, P < 0.01), indicating that comprehensive separate feeding plus chronic unpredictable mild stress (CUMS) after stroke seriously affected the recovery of neurological function of mice. Treatment with TFA (160, 80, and 40 mg/kg) or Flu (2.5 mg/kg) clearly improved the impairment in neurological function in PSD mice compared with model group (P < 0.05, P < 0.01). Effect of TFA on Immobility Time in PSD Mice The immobility time in the forced swimming test was signiﬁcantly longer in the PSD group compared with the sham group (Fig. 2). TFA treatment at 160, 80, and 40 mg/kg signiﬁcantly reduced the total immobility time. This suggested that TFA could reverse the behavior despair induced by PSD, similar to effect of Flu. Effect of TFA on Learning and Memory Deﬁcits in PSD Mice The passive avoidance test is a standard test used in passive learning and memory research. Twenty-four hours after the last time stress, the passive avoidance test was conducted. The results showed that mice in the model group developed a remarkable loss of learning and memory compared with the sham-operated (P < 0.01; Table 2). After administration of various doses of TFA, L was clearly increased, and NR was signiﬁcantly reduced (P < 0.01, P < 0.05). This result suggested that the impairment in learning and memory of PSD mice were partly restored by TFA treatment. RESULTS Effect of TFA on Neurological Status Effect of TFA on MDA Levels, SOD and GSH-Px Activity in PSD Mice Mice usually exhibited some neurological function deficits after cerebral ischemia (e.g., hemiplegia, loss of balance, no spontaneous motor activity). Two hours after MCAO 30 min/reperfusion, we immediately noted the neurological function score (Table 1). Sham-operated mice did not show neurological deﬁcits (neurological MDA levels were elevated, and the activity of SOD and GSH-Px was signiﬁcantly decreased by 61% and 69.8%, respectively, in PSD mice (Table 3). Treatment with TFA (160, 80, and 40 mg/kg) or Flu (2.5 mg/kg) could markedly reduce lipid peroxidation, and elevate the activity of the two enzymes to near control values. 417 POSTSTROKE DEPRESSION Effect of TFA on Histopathological Response of Hippocampal Tissues in PSD Mice H&E staining revealed that in the sham-operated group the morphology of neurons in CA1, CA3, and DG of hippocampal tissues were normal, which manifested as intact cellular bodies, clear structure, light-red cytoplasm and deep-blue nucleus (Fig. 3A). In the model group, pyra- mid neurons of the CA1, CA3, and granule neurons in dentate gyrus were severely degenerative, including cellular swelling, loss or death of neurons, partial karyopyknosis in neurons, and proliferation of microglial cells (Fig. 3C). Treatment with TFA at different doses or Flu clearly improved pathomorphological change of hippocampal neurons, such as reduction of neuronal degeneration and loss or death of neurons (Fig. 3E–G). Effect of TFA on Hippocampal BDNF Immunoreactivity in PSD Mice Immunohistochemical staining demonstrated normalsized BDNF-positive neurons arranged closely in the sham-operated group that were mainly in pyramid cell layer of the CA1 and CA3, and in the granule neuron layer in DG (Fig. 4A). Neurons of model mice were in a loose arrangement; some had degenerated into a vacuole- or atrophic-like shape, and the relative number of BDNF-positive neurons was signiﬁcantly reduced (Figs. 4C, 5). Long-term TFA treatment at 160, 80, or 40 mg/kg or Flu (2.5 mg/kg) signiﬁcantly improved pathomorphological change induced by PSD and increased the relative number of BDNF-positive neurons (Figs. 4E–G, 5). Fig. 2. Effect of TFA on the immobility time in PSD mice in the forced swimming test. Procedures are described in ‘‘Materials and Methods’’ section. Data are mean 6 SD of 15 mice per group. DP < 0.05 versus the sham; *P < 0.05 and **P < 0.01 versus the model. S, sham; I, ischemia; M, model. Effect of TFA on the Expressions of BDNF mRNA and CREB mRNA in PSD Mice We examined the levels of BDNF mRNA and CREB mRNA by RT-PCR to further investigate the protective TABLE 2. Effect of TFA on impairment of learning and memory in PSD mice Learning test Group Sham Ischemia Model Flu TFA 21 L (S) Dose/mg kg 2.5 160 80 40 80.8 71.7 14.3 69.2 70.1 71.6 72.0 6 6 6 6 6 6 6 Memory test NR 44.4 54.9 7.8D 63.7** 54.8** 58.8** 66.8** 3.8 4.8 10.6 6.0 5.6 5.7 6.3 6 6 6 6 6 6 6 L(S) 2.2 4.4 5.3D 3.3* 4.4* 5.9* 3.6* 241.9 225.3 99.8 213.0 219.9 190.5 217.4 6 6 6 6 6 6 6 NR 93.5 96.7 62.4D 110.5** 95.6** 120.6* 104.2** 1.0 1.4 2.4 1.5 1.3 1.4 1.6 6 6 6 6 6 6 6 0.5 0.7 1.0D 0.6** 0.8** 0.5** 0.9* Note: Twenty-four hours after the last time stress, mice were trained individually in a step-through passive avoidance apparatus termed a ‘‘shuttle-box’’. A foot shock (40 V) was delivered, and the L and NR for the mice to move from the light compartment to the dark compartment was recorded. Data are mean 6 SD (n 5 15). D P < 0.01 versus the Sham; *P < 0.05 and **P < 0.01 versus the model. TABLE 3. Effect of TFA on MDA levels, and the activity of SOD and GSH-Px in brain tissue of PSD mice Group Sham Ischemia Model Flu TFA Dose/ mgkg21 2.5 160 80 40 MDA/Ug21pr 2.57 2.88 4.56 3.40 2.52 2.78 3.70 6 6 6 6 6 6 6 0.57 0.88 0.93DD 0.63** 0.64** 0.87** 0.62** SOD/NUmg21pr 56.81 47.28 34.70 56.00 89.00 55.16 39.53 6 6 6 6 6 6 6 15.88 15.28 10.45DD 12.89** 15.60** 17.78** 16.53 GSH-PX/activity U 22.79 19.8 15.90 21.86 26.95 21.92 23.15 6 6 6 6 6 6 6 7.81 8.93 5.96D 4.43* 6.39** 3.50* 5.63* Note: brain samples were collected after the passive avoidance task and 10% (w/v) homogenates were prepared as described in the ‘‘Materials and Methods’’ section. MDA level, and the activity of SOD and GSH-Px was measured. Data are mean 6 SD (n 5 8). D P < 0.05 and DDP < 0.01 versus the sham. *P < 0.05 and **P < 0.01 versus the model. 418 LIU ET AL. Fig. 3. Effect of TFA on the histopathology of the hippocampus of PSD mice. Mice were decapitated and hippocampus tissue quickly removed for histological analysis after the passive avoidance task. Sections (5 mm) were cut. Parafﬁn sections were stained with hematoxylin and eosin and observed with a microscope for conventional morphology (Original magniﬁcation A–G, 3400, scale bar 5 50 mm). (A) Sham, (B) Ischemia, (C) Model, (D) Flu 2.5 mg/kg, (E) TFA 160 mg/ kg, (F) TFA 80 mg/kg, and (G) TFA 40 mg/kg. effect of TFA on hippocampal neurons. Densitometric analysis of the levels of BDNF mRNA and CREB mRNA (which were normalized to expression levels of the housekeeping gene b-actin and expressed as the ratio of b-actin) revealed a signiﬁcant increase of the levels of BDNF mRNA and CREBmRNA in PSD mice upon treatment with TFA. A signiﬁcant reduction in BDNF mRNA and CREB mRNA levels in the hippocampus were observed after the PSD procedure compared with the sham-operated or ischemia group. TFA (160, 80 mg/kg) or Flu (2.5 mg/kg) could clearly increase the expression of BDNF mRNA and CREB mRNA in the hippocampus (Fig. 6). DISCUSSION Experimental models of depression are essential for investigating the psychobiology of depression and for screening new antidepressants. Having access to highly Fig. 4. Effect of TFA on the immunohistochemical examination of BDNF in the hippocampus of PSD mice. Mice were decapitated and hippocampus tissue quickly removed for immunohistochemical analysis after the passive avoidance task. Expression of BDNF protein in the hippocampal CA1, CA3, and DG regions was evaluated by mean optical density values with an Olympus microscope. Scale bar 5 50 mm. Original magniﬁcation A–G, 3400. (A) Sham, (B) Ischemia, (C) Model, (D) Flu 2.5 mg/kg, (E) TFA 160 mg/kg, (F) TFA 80 mg/kg, and (G) TFA 40 mg/kg. valid animal models is very important (Willner, 1991). Separate feeding plus stressors based on cerebral ischemia was developed as a well-validated animal models of depression for mimicking biological and psychosocial syndrome resemblances between the animal model and clinical depressive condition in humans (Liu et al., 2004; POSTSTROKE DEPRESSION 419 Fig. 6. Effect of TFA on the levels of hippocampal BDNF mRNA and CREB mRNA expression in PSD mice. BDNF mRNA and CREB mRNA expression was detected by RT-PCR. (A), (B): PCR products were separated on a 2% agarose gel and stained with ethidium bromide. (C), BDNF mRNA expression, and (D), CREB mRNA expression, upper panel shows RT-PCR example from three experiments and lower panel shows relative ratios of BDNF mRNA and CREB mRNA to b-actin, respectively. Data are mean 6 SD of BDNF mRNA and CREB mRNA to b-actin from at least three experiments. DP < 0.01 versus the Sham; *P < 0.05 and **P < 0.01 versus the model. S, sham; I, ischemia; M, model. M, Marker; 1, sham; 2, ischemia; 3, model; 4, ﬂu; 5, TFA 160 mg/kg; 6, TFA 80 mg/kg; and 7, TFA 40 mg/kg. Fig. 5. Effect of TFA on mean optical density of BDNF immunoreactivity in DG and CA1, CA3, regions in PSD mice. The mean optical density value was analyzed in three ﬁelds of the same magniﬁcation with Image J 1.34 analysis system software. Data are mean 6 SD of six mice per group. DP < 0.01 versus the sham; *P < 0.05 and **P < 0.01 versus the model. S, sham; I, ischemia; M, model. Tang et al., 2004). We think that the key feature of a PSD model is to ﬁrst set-up a model of cerebral ischemia. Garcia (Garcia, 1984) thought that the ideal cerebral ischemia model should meet following requirements: (a) a single artery that can be occluded reproducibly; (b) that vascular occlusion results in predictable changes in blood ﬂow, i.e., focal or regional ischemia; and (c) reperfusion is possible after cerebral ischemia. The MCAO model produced by an intraluminal thread not only meets these requirements, but also has other advantages: (i) the skull is not opened, thereby avoiding a change in the internal environment of the skull and intracranial infection; (ii) the model is reliable, causes little trauma, and is stable; and (iii) mortality is very low and neurological function recovers 1 week after surgery. We chose a MCAO 30 min/reperfusion procedure to produce a model of cerebral ischemia in the mouse. There were no neurological deﬁcits in the sham-operated group; and all ischemia-operated mice displayed high neurological function scores 2 hr after surgery. Stress is considered to be important for the genesis of depression, as well as a risk factor in depressive disorder development (Barden, 2004). The CUMS regimen, an experimental model of depression with good validity, was employed in conjunction with single rearing to mimic psychosocial stress including ‘‘daily hassles’’ and the social isolation of stroke survivors (Willner, 1991). Separate feeding may cause social memory impairment, and a reduction in social interaction (Von Frijtag et al., 2000). The PSD mouse model was set-up on the basis of focal cerebral ischemia by means of a MCAO 30 min/ reperfusion procedure followed by separate feeding plus CUMS. Mice were then subjected to forced swimming 420 LIU ET AL. and a passive avoidance test to screen behavioral change. The forced swimming test was considered a relatively quick and simple test to the predict antidepressant activity of drugs (Borsini and Meli, 1998; David et al., 2003; Redrobe and Bourin, 1999). The immobility in this test reﬂected a failure of persistence in escapedirected behavior, or the development of passive behavior that disengages the animal from active forms of coping with stressful stimuli, and this behavior is sensitive for antidepressants. The passive avoidance test was considered to reﬂect hippocampus-dependent learning and memory ability. Consistent with previous reports, the present study showed that PSD animals failed to acquire a normal escape response as demonstrated by the increased immobility time in forced swimming test, and an increased number of errors and reduced latency to a darkened environment in the passive avoidance task, which simulated core symptoms in PSD patients. These behavioral abnormalities induced by PSD were reversed by long-term treatment with TFA at 160, 80, and 40 mg/ kg or Flu (2.5 mg/kg). Taken together, the results from the current study conﬁrmed and extended previous ﬁndings that TFA could effectively inhibit depression behaviors, and showed a potential protective effect against injury by PSD insult, similar to the classical antidepressant Flu. Full understanding of the mechanisms responsible for the protective effect of TFA is of great relevance with respect to its therapeutic potential. Oxidative stress has emerged as a potentially important factor in the pathogenesis of PSD. The brain is entirely dependent upon oxidative metabolism for cell survival, and is particularly sensitive to oxidative damage because of its high content of iron, polyunsaturated fatty acids, catecholamines and excitatory amino acids, all of which may mediate oxidative stress and ROS production (e.g., superoxide, hydroxyl ion, nitric oxide) (Marciniak and Petty, 1996; Juurlink and Sweeney, 1997). Recognition of the involvement of oxidative stress in neurodegenerative diseases has generated substantial interest in investigating the ability of naturally occurring antioxidants (photochemical) to ameliorate neuronal damage associated with neurodegeneration and aging (Youdim and Joseph, 2001; Joseph et al., 2005). MDA, the product of lipid peroxidation of the cell membrane, indirectly represents the degree of free-radical production and lipid peroxidation. SOD and GSH-Px, indicators of the activity of antioxidative enzymes, are very important metabolic enzymes in neurons, and directly reﬂect the scavenging capacity of some ROS. We measured MDA levels, and the activity of SOD and GSH-Px to show the extent of peroxidation and neuronal damage in PSD mice. TFA treatment (160, 80, and 40 mg/kg) inhibited the increases in levels of brain MDA and elevated the activity of SOD and GSH-Px. This may be an important mechanism by which TFA exerts an antidepressant effect. Hippocampal formation is a major target tissue in neurodegenerative disorders and is particularly susceptible to the deleterious effect of stressors (Smith et al., 1991; Smith et al., 1996). Recent evidence suggests that depression also has a close association with neuroplasticity and cellular adaptation which may be because of the downstream events beyond the receptors at the levels of intracellular signaling molecules (Duman et al., 2000). It is therefore of great signiﬁcance to study and develop antidepressant drugs that can promote hippocampal neurogenesis. Neurotrophic factors such as BDNF are known to be crucial for the survival of neurons during development. The BDNF gene contains a cAMP response element to which phosphorylated CREB binds and thereby enhances transcription (Korte et al., 1995). CREB is a transcription factor regulating expressions of several genes involved in neuroplasticity, cell survival, and cognition; CREB can be modulated by BDNF. The AMPc-MAPkinase-CREB-BDNF cellular cascade may have a signiﬁcant role in the mechanisms of dendritic restructuring, increase in hippocampal neurogenesis and survival of nerve cells, and may serve as protective factor in neuronal development. Recent basic and clinical studies provide evidence for a neurotrophic hypothesis of depression and antidepressant action. According to this hypothesis, reduced expression of BDNF or CREB could contribute to the atrophy of hippocampal neurons in response to stress in stroke patients, and the upregulation of BDNF and CREB could help to protect neurons from damage and death, which may be the mechanism of antidepressants (Nestler et al., 2002a; Nestler et al., 2002b). Cerebral ischemia/reperfusion and stress resulting in reduction of hippocampal BDNF levels are accompanied by structural change and neuronal damage in certain brain areas (e.g., hippocampus, frontal cortex) (Vollmayr et al., 2001). Long-term administration of various antidepressants increases the expression, phosphorylation and function of CREB and its downstream target gene BDNF in rat hippocampus, and other limbic brain regions are thought to be involved in depression (Thomel et al., 2000; Duman, 1996). Some molecular events involved in PSD pathology are known to affect the process of neurogenesis entirely or partially, for example, exposure to psychosocial or stress decreases the number of new neurons in the DG, CA1, and CA3 hippocampus subﬁelds by downregulating the proliferation of granule cell precursors. Histological analysis in our study indicated that many cells characterized by pyknosis or karyorrhexis were most frequently seen in the hippocampal CA1, CA3, and DG subﬁelds (H&E staining) in the model groups. Expression of BDNF mRNA and CREB mRNA were reduced in PSD mice by RT-PCR. Long-term TFA treatment (80, 160 mg/kg) could effectively ameliorate neuron injury, and increase expression of BDNF mRNA and CREB mRNA. These results are in accordance with studies showing that certain classical antidepressants (e.g., Flu) can increase BDNF and CREB levels in rat hippocampus (Tiraboschi et al., 2004). These observations suggest that TFA can upregulate BDNF-CREB levels, and may have an important role in protecting hippocampal neurons against PSD-induced injury. In summary, these ﬁndings raise the possibility that TFA treatment, via antilipid oxidation, and upregulation of BDNF and CREB levels, produces a protective effect against injury in response to chronic stress or other environmental stimuli after stroke, thereby exerting an antidepressant-like effect. These ﬁndings shed light on the mechanisms by which TFA acts as a neuroprotective agent, particularly as an antioxidant and modulator of the hippocampal neurotrophin and transcription factors. These data may help to support future development of TFA as a potential treatment for PSD. POSTSTROKE DEPRESSION ACKNOWLEDGMENTS The author is grateful to De-yi Cen (Department of Pharmacology, Basic Medical College, Anhui Medical University, Hefei, China) and Hai-yan Yan (Department of Pharmacology, Basic Medical College, Anhui Medical University) for their assistance in this study. LITERATURE CITED Barden N, 2004. Implication of the hypothalamic-pituitary-adrenal axis in the physiopathology of depression. J Psychiatry Neurosci 29:185–193. Blum R, Konnerth A. 2005. Neurotrophin-mediated rapid signaling in the central nervous system: mechanisms and functions. Physiology (Bethesda) 20:70–78. Borsini F, Meli A, 1998. Is the forced swimming test a suitable model for revealing antidepressant activity? Psychopharmacology (Berl) 94:147–160. David DJ, Renard CE, Jolliet P, et al. 2003. Antidepressant-like effects in various mice strains in the forced swimming test. Psychopharmacology 166:373–382. Duman RS. 1996. A role for CREB and BDNF in the actions of antidepressants. Biol Psychiatry 39:558–559. Duman RS, Heninger GR, Nestler EJ. 1997. A molecular cellular theory of depression. Arch Gen Psychiatry 54:597–606. Duman RS, Malbery J, Nakagawa S, D’Sa C. 2000. Neuronal plasticity and survival in mood disorders. Biol Psychiatry 48:732–739. Finkbeiner S, Tavazoie SF, Maloratsky A, Jacobs KM, Harris KM, Greenberg ME. 1997. CREB: a major mediator of neuronal neurotrophin responses. Neuron 19:1031–1047. Fumagalli F, Molteni R, Calabrese F, Frasca A, Racagni G, Riva MA. 2005. Chronic ﬂuoxetine administration inhibits extracellular signal-regulated kinase1/2 phosphorylation in rat brain. J Neurochem 93:155–160. Gainotti G, Azzoni A, Marra C. 1999. Frequency, phenomenology and anatomical-clinical correlates of major post-stroke depression. Br J Psychiatry 175:163–167. Garcia JH. 1984. Experimental ischemia stroke: a review. Stroke 15:5–14. Guo Y, Chen ZW. 2002. Protective effect of Abelmoschus manihot L. medic against cerebral ischemia-reperfusion injury. Chin Pharmacol Bull 18:692–695. Hao JL, Yan HY, Zhou LL. 2007. Effect of total ﬂavone of Abelmoschus manihot L. medic on behavior and plasma ACTH, CORT levels in post-stroke depression in rats. Acta Univ Med Anhui (Chin) 42:48–51. He P, He WH, Wang A, Xia T, Xu B, Zhang M, Chen X. 2008. PBDE-47-induced oxidative stress, DNA damage and apoptosis in primary cultured rat hippocampal neurons. Neurotoxicology 29:124–129. House A. 1996. Depression associated with stroke. J Neuropsychiatry Clin Neurosci 8:453–457. Joseph JA, Shukitt-Hale B, Casadesus G. 2005. Reversing the deleterious effect of aging on neuronal communication and behavior: beneﬁcial properties of fruit polyphenolic compounds. Am J Clin Nutr 81 (Suppl 1):313S–316S. Juurlink BHJ, Sweeney MI. 1997. Mechanisms that result in damage during and following cerebral ischemia. Neurosci Biobehav Rev 21:121–128. Kauhanen ML, Korpelainen JT, Hiltunen P, Brusin E, Mononen H, Määttä R, Nieminen P, Sotaniemi KA, Myllylä VV. 1999. Poststroke depression correlates with cognitive impairment and neurological deﬁcits. Stroke 30:1875–1880. Kilic E, Bahr M, Hermann DM, 2001. Effect of recombinant tissue plasminogen activation after intraluminal thread occlusion in mice: role of hemodynamic alterations. Stroke 32:2641–2647. Kimura M, Robinson RG, Kosier JT. 2000. Treatment of cognitive impairment after post-stroke depression: a double-blind treatment trial. Stroke 31:1482–1486. 421 Korte M, Carroll P, Wolf E, Brem G, Thoenen H, Bonhoeffer T. 1995. Hippocampal long-term potentiation is impaired in mice lacking brain-derived neurotrophic factor. Proc Natl Acad Sci USA 92:8856–8860. Lewin GR, Barde YA. 1996. Physiology of the neurotrophins. Annu Rev Neurosci 19:289–317. Liu FY, Yang S, Chen WY, Tang Y. 2004. Effect of Yu Le Shu on neurotransmitter of PSD hamster. Pract J Clin Med 1:30–31. Longa EZ, Weinstein PR, Carlson S, Cummins R. 1989. Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke 20:84–91. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. 1951. Protein measurement with the Folin phenol reagent. J Biol Chem 193:265–275. Marciniak G, Petty MA. 1996. Design and biological evaluation of new anti-oxidants for use in cerebrovascular disorders. Drugs Future 21:1037–1046. McAllisler AK, Katz LC, Lo DC. 1996. Neurotrophin regulation of cortical dendritic growth requires activity. Neuron 17:1057–1064. Nestler EJ, Barrot M, DiLeone RJ, Eisch AJ, Gold SJ, Monteggia LM. 2002a. Neurobiology of depression. Neuron 34:13–25. Nestler EJ, Buncan M, Duman RS, Greshenfeld HK, Hen R, Koester S, Lederhendler I, Meaney M, Robbins T, Winsky L, Zalcman S. 2002b. Preclinical models status of basic research in depression. Biol Psychiatry 52:503–528. Porsolt RD, Bertin A, Jalfre M. 1977. Behavioral despair in mice: a primary screening test for antidepressants. Arch Int Pharmacodyn Ther 229:327–336 Ramasubbu R, Robinson RG, Flint AJ, Kosier T, Price TR. 1998. Functional impairment associated with acute post-stroke depression: the stroke data bank study. J Neuropsychiatry Clin Neurosci 10:26–33. Redrobe JP, Bourin M. 1999. Evidence of the activity of lithium on 5-HT1B receptors in the mouse forced swimming test: comparison with carbamazepine and sodium valproate. Psychopharmacology 141:370–377. Sapolsky RM. 2001. Depression, antidepressants, and the shrinking hippocampus. Proc Natl Acad Sci USA 98:12320–12322. Simona C, Claudio C. 1997. Impairments produced by amphetamine and stresses on memory storage are reduced following a chronic stressful experience. Psychopharmacology 129:161–167. Smith CD, Camey JM, Starke-Reed PE, Olivier CN, Stadtman ER, Floyd RA, Markesbery WR. 1991. Excess brain protein oxidation and enzyme dysfunction in normal aging and Alzheimer disease. Proc Natl Acad Sci USA 88:10540–10543. Smith MA, Makino S, Kvetmasky R, Post RM. 1995. Stress alters the expression of brain-derived neurotrophic factor and neurotrophin-3 mRNA in the hippocampus. J Neurosci 15:1768–1777. Smith MA, Perry G, Richey PL, Sayre LM, Anderson VE, Beal MF, Kowall N. 1996. Oxidative damage in Alzheimer’s disease. Nature 382:120–121. Spalletta G, Ripa A, Caltagirone C. 2005. Symptom proﬁle of DSMIV major and minor depressive disorders in ﬁrst-ever stroke patients. Am J Geriatr Psychiatry 13:108–115. Sun Y, Oberley LW, Li Y. 1988. A simple method for clinical assay of superoxide dismutase. Clin Chem 34:497–500. Tang QS, Pei QH, Hou XJ, Qu M. 2004. Establishment of the animal model of post-stroke depression. J Beijing Univ Tradit Chin Med 27:33–36. Thoenen H. 1995. Neurotrophins and neuronal plasticity. Science 270:593–598. Thomel N, Sakai KH, Shin C, Thome J, Sakai N, Shin K, Stenffen C, Zhang YJ, Impey S, Storm D, Duman RS. 2000. CAMP response element-mediated gene transcription is upregulated by chronic antidepressant treatment. J Neurosci 20:4030–4036. Tiraboschi E, Tardito D, Kasahara J, Moraschi S, Pruneri P, Gennarelli M, Racaqni G, Popoli M. 2004. Selective effect of antidepressants on CREB phosphorylation and related kinase cascade. Eur Neuropsychopharmacol 14:S10–S11. Vollmayr BH, Henn FA, Lewicka S, Faust H. 2001. Brain-derived neurotrophic factor (BDNF) stress response in rats bred for learned helplessness. Mol Psychiatry 6:471–474. 422 LIU ET AL. Von Frijtag JC, Reijmers LG, Van der Harst JE, Van den Bos R, Spruijt BM. 2000. Defeat followed by individual housing results in long-term impaired reward- and cognition-related behaviors in rats. Behav Brain Res 117:137–146. Warner-Schmidt JL, Duman RS. 2006. Hippocampal neurogenesis: opposing effect of stress and antidepressant treatment. Hippocampus 16:239–249. Wen JY, Chen ZW. 2007. Protective effect of pharmacological preconditioning of total ﬂavones of Abelmoschl manihot on cerebral ischemia reperfusion injury in rats. Am J Chin Med 35:653–661. Willner P. 1991. Animal models as simulations of depression. Trends Pharmacol Sci 12:131–136. Willner P, Towell A, Sampson D, Sophokleous S, Muscat R. 1987. Reduction of sucrose preference by chronic unpredictable mild stress and its restoration by a tricyclic antidepressant. Psychopharmacology 93:358–364. Wong SH, Knighet JA, Hopfer SM, Zaharia O, Leach CN, Jr., Sunderman FW, Jr. 1987. Lipoperoxides in plasma as measured by liquidchromatographic separation of malondialdehyde-thiobarbituric acid adduct. Clin Chem 33:214–220. PMID: 3802504. Youdim KA, Joseph J. 2001. A possible emerging role of phytochemicals in improving age-related neurological dysfunction: a multiplicity of effect. Free Radic Biol Med 30:583–594. Zhou LL, Ming L, Ma CG, Jiang Q. 2000. Protective effect of extract of folium ginkgo on repeated cerebral ischemia-reperfusion injury. Chin J Integrated Tradit West Med (Chin) 20:356–358.