AMPK activation prevents prenatal stress-induced cognitive impairment: Modulation of mitochondrial content and oxidative stress

Free Radical Biology and Medicine 75 (2014) 156–166

Contents lists available at ScienceDirect

Free Radical Biology and Medicine journal homepage: www.elsevier.com/locate/freeradbiomed

Original Contribution

AMPK activation prevents prenatal stress-induced cognitive impairment: Modulation of mitochondrial content and oxidative stress Ke Cao a, Adi Zheng a, Jie Xu a, Hao Li a, Jing Liu a, Yunhua Peng a, Jiangang Long a, Xuan Zou b, Yuan Li a, Cong Chen a, Jiankang Liu a, Zhihui Feng a,n a Center for Mitochondrial Biology and Medicine, The Key Laboratory of Biomedical Information Engineering of Ministry of Education, School of Life Science and Technology and Frontier Institute of Science and Technology, Xi’an Jiaotong University, Xi’an, China b Center for Translational Medicine, The Key Laboratory of Biomedical Information Engineering of Ministry of Education, School of Life Science and Technology and Frontier Institute of Science and Technology, FIST, Xi’an Jiaotong University, Xi’an, China

art ic l e i nf o

a b s t r a c t

Article history: Received 12 June 2014 Received in revised form 10 July 2014 Accepted 23 July 2014 Available online 1 August 2014

Prenatal stress induces cognitive functional impairment in offspring, an eventuality in which mitochondrial dysfunction and oxidative stress are believed to be closely involved. In this study, the involvement of the AMP-activated protein kinase (AMPK) pathway was investigated. A well-known activator, resveratrol (Res), was used to induce AMPK activation in SH-SY-5Y cells. Significant mitochondrial biogenesis and phase II enzyme activation, accompanied by decreased protein oxidation and GSSG content, were observed after Res treatment, and inhibition of AMPK with Compound c abolished the induction effects of Res. Further study utilizing a prenatal restraint stress (PRS) animal model indicated that maternal supplementation of Res may activate AMPK in the hippocampi of both male and female offspring, and that PRS-induced mitochondrial loss in the offspring hippocampus was inhibited by Res maternal supplementation. In addition, Res activated Nrf2-mediated phase II enzymes and reduced PRS-induced oxidative damage in both male and female offspring. Moreover, PRS markedly decreased mRNA levels of various neuron markers, as well as resultant offspring cognitive function, based on spontaneous alternation performance and Morris water maze tests, the results of which were significantly improved by maternal Res supplementation. Our results provide evidence indicating that AMPK may modulate mitochondrial content and phase II enzymes in neuronal cells, a process which may play an essential role in preventing PRS-induced cognitive impairment. Through the coupling of mitochondrial biogenesis and the Nrf2 pathway, AMPK may modulate oxidative stress and be a promising target against neurological disorders. & 2014 Elsevier Inc. All rights reserved.

Keywords: AMPK Resveratrol Mitochondrial biogenesis Oxidative stress Prenatal stress Cognitive function

Introduction Prenatal stress during gestation has many deleterious effects on both the development and the behavior of offspring [1]. Clinical studies demonstrate that exposure of pregnant mothers to stressful conditions increases the susceptibility of their offspring to mental Abbreviations: AMPK, AMP-activated protein kinase; Arc, activity-regulated cytoskeleton-associated protein; BDNF, brain-derived neurontrophic factor; EMX2, empty spiracles homeobox 2; GAP43, growth-associated protein-43; GCL, glutamate-cysteine ligase; HO-1, heme oxygenase 1; NMDAR, N-methyl-D-aspartic acid receptor; Nrf2, NF-E2-related factor; NQO-1, NAD(P)H dehydrogenase (quinone 1); PGC-1α, peroxisome proliferator-activated receptor gamma, coactivator 1 alpha; PRS, prenatal restraint stress; Res, resveratrol; SCG10, stathmin-like 2. n Correspondence to: Center for Mitochondrial Biology and Medicine, School of Life Science and Technology and Frontier Institute of Science and Technology, Xi’an Jiaotong University, 28 W, Xian-ning Road, Xi’an 710049, China. Fax: 86 029 82665849. E-mail address: [email protected] (Z. Feng). http://dx.doi.org/10.1016/j.freeradbiomed.2014.07.029 0891-5849/& 2014 Elsevier Inc. All rights reserved.

disorders, such as depression, schizophrenia, and cognitive deficits [2]. Animal studies show that prenatal stress causes alterations of both the hypothalamic–pituitary adrenocortical axis and the brain neurotransmitter systems and also impairs hippocampal-dependent spatial learning and memory abilities in offspring [3,4]. Pregnant rodents that have suffered restraint stress represent a valid model of stress with neurobiological and behavioral consequences [5,6]. Although the specific mechanisms of prenatal restraint stress (PRS) remain unclear, evidence suggests that both oxidative stress and mitochondrial dysfunction may be involved in PRS-induced neurological damage and cognitive impairment [7,8]. However, the detailed mechanisms regulating each of these processes have not been elucidated in the rodent model. The involvement of cellular energy metabolism in different conditions has become an area of intense interest [9]. As a sensor of cellular energy status, AMP-activated protein kinase (AMPK) is an attractive target for a range of diseases, such as cancer [10], diabetes [11], and cardiovascular disease [12]. Studies also implicate a neuroprotective

K. Cao et al. / Free Radical Biology and Medicine 75 (2014) 156–166

effect of AMPK both in vitro [13] and in vivo [14]. However, whether AMPK plays a role in PRS-induced neurological impairment is still unknown. It has been demonstrated that AMPK activity correlates strongly with mitochondrial function by promoting a mitochondrial biogenesis pathway. Activation of AMPK is dependent on the upregulation of peroxisome proliferator-activated receptor gamma coactivator-1alpha (PGC-1α) and nuclear respiratory factor 1 (NRF1) expression in rat visual cortical neurons [15]. In addition to mitochondrial dysfunction, oxidative stress was also involved in PRS-induced neurological damage and cognitive impairment [7,8]. Induction of phase II detoxifying enzymes is one of the most important pathways for cells to fight against oxidative stress. Nuclear factor erythroid-2related factor-2 (Nrf2) is an antioxidant transcription factor mediating the expression of antioxidant enzymes, such as NADPH quinineoxido reductase-1 (NQO1) and heme oxygenase-1 (HO-1). It has a wide range of activities in regulating redox state and energy metabolism in cells [16]. The response of AMPK to oxidative stress has been recently reported, but the downstream signals of this response are largely unknown. The potential for cross talk between the AMPK and the Nrf2 cascades has been reported in Caenorhabditis elegans [17], in human endothelial cells [18], and in mammalian inflammatory systems [19].

157

However, no information about the potential for convergence between the AMPK and the Nrf2 pathways or the subsequent exertion of a neuroprotective effect exists. Resveratrol (3,40 ,5 trihydroxystilbene, Res), a naturally occurring phytoalexin compound present in almost 70 plant species (such as grape, peanut, and soya beans), is considered one of the most effective known antioxidants. Similar to other chemical activators of AMPK, such as 5-aminoimidazole-4-carboxamide ribonucleoside (AICAR) and metformin, Res has been widely accepted as a natural AMPK activator [20,21]. The impact of Res on the prevention of prenatal stress-induced cognitive impairment has been recently reported, but the exact mechanisms involved in its neuroprotective effects are still poorly characterized [22–24]. In the present study, to investigate the potential regulatory effect of AMPK on mitochondrial biogenesis and phase II enzyme induction, and their subsequent involvement in prenatal stressinduced cognitive dysfunction, an AMPK activator was employed both in SH-SY-5Y cells and in a PRS animal model. We propose that the upregulation of mitochondrial biogenesis and Nrf2 pathways by AMPK activation may play an important role in promoting neuron survival and related improvement in cognitive function.

Fig. 1. Effects of Res on mitochondria and phase II enzymes in SH-SY-5Y cells. Cells were treated with Res at doses of 10, 50, and 100 μM for 24 h. Mitochondrial biogenesisrelated proteins were analyzed by Western blot (A, Western blot image; B, statistical analysis of PGC-1α; C, statistical analysis of Complex I; D, statistical analysis of Complex III). Phase II enzyme-related proteins were determined by Western blot (E, Western blot image; F, statistical analysis of Nrf2; G, statistical analysis of NQO1; H, statistical analysis of HO-1). Protein oxidation was determined by measuring carbonyl protein content (I, Western blot image; J, statistical analysis). GSSG was evaluated (K). GSH/GSSG ratio was calculated (L). Values are means 7 SEM from at least three independent experiments. nP o 0.05, nnP o 0.01.

158

K. Cao et al. / Free Radical Biology and Medicine 75 (2014) 156–166

Material and methods

Animals and treatments

Chemicals

Sprague-Dawley female and male rats were purchased from the SLAC laboratory Animal Co. Ltd. (Shanghai, China). Female rats weighing 230–250 g and male rats weighing 280–350 g were used. All animals were housed in a temperature (24–27 1C) and humidity (60%) controlled animal room and maintained on a 12 h light/12 h dark cycle (light from 8:00 to 20:00), with food and water provided during the experiments. All of the procedures were performed in accordance with the United States Public Health Services Guide for the Care and Use of Laboratory Animals, and all efforts were made to minimize both the suffering and number of animals used in this study. After 1 week of acclimatization, female rats were randomly divided into the following three groups: Control, Stress, and Res treatment (100 mg/kg/day).

Antibodies against β-actin were purchased from Sigma (St. Louis, MO, USA). Antibodies against complexes I (30 kDa), II (30 kDa), III (51 kDa), IV (40 kDa), and V (55 kDa) were purchased from Invitrogen (Carlsbad, CA, USA). Antibodies against PGC-1α, Nrf2, NQO1, and HO-1 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Antibodies against AMPK and p-AMPK (Thr172) were purchased from Cell Signaling Technology (Danvers, MA, USA). PCR primers were synthesized by Baiaoke Biotech (Beijing, China). TRIzol and other reagents were purchased from Invitrogen (Carlsbad, CA). Resveratrol was purchased from APP-Chem Bio (Xi’an, China).

Fig. 2. Effects of AMPK on mitochondrial biogenesis and Phase II enzyme induction by Res in SH-SY-5Y cells. Cells were treated with Res at doses of 10, 50, and 100 μM for 24 h, and AMPK activation was determined by measuring the expression of p-AMPK and AMPK (A, Western blot image; B, statistical analysis of p-AMPK; C, statistical analysis of p-AMPK/AMPK ratio). Cells were treated with 100 μM Res, with or without Compound c (10 μmol/L) for 24 h, and the protein expression of p-AMPK, AMPK, PGC-1α, Complex I, Nrf2, and HO-1 was determined by Western blotting (D, Western blot image; E, statistical analysis of PGC-1α; F, statistical analysis of Complex I; G, statistical analysis of Nrf2; H, statistical analysis of HO-1). Values are means 7 SEM from at least three independent experiments. nP o 0.05, nnP o 0.01.

K. Cao et al. / Free Radical Biology and Medicine 75 (2014) 156–166

For each group, 14 female and 3 male rats were used. Virgin female rats were placed with adult male rats (3:1) overnight for mating. Vaginal smears were examined on the following morning. The day on which a smear was determined to be sperm-positive was set as embryonic day 0. Each pregnant rat was then housed separately, and Res gavages were implemented on a daily basis until baby rats were born.

Stress procedure Each group of pregnant rats, except the Control group, was exposed to restraint stress on days 14–20 of pregnancy, three times each day and for 2 h each time. To prevent the habituation of the animals as a result of being subjected to daily procedures, restraint periods were randomly shifted to different time periods (08:00 AM to 11:00 AM, 11:00 AM to 02:00 PM, and 04:00 PM to 07:00 PM). The restraint device was a transparent plastic tube (6.8 cm in diameter) with closed ends and air holes for breathing. The length could be adjusted to accommodate the size of the animal. After birth, the offspring of all groups were housed in the same animal room and kept together with their biological mothers. The pregnant rats of the Control group were left undisturbed. On day 21, after all offspring were weaned, male and female pups were housed separately until testing at 1 month of age.

159

Morris water maze The Morris water maze (MWM) was performed as previously reported [8]. The day before beginning the MWM test, each rat was allowed to swim freely for 120 s in order to become familiar with the novel environment of the maze and to locate and climb onto the escape platform. For four and half consecutive days, each animal performed eight swimming trials per day, with two trials beginning at each of the four start positions (N, E, S, W). The average result of the four start positions was defined as the result for that session. The time lapse between 2 sessions was 6 h. The order of start positions was randomized each day, and across days, for all animals. Latency and swim patterns were digitally tracked by HVS Image Water 2020. For each trial, rats were allowed to swim until they reached the platform and climbed onto it, subject to a 120 s cutoff. Any animal that failed was guided to find the platform and allowed to stay on it for 3 s. The variable recorded was escape latency, a measure of the time required for an animal to reach the platform and remain there for more than 3 s. For cases in which rats did not reach the platform, the latency value was set at 120 s. Data were recorded and then analyzed by parametric ANOVA (repeated measures and multivariate measures).

T maze for spontaneous alternation performance Spontaneous alternation is the natural tendency of rats and mice to alternate their choice of goal arm (left or right arm) during

Fig. 3. Effects of PRS and Res on the AMPK pathway in both male and female offspring hippocampus. Total proteins were prepared from offspring hippocampus, and p-AMPK and AMPK levels were detected by Western blot. In male offspring as follows: (A) Western blot image, (B) statistical analysis of p-AMPK, and (C) ratio of p-AMPK/AMPK; in female offspring as follows: (D) Western blot image, (E) statistical analysis of p-AMPK, and (F) ratio of p-AMPK/AMPK. Values are means 7 SEM; n Z 8. nP o 0.05, nn P o 0.01.

160

K. Cao et al. / Free Radical Biology and Medicine 75 (2014) 156–166

exploration of a T maze. The size of the maze was based on parameters used for rats in a previous publication [25]. The rats were confined to the start area for 30 s before testing. The sliding door was then removed, and the rats were allowed to freely explore the rest of the maze for an 8-min trial [26,27]. Numbers of visits to each goal arm and spontaneous alternation performance were recorded by the observer. The apparatus was cleaned after each rat’s use to remove any residual odors and debris.

Cell culture and treatment The human SH-SY-5Y cell line was obtained from ATCC (Manassas, VA, USA). Cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (v/v) fetal bovine serum and maintained at 37 1C in a humidified atmosphere of 95% air and 5% CO2 by incubator. The culture medium was changed every other day. Resveratrol was dissolved in dimethyl sulfoxide (DMSO) at 100 mM and was stored at  20 1C. For

cellular treatment, resveratrol was dissolved in DMEM (with 10% fetal bovine serum) to reach a final concentration of 100 μM. Biochemical analysis Small portions of hippocampal tissue were collected and homogenized in ice cold phosphate buffered saline (PBS). After centrifugation (1000g, 10 min), the supernatant was collected for analysis. GSH and GSSG content was analyzed using commercial clinical diagnosis kits according to the manufacturer’s standards and protocols (Jiancheng, Nanjing, China). Briefly, total glutathione (oxidized þ reduced) assay is based on the reaction with the thiolspecific reagent dithionitrobenzoic acid after glutathione reductase treatment to change GSSG to GSH. The adduct was measured spectrophotometrically at 412 nm. GSSG assay is based on the same reaction with a pretreatment for clearance of GSH. Thus, the GSH content is equal to total glutathione content minus GSSG content.

Fig. 4. Effects of PRS and Res on mitochondrial content in both male and female offspring hippocampus. Total proteins and DNA were prepared from offspring hippocampus, and mitochondria subunit expression and mtDNA contents were detected by Western blot or real-time PCR. In male offspring as follows: (A) Western blot image, (B) statistical analysis of PGC-1α, Complex I, III, IV, and V subunits, and (C) mtDNA content; in female offspring as follows: (D) Western blot image, (E) statistical analysis of PGC-1α, Complex I, and IV, and (F) mtDNA content. Values are means 7 SEM; n Z 8. nP o 0.05, nnP o 0.01.

K. Cao et al. / Free Radical Biology and Medicine 75 (2014) 156–166

Protein carbonylation assay Protein carbonyls in soluble proteins were assayed using the Oxyblot protein oxidation detection kit (Cell Biolabs, San Diego, CA, USA). Protein carbonyls were labeled with 2,4-dinitrophenylhydrazine and detected by Western blot. As a negative loading control, equal amounts of samples were subjected to 10% SDSPAGE and stained with Coomassie brilliant blue.

161

mtDNA copy analysis, total DNA was extracted using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany), followed by quantitative real-time PCR with mitochondrial D-loop primers. Data were normalized to the mRNA of actin as a housekeeping gene and were analyzed by the 2-△△Ct method. Final results were presented as a percentage of the control. The specific primers used in the study are presented in Supplemental Table 1. Western blotting

Real-time PCR Total RNA was isolated from hippocampal samples using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA), and 1 μg of RNA was reverse-transcribed into cDNA using a RT-PCR kit (TaKaRa, DaLian, China) according to the manufacturer’s protocol, followed by semiquantitative real-time PCR with specific primers. For the

Hippocampal samples and SH-SY-5Y cellular samples were lysed with Western and IP lysis buffer (Beyotime, Jiangsu, China). The lysates were homogenized, and the homogenates were centrifuged at 13,000g for 15 min at 4 1C. The supernatants were collected, and the protein concentrations were determined with a BCA protein assay kit. Equal aliquots (20 μg) of the protein samples

Fig. 5. Effects of PRS and Res on phase II enzymes in both male and female offspring hippocampus. Total RNA and proteins were prepared from offspring hippocampus, and phase II enzyme regulator Nrf2 and its target genes were analyzed. In male offspring as follows: (A) mRNA expression of Nrf2, NQO1, HO-1, GCLc, and GCLm, (C) Western blot image, (D) statistical analysis of Nrf2, (E) statistical analysis of NQO-1, and (F) statistical analysis of HO-1; in female offspring as follows: (B) mRNA expression of Nrf2, NQO1, HO-1, GCLc, and GCLm, (G) Western blot image, (H) statistical analysis of Nrf2, (I) statistical analysis of NQO-1, and (J) statistical analysis of HO-1. Values are means 7 SEM; n Z 8. nP o 0.05, nnP o 0.01.

162

K. Cao et al. / Free Radical Biology and Medicine 75 (2014) 156–166

were loaded for Western blotting. Chemiluminescent detection was performed using an ECL Western blotting detection kit and quantified by scanning densitometry. Statistical analysis Data are presented as the means 7 SEM. Statistical analyses were conducted using one-way ANOVA followed by least significant difference post hoc analysis. For water maze data, statistical analysis was conducted using two-way ANOVA repeated measures. For all analyses, values of P o0.05 were considered statistically significant.

a dose-dependent manner. Similar induction was observed with Nrf2, the key regulator of phase II enzymes, as well as its target genes, NQO1 and HO-1 (Fig. 1E–H), indicating the simultaneous activation of mitochondrial biogenesis and Nrf2 pathways by Res. In addition, Res may effectively ameliorate oxidative stress in SH-SY-5Y cells by decreasing both carbonyl protein (Fig. 1I and J) and GSSG content (Fig. 1K) in a dose-dependent manner. Although GSH content was not affected after Res treatment (data not shown), GSH/GSSG ratio exhibited a dose-dependent increase after Res treatment, indicating an enhancement in antioxidant system (Fig. 1L). AMPK inhibition diminishes the induction effect of Res on mitochondria and phase II enzymes

Results Res promotes mitochondrial biogenesis and induces phase II enzymes in SH-SY-5Y cells As an activator of AMPK, Res was used to treat SH-SY-5Y cells for 24 h. As illustrated in Fig. 1, Res may significantly increase expression of PGC-1α and Complex I and III subunits (Fig. 1A–D) in

To further confirm that the induction effect of Res on mitochondria and phase II enzymes was a result of AMPK activation, we first confirmed that Res may significantly induce AMPK activation by increasing p-AMPK content and the p-AMPK/AMPK ratio in a dose-dependent manner in SH-SY-5Y cells (Fig. 2A–C). The application of Compound C, a common AMPK inhibitor, sufficiently blocked the induction effect of Res, as evidenced by

Fig. 6. Effects of PRS and Res on oxidative stress in both male and female offspring hippocampus. Total proteins were prepared from offspring hippocampus, and carbonyl protein as marker of protein oxidation was measured by Western blot: (A) Western blot image and (B) statistical analysis in male offspring; (D) Western blot image and (E) statistical analysis in female offspring. Tissue homogenates were prepared from offspring: GSH and GSSG were measured, and the ratio of GSH/GSSG was determined in male (C) and female (F) offspring. Values are means 7 SEM; n Z 8. nP o 0.05, nnP o 0.01.

K. Cao et al. / Free Radical Biology and Medicine 75 (2014) 156–166

the resulting expression of PGC-1α, Complex I, Nrf2, and HO-1 (Fig. 2D–H). PRS suppresses AMPK activation in offspring hippocampus To study the involvement of AMPK in prenatal restraint stressinduced offspring cognitive dysfunction, Res was supplemented during the pregnancy at a dose of 100 mg/kg/day. As shown in Fig. 3A, PRS induced significant declines in both p-AMPK content and p-AMPK/AMPK ratio in the hippocampus of male offspring (Fig. 3A–C). In female offspring, PRS did not have a significant effect on p-AMPK content compared with controls, whereas the p-AMPK/AMPK ratio was decreased (Fig. 3D–F). Supplementation of Res effectively increased both p-AMPK content and the p-AMPK/AMPK ratio in the hippocampus of both male and female offspring (Fig. 3), suggesting significant activation of AMPK by Res. Maternal Res supplementation inhibits PRS-induced mitochondrial loss in offspring hippocampus To determine whether mitochondrial biogenesis was affected in the PRS-induced cognitive impairment model, expression of PGC-1α, mitochondrial Complex subunits, and mtDNA content were examined. As shown in Fig. 4A-C, PRS decreased protein expression of PGC-1α, Complex I, III, IV, and V subunits in male offspring, as well as mtDNA content, indicating decreased mitochondrial content. As expected, Res significantly increased the expression of those mitochondrial biogenesis-related proteins, as well as mtDNA content (Fig. 4A–C). In female offspring, Res demonstrated the ability to normalize the decline of PGC-1α, Complex IV, and mtDNA content induced by PRS (Fig. 4D–F). Although PRS had no significant effect on Complex I, Res nonetheless markedly increased its expression (Fig. 4D and E). Moreover, PRS induced a decrease in the mRNA content of PGC-1α, PGC-1β, NRF1, NRF2α, NRF2β, TFAM, and TFB1M in both male and female offspring, an eventuality that could also be normalized by Res treatment (Fig. S1). Taken together, these data indicated that PRS-induced mitochondrial loss may be prevented by Res supplementation.

163

and a decrease in the GSH/GSSG ratio (Fig. 6C) in male offspring, whereas no significant effects were observed in female offspring (Fig. 6D–F). Despite the difference between male and female offspring, Res supplementation sufficiently decreased oxidative damage, as evidenced by the decrease in carbonyl protein and the increased GSH/GSSG ratio in both male and female offspring compared to the PRS control group (Fig. 6). Effects of PRS and Res on mRNA levels of neurotrophic factors Oxidative damage is known to contribute to neuron loss and associated disease progression. We therefore examined neurotrophic factors in offspring, as their presence may partially reflect neuron condition. As shown in Fig. 7A and B, PRS significantly decreased the mRNA content of several factors, including Arc, BDNF, EMX2, GAP43, NMDAR, and SCG10 in male offspring, as well as Arc, EMX2, and NMDAR in female offspring. Res treatment significantly increased the mRNA levels of each of those factors in both male and female offspring compared with PRS control groups. Effects of PRS and Res on cognitive function The T maze and Morris water maze tests were used to assess spatial learning and memory abilities in rat offspring. As shown in Fig. 8A, B, D, and E, PRS caused decreased learning and memory abilities compared to the T maze controls in both male and female offspring, although the total amount of exploration in female offspring was not significant compared with the control group

Maternal Res supplementation activates Nrf2-mediated phase II enzymes in offspring In vitro results indicated that AMPK may regulate phase II enzyme expression, and decreased AMPK activation was observed in PRS-treated offspring (Fig. 3). Therefore, we further evaluate phase II enzymes expression in offspring hippocampus. As shown in Figs. 5A and B, PRS significantly decreased the mRNA content of NQO1, HO-1, GCLc, and GCLm in male offspring, as well as GCLm in female offspring, whereas Res supplementation effectively increased mRNA levels of Nrf2 and its target genes, NQO1, HO-1, GCLc, and GCLm, in both male and female offspring compared with a PRS control (Fig. 5A and B).Western blot results confirmed that the protein levels of Nrf2, NQO-1, and HO-1 were all markedly increased in the Res-treated group compared to the PRS control (Fig. 5C–J), although the decline of these proteins induced by PRS was not significant in female offspring (Fig. 5G–J). Taken together, these data suggested that Res may activate the Nrf2 pathway in the offspring hippocampus. Effects of PRS and Res on oxidative stress Oxidative stress was suggested to be involved in PRS-induced neurological damage and cognitive impairment [7,8], and mitochondrial dysfunction and inhibition of phase II enzymes may each contribute to increased oxidative damage. As shown in Fig. 6, PRS induced a significant increase in carbonyl protein (Fig. 6A and B)

Fig. 7. Effects of PRS and Res on mRNA levels of neurotrophic factors in both male and female offspring hippocampus. Total RNA were prepared from offspring hippocampus: mRNA expressions of Arc, BDNF, EMX2, GAP43, NMDAR, and SCG10 were determined by real-time PCR in both male (A) and female (B) offspring. Values are means 7 SEM; n Z 8. nP o 0.05, nnP o 0.01.

164

K. Cao et al. / Free Radical Biology and Medicine 75 (2014) 156–166

Fig. 8. Effects of PRS and Res on cognitive impairment in both male and female offspring. After baby rats were born and grown to 1 month of age, both male and female offspring were tested for cognitive function. In male offspring as follows: (A) Total amount of exploration and (B) success rate of exploration in T maze, and (C) escape latency in Morris water maze; in female offspring as follows: Total amount of exploration and (D) success rate of exploration in T maze, and (E) escape latency in Morris water maze (F). All groups of animals were able to learn the task through consecutive trials. The latency per testing session represents the average of four trials of all animals in each group. Values are means 7 SEM. For T maze (n Z 10) and for Morris water maze (n Z 15). nP o 0.05, nnP o 0.01.

(Fig. 8D). Res sufficiently increased the total amount of and success rate of exploration of the T maze, indicating improved spatial learning and memory abilities (Fig. 8A, B, and E). Similar to the T maze test, the results of the Morris water maze test showed that PRS significantly increased escape latency time in both male and female offspring (Fig. 8C and F). Res significantly decreased escape latency time compared with the control group in both male and female offspring (Fig. 8C and F). Taken together, these data illustrated that the cognitive impairment induced by PRS was efficiently ameliorated through maternal Res supplementation.

Discussion Prenatal stress reportedly has many deleterious effects on the development and behavior of offspring [1,5,6]. Although oxidative stress and mitochondrial dysfunction were believed to be involved in PRS-induced neurological damage and cognitive impairment [7,8], the details regarding specific mechanisms remain largely unknown. Recent studies reveal AMPK, a major regulator of cellular and whole-body energy homeostasis, as an attractive therapeutic target for a range of diseases, including neurological disorders [28,29]. Therefore, in the current study, we investigated

the involvement of AMPK in PRS-induced neurological impairment, and we proposed that AMPK may affect neuron survival and cognitive function through the modulation of mitochondrial biogenesis and phase II enzyme expression. As a well-known ATP producer, mitochondria are also a major source of ROS. Studies suggest that oxidative stress resulting from mitochondrial dysfunction may play an important role in neurological impairment and cognitive decline [30–32]. Previously, we found that PRS-induced cognitive decline was closely associated with a functional disorder in hippocampal mitochondria [8,33]. Recent data indicate that the capacity to undergo mitochondrial biogenesis, which is dysregulated in disease states, may play a vital role in determining neuron cell survival [34]. As a key regulator of mitochondrial biogenesis, PGC-1α was believed activated by AMPK in rat visual cortical neurons, as previously reported [15]. We consistently observed the dose-dependent induction of PGC-1α through activation of AMPK by Res, a known natural AMPK activator, accompanied by the increased presence of mitochondrial complex subunits in SH-SY-5Y cells. Mitochondrial biogenesis induction was efficiently abolished with an AMPK inhibitor, Compound c. In further animal analysis, we found decreased p-AMPK/AMPK levels in the hippocampus of both male and female offspring, suggesting the PRS-induced inactivation of

K. Cao et al. / Free Radical Biology and Medicine 75 (2014) 156–166

AMPK. As a result, mitochondrial content was decreased, as evidenced by the decreased amount of mitochondrial complex subunits and mtDNA copy number. However, via maternal supplementation of Res, increased AMPK activation was observed in offspring, accompanied by increased mitochondrial biogenesis. Therefore, we proposed that modulation of AMPK activity may affect mitochondrial biogenesis both in vivo and in vitro. Nrf2 is an antioxidant transcription factor mediating the expression of phase II antioxidative enzymes, and has a wide range of activities in regulating redox state and energy metabolism in cells [16]. Upregulation of Nrf2 and its downstream genes (such as NQO1, HO-1, and GCL) is usually considered evidence for Nrf2 pathway activation. Recent studies indicate the existence of cross talk between the AMPK and the Nrf2 pathways in Caenorhabditis elegans [17], human endothelial cells [18], and mammalian inflammatory systems [19]. In the current study, we found that activation of AMPK may increase the protein content of Nrf2 and its target genes, NQO-1 and HO-1, in SH-SY-5Y cells in a dose-dependent manner. Through maternal Res supplementation, significant increases in both mRNA and protein content were observed in both male and female offspring of the Res-treated group compared with the PRS-only group, suggesting the obvious activation of Nrf2-mediated phase II enzymes. Although Res is known as an activator of the AMPK pathway, and previous studies have indicated the potential neuroprotective effects of Res both in vitro [35,36] and in vivo [37], the relationship between AMPK and Nrf2 was not elucidated. Therefore, combining cellular and animal data, we proposed for the first time that AMPK may regulate neuronal phase II antioxidative enzyme expression, which may contribute to reduced oxidative damage. Oxidative stress is usually defined as a state of imbalance favoring the factors that generate reactive oxygen species (e.g., superoxide or hydroxyl radicals) and working against the factors that protect cellular macromolecules from these reactants, including antioxidants. An increasing number of studies have revealed the involvement of oxidative stress in the pathogenesis of various diseases [38,39]. Our previous studies have shown that cognitive dysfunction was related to neuronal oxidative damage in the hippocampus, which may account for impaired spatial learning and memory in stressed offspring [8,32,33]. In the current study, we found that AMPK activation induced both mitochondrial biogenesis and phase II enzyme activation, accompanied by decreased protein oxidation and GSSG content in SH-SY-5Y cells. Although GSH content was unchanged by Res treatment, GSH/GSSG ratio exhibited a dosedependent increase by Res treatment. Consequently, maternal supplementation of Res may activate AMPK, induce mitochondrial biogenesis, and trigger antixodative enzymes in offspring. As a result, significant reduction in carbonyl protein levels, as well an increased GSH/GSSG ratio, was observed in Res-supplemented offspring. Eventually, Res efficiently inhibited the decreased expression of neuronrelated factors, indicating the promotion of neuron survival, and results from both the T maze and the Morris water maze showed significant improvement in cognitive function facilitated by supplementation of Res. Collective data from the current study have suggested a functional association between AMPK and oxidative stress through the modulation of mitochondrial content and the Nrf2 pathway. However, the details regarding the mechanism of AMPK’s modulation of Nrf2 were not elucidated in this study, and whether there is potential cross talk between PGC-1α-mediated mitochondrial biogenesis and Nrf2-mediated phase II enzyme activation should be investigated further. Our results showed that Res could activate AMPK pathways, and then may activate mitochondrial biogenesis and Nrf2 pathways in both male and female offspring, although in female offspring the difference between the Stress group and the Control group was not so significant compared with the male offspring.

165

The similar activation of AMPK, mitochondrial biogenesis, and Nrf2 pathways in both male and female offspring confirmed the cross talk among AMPK, mitochondrial biogenesis, and Nrf2 pathways. As for the difference, it has been reported that estrogen is involved in protecting neural function and regulating learning and memory mechanisms, and therefore may reduce the damage caused by prenatal stress [40,41]. Thus, female offspring may exhibit less sensitivity to prenatal stress. In conclusion, our study demonstrated for the first time that Res-mediated AMPK activation is closely associated with mitochondrial biogenesis and Nrf2 pathways in both SH-SY-5Y cells and PRS-induced rat offspring and that AMPK may work upstream of these two pathways simultaneously. This finding provides the evidence necessary to further explore the mechanism underlying the neuroprotective effects of Res, as well as the subsequent prevention of PRS-induced cognitive impairment, and the cross talk among energy homeostasis, mitochondrial dysfunction, and oxidative clearance. Understanding and rationally utilizing this functional relationship may be helpful in the establishment of effective therapies for neurological diseases.

Acknowledgments The authors are supported by the National Natural Science Foundation of China (81201023, 31370844, and 31070740), the 973 program for young scientists (No. 2014CB548200), National "Twelfth Five-Year" Plan for Science & Technology Support (2012BAH30F03), the Fundamental Research Funds for the Central Universities, and the 985 and 211 projects of Xi’an Jiaotong University.

Appendix A.

Supplementary Information

Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.freeradbiomed. 2014.07.029. References [1] Weinstock, M. Alterations induced by gestational stress in brain morphology and behaviour of the offspring. Prog. Neurobiol. 65:427–451; 2001. [2] O'Connor, T. G.; Heron, J.; Golding, J.; Glover, V. Maternal antenatal anxiety and behavioural/emotional problems in children: a test of a programming hypothesis. J. Child Psychol. Psychiatry 44:1025–1036; 2003. [3] Beydoun, H.; Saftlas, A. F. Physical and mental health outcomes of prenatal maternal stress in human and animal studies: a review of recent evidence. Paediatr. Perinat. Epidemiol. 22:438–466; 2008. [4] Bustamante, C.; Bilbao, P.; Contreras, W.; Martinez, M.; Mendoza, A.; Reyes, A.; Pascual, R. Effects of prenatal stress and exercise on dentate granule cells maturation and spatial memory in adolescent mice. Int. J. Dev. Neurosci. 28:605–609; 2010. [5] Lemaire, V.; Koehl, M.; Le Moal, M.; Abrous, D. N. Prenatal stress produces learning deficits associated with an inhibition of neurogenesis in the hippocampus. Proc. Natl. Acad. Sci. USA 97:11032–11037; 2000. [6] Zuena, A. R.; Mairesse, J.; Casolini, P.; Cinque, C.; Alema, G. S.; Morley-Fletcher, S.; Chiodi, V.; Spagnoli, L. G.; Gradini, R.; Catalani, A.; Nicoletti, F.; Maccari, S. Prenatal restraint stress generates two distinct behavioral and neurochemical profiles in male and female rats. PLoS One 3:e2170; 2008. [7] Zhu, Z.; Li, X.; Chen, W.; Zhao, Y.; Li, H.; Qing, C.; Jia, N.; Bai, Z.; Liu, J. Prenatal stress causes gender-dependent neuronal loss and oxidative stress in rat hippocampus. J. Neurosci. Res. 78:837–844; 2004. [8] Feng, Z.; Zou, X.; Jia, H.; Li, X.; Zhu, Z.; Liu, X.; Bucheli, P.; Ballevre, O.; Hou, Y.; Zhang, W.; Wang, J.; Chen, Y.; Liu, J. Maternal docosahexaenoic acid feeding protects against impairment of learning and memory and oxidative stress in prenatally stressed rats: possible role of neuronal mitochondria metabolism. Antioxid. Redox Signal. 16:275–289; 2012. [9] Ye, J.; Keller, J. N. Regulation of energy metabolism by inflammation: a feedback response in obesity and calorie restriction. Aging (Albany NY) 2:361–368; 2010. [10] Kim, J. H.; Lee, J. O.; Lee, S. K.; Kim, N.; You, G. Y.; Moon, J. W.; Sha, J.; Kim, S. J.; Park, S. H.; Kim, H. S. Celastrol suppresses breast cancer MCF-7 cell viability via

166

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

K. Cao et al. / Free Radical Biology and Medicine 75 (2014) 156–166

the AMP-activated protein kinase (AMPK)-induced p53-polo like kinase 2 (PLK-2) pathway. Cell Signal. 25:805–813; 2013. Li, Y. Y.; Yu, L. F.; Zhang, L. N.; Qiu, B. Y.; Su, M. B.; Wu, F.; Chen, D. K.; Pang, T.; Gu, M.; Zhang, W.; Ma, W. P.; Jiang, H. W.; Li, J. Y.; Nan, F. J.; Li, J. Novel smallmolecule AMPK activator orally exerts beneficial effects on diabetic db/db mice. Toxicol. Appl. Pharmacol. 273:325–334; 2013. Sasaki, H.; Asanuma, H.; Fujita, M.; Takahama, H.; Wakeno, M.; Ito, S.; Ogai, A.; Asakura, M.; Kim, J.; Minamino, T.; Takashima, S.; Sanada, S.; Sugimachi, M.; Komamura, K.; Mochizuki, N.; Kitakaze, M. Metformin prevents progression of heart failure in dogs: role of AMP-activated protein kinase. Circulation 119:2568–2577; 2009. Park, Y. J.; Ko, J. W.; Jang, Y.; Kwon, Y. H. Activation of AMP-activated protein kinase alleviates homocysteine-mediated neurotoxicity in SH-SY5Y cells. Neurochem. Res. 38:1561–1571; 2013. Liu, F.; Benashski, S. E.; Persky, R.; Xu, Y.; Li, J.; McCullough, L. D. Age-related changes in AMP-activated protein kinase after stroke. Age (Dordr.) 34:157–168; 2012. Yu, L.; Yang, S. J. AMP-activated protein kinase mediates activity-dependent regulation of peroxisome proliferator-activated receptor gamma coactivator1alpha and nuclear respiratory factor 1 expression in rat visual cortical neurons. Neuroscience 169:23–38; 2010. Vomhof-Dekrey, E. E.; Picklo Sr M. J. The Nrf2-antioxidant response element pathway: a target for regulating energy metabolism. J. Nutr. Biochem. 23:1201–1206; 2012. Onken, B.; Driscoll, M. Metformin induces a dietary restriction-like state and the oxidative stress response to extend C. elegans health span via AMPK, LKB1, and SKN-1. PLoS One 5:e8758; 2010. Liu, X. M.; Peyton, K. J.; Shebib, A. R.; Wang, H.; Korthuis, R. J.; Durante, W. Activation of AMPK stimulates heme oxygenase-1 gene expression and human endothelial cell survival. Am. J. Physiol. Heart Circ. Physiol 300:H84–H93; 2011. Mo, C.; Wang, L.; Zhang, J.; Numazawa, S.; Tang, H.; Tang, X.; Han, X.; Li, J.; Yang, M.; Wang, Z.; Wei, D.; Xiao, H. The crosstalk between Nrf2 and AMPK signal pathways is important for the anti-inflammatory effect of berberine in LPS-stimulated macrophages and endotoxin-shocked mice. Antioxid. Redox Signal. 20:574–588; 2014. Kim, M. Y.; Lim, J. H.; Youn, H. H.; Hong, Y. A.; Yang, K. S.; Park, H. S.; Chung, S.; Ko, S. H.; Shin, S. J.; Choi, B. S.; Kim, H. W.; Kim, Y. S.; Lee, J. H.; Chang, Y. S.; Park, C. W. Resveratrol prevents renal lipotoxicity and inhibits mesangial cell glucotoxicity in a manner dependent on the AMPK-SIRT1-PGC1alpha axis in db/db mice. Diabetologia 56:204–217; 2013. Tillu, D. V.; Melemedjian, O. K.; Asiedu, M. N.; Qu, N.; De Felice, M.; Dussor, G.; Price, T. J. Resveratrol engages AMPK to attenuate ERK and mTOR signaling in sensory neurons and inhibits incision-induced acute and chronic pain. Mol. Pain 8:5; 2012. Sahu, S. S.; Madhyastha, S.; Rao, G. M. Neuroprotective effect of resveratrol against prenatal stress induced cognitive impairment and possible involvement of Na(þ ), K(þ )-ATPase activity. Pharmacol. Biochem. Behav. 103:520–525; 2013. Madhyastha, S.; Sekhar, S.; Rao, G. Resveratrol improves postnatal hippocampal neurogenesis and brain derived neurotrophic factor in prenatally stressed rats. Int. J. Dev. Neurosci. 31:580–585; 2013. Madhyastha, S.; Sahu, S. S.; Rao, G. Resveratrol for prenatal-stress-induced oxidative damage in growing brain and its consequences on survival of neurons. J. Basic Clin. Physiol. Pharmacol. :1–10; 2013. Deacon, R. M.; Rawlins, J. N. P. T-maze alternation in the rodent. Nat. Protoc. 1:7–12; 2006.

[26] Fernandez, F.; Morishita, W.; Zuniga, E.; Nguyen, J.; Blank, M.; Malenka, R. C.; Garner, C. C. Pharmacotherapy for cognitive impairment in a mouse model of Down syndrome. Nat. Neurosci. 10:411–413; 2007. [27] Gué, M.; Bravard, A.; Meunier, J.; Veyrier, R.; Gaillet, S.; Recasens, M.; Maurice, T. Sex differences in learning deficits induced by prenatal stress in juvenile rats. Behav. Brain Res. 150:149–157; 2004. [28] Paintlia, A. S.; Paintlia, M. K.; Mohan, S.; Singh, A. K.; Singh, I. AMP-activated protein kinase signaling protects oligodendrocytes that restore central nervous system functions in an experimental autoimmune encephalomyelitis model. Am. J. Pathol. 183:526–541; 2013. [29] Frosig, C.; Jensen, T. E.; Jeppesen, J.; Pehmoller, C.; Treebak, J. T.; Maarbjerg, S. J.; Kristensen, J. M.; Sylow, L.; Alsted, T. J.; Schjerling, P.; Kiens, B.; Wojtaszewski, J. F.; Richter, E. A. AMPK and insulin action— responses to ageing and high fat diet. PLoS One 8:e62338; 2013. [30] Bishop, N. A.; Lu, T.; Yankner, B. A. Neural mechanisms of ageing and cognitive decline. Nature 464:529–535; 2010. [31] Gibson, G. E.; Starkov, A.; Blass, J. P.; Ratan, R. R.; Beal, M. F. Cause and consequence: mitochondrial dysfunction initiates and propagates neuronal dysfunction, neuronal death and behavioral abnormalities in age-associated neurodegenerative diseases. Biochim. Biophys. Acta 1802:122–134; 2010. [32] Liu, J.; Head, E.; Gharib, A. M.; Yuan, W.; Ingersoll, R. T.; Hagen, T. M.; Cotman, C. W.; Ames, B. N. Memory loss in old rats is associated with brain mitochondrial decay and RNA/DNA oxidation: partial reversal by feeding acetyl-L-carnitine and/or R-alpha -lipoic acid. Proc. Natl. Acad. Sci. USA 99:2356–2361; 2002. [33] Feng, Z.; Jia, H.; Li, X.; Bai, Z.; Liu, Z.; Sun, L.; Zhu, Z.; Bucheli, P.; Ballevre, O.; Wang, J.; Liu, J. A milk-based wolfberry preparation prevents prenatal stressinduced cognitive impairment of offspring rats, and inhibits oxidative damage and mitochondrial dysfunction in vitro. Neurochem. Res. 35:702–711; 2010. [34] Zhu, J.; Wang, K. Z.; Chu, C. T. After the banquet: mitochondrial biogenesis, mitophagy, and cell survival. Autophagy 9:1663–1676; 2013. [35] Kim, D. W.; Kim, Y. M.; Kang, S. D.; Han, Y. M.; Pae, H. O. Effects of resveratrol and trans-3,5,40 -trimethoxystilbene on glutamate-induced cytotoxicity, heme oxygenase-1, and sirtuin 1 in HT22 neuronal cells. Biomol. Ther. (Seoul) 20: 306–312; 2012. [36] Son, Y.; Byun, S. J.; Pae, H. O. Involvement of heme oxygenase-1 expression in neuroprotection by piceatannol, a natural analog and a metabolite of resveratrol, against glutamate-mediated oxidative injury in HT22 neuronal cells. Amino Acids 45:393–401; 2013. [37] Ren, J.; Fan, C.; Chen, N.; Huang, J.; Yang, Q. Resveratrol pretreatment attenuates cerebral ischemic injury by upregulating expression of transcription factor Nrf2 and HO-1 in rats. Neurochem. Res. 36:2352–2362; 2011. [38] Sims-Robinson, C.; Hur, J.; Hayes, J. M.; Dauch, J. R.; Keller, P. J.; Brooks, S. V.; Feldman, E. L. The role of oxidative stress in nervous system aging. PLoS One 8: e68011; 2013. [39] Yavari, A.; Azizova, G. I. Effect of oxidative stress on immunological parameters in type 2 diabetes mellitus in the Azerbaijan Republic. Diabetes Metab. Syndr 6:195–198; 2012. [40] Suzuki, S.; Brown, C. M.; Dela Cruz, C. D.; Yang, E.; Bridwell, D. A.; Wise, P. M. Timing of estrogen therapy after ovariectomy dictates the efficacy of its neuroprotective and antiinflammatory actions. Proc. Natl. Acad. Sci. USA 104:6013–6018; 2007. [41] Gulinello, M.; Lebesgue, D.; Jover-Mengual, T.; Zukin, R. S.; Etgen, A. M. Acute and chronic estradiol treatments reduce memory deficits induced by transient global ischemia in female rats. Horm. Behav. 49:246–260; 2006.