A total of 48 male C57BL/6 mice (13 ± 2 g, aged 3 weeks) were purchased from the Vital River Laboratory Animal Technology (Beijing, China) and group-housed in accordance with the Guide for the Care and Use of Laboratory Animals published by Ministry of Health of People's Republic of China. All experimental procedures and protocols were reviewed and approved by Life Science Ethics Review Committee of Zhengzhou University.

Lead acetate [Pb(Ac)₂] with purity of ≥99.5% was purchased from Aladdin Bio-Chem Technology (Shanghai, China). Resveratrol was obtained from Sigma Aldrich (St. Louis, MO, USA). Pb(Ac)₂ was dissolved into deionized water at a concentration of 0.2%. Resveratrol was diluted in 5% carboxymethyl cellulose sodium (CMC-Na). GSH-Px, MDA and GSH assay kits were purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). ELISA kit for Aβ₁₋₋₄₀ was obtained from Biosource International (Camarillo, USA). Antibody against BDNF, TrkB, phosphorylated TrkB (pTrkB, Y515), β-actin and Lamin B was purchased from Abcam Company (Cambrige, UK). Antibodies against 8-hydroxy-2-deoxyguanosine (8-OHdG) and phosphorylated SIRT1(pSIRT1, Ser 47) was obtained from Beijing Biosynthesis Biotechnology (Beijing, China) while SIRT1, AMPK and phosphorylated AMPK (pAMPK, Thr 172) were purchased from Cell Signaling Technology (Danvas, MA, USA).

After 1 week of normal feeding to allow for accommodation, 4-week old mice were divided into four groups of 12 mice each: Control group (Cont), Pb-exposed group (Pb), Pb + Resveratrol group (PbR), Control + Resveratrol group (CR). For Pb group, lead exposure was induced by providing a free drink of 0.2% Pb(Ac)₂ solution for 12 weeks. For CR group, mice were housed until the age of 16 weeks, and then were gavaged with 50 mg/kgBw•d of resveratrol every other day for another 48 weeks. Both Cont group and Pb group were also administered with the same volume of CMC-Na starting at 16 weeks old. For PbR group, mice were similarly treated with Pb(Ac)₂ but subject to a 48-week oral gavages' supplementation with resveratrol at the dose of 50 mg/kgBw•d. A timeline with our experimental design can be seen in Figure 1.

The mice were subjected to assess spatial learning in the MWM test at the age of 64 weeks, and the behavioral test was performed double blinded according to the standard protocol. Water was made opaque by the addition of 200 g of powdered milk. During the platform trial days, each mouse was trained in the water maze and allowed a maximum of 60 s to find a hidden platform for 5 consecutive days, 4 trials per day with a 30-s interval. On each trial, the mice started from one of the middle of the four quadrants (N, E, S, W) facing the wall of the pool. Escape latencies and swim patterns were digitally monitored. On the 6th day, the probe test was performed, in which the mice were subjected to a 60-s free swim to find the previous location of the platform starting from each of the four quadrants. The pathway that the mice passed through the previous platform quadrant was recorded by a video camera which was connected to a digital-tracking device attached to the computer loaded with the water maze software (Zhenghua Biologic Apparatus Facilities, Huaibei, China).

At the end of the behavioral test, the mice were sacrificed with 10% chloral hydrate and brains were removed and rinsed with precooling isotonic saline. Three perfused brain samples from each group were fixed with 4% paraformaldehyde-PBS solution (pH 7.4) at 4°C overnight for further analysis. The remaining hippocampus and cortex were quickly separated and the tissues were stored at −80°C until analysis.

Small portions of cortical tissue were carefully dissected and homogenized with ice-cold buffer (50 mM Tris-HCl buffer, pH7.4) to prepare a 10% homogenate solution. The homogenate was centrifuged (10,000 g, 15 min) and aliquots of supernatant were separated for estimation the activities GSH-Px and catalase (CAT) as well as the levels of MDA and GSH using assay kits. All the operations followed the protocols provided in the kits.

For brain Aβ₁₋₋₄₀, 60 mg of fresh-frozen mouse cortex was serially homogenized into detergent-soluble and guanidine HCl-soluble fractions. All samples were assayed for Aβ₁₋₋₄₀ using a commercially available ELISA assay kit according to the manufacturer's instructions.

Paraffin histological sections (5 μm thick) with an average distance of 5 μm apart were obtained from each brain sample. The sections were rinsed and incubated with permeabilization solution, followed by blocking with 10% normal goat serum. IHF staining was performed using primary antibody (rabbit polyclonal antibody 1:400 dilution, at 4°C, overnight) for 8-OHdG or SIRT1, followed by the addition of anti-rabbit Alexa Fluor 488 FITC (Molecular Probes) secondary antibody. The sections were counterstained with 4', 6-Diamidino-2-phenylindole (DAPI, dilution 1:500, Sigma) to identify cellular nuclei. Additionally, mouse monoclonal antibody (dilution 1:200 at 4°C overnight) was used as primary antibody for pSIRT1, followed by the addition of Cy3 goat anti-mouse IgG (Abbkine, CA, USA) at a 1:200 dilution for 2 h at 37°C in the dark. Next, the sections were washed with PBS three times and fluorescence quencher was applied. Subsequently, the images (three images per section) were acquired by using the fluorescent microscope system (Olympus, FV1000, Tokyo, Japan). For negative control experiments, the primary antibodies were omitted. In addition, three sections per brain were used for analyzing the mean fluorescence value of 8-OHdG-labeled cells by Image-Pro Plus software (IPP 6.0) in mice brain from five views under the fluorescence microscope (Nikon, Japan).

The total RNA from hippocampus and cortex tissues prepared in advance was extracted using TRIZOL reagent (Invitrogen, USA). Then 2 μg of RNA of each sample was transcribed to cDNA using the TaKaRa RNA PCR™ kit (Takara, Japan). The mRNA expression of Bdnf, Sirt1, App, Bace1 and Gadph was inspected using cDNA as a template for amplification utilizing the following primers: Bdnf IV (forward primer 5'-CAGAGCAGCTGCCTTGATGTT-3' and reverse primer 5'-GCCTTGTCCGTGGACGTTTA-3'); Sirt1 (forward primer 5'-GAGGTCTCGA TATGTGCTGGA-3' and reverse primer 5'-TTCCTGCAACCTGCTCCAAG-3'); App (forward primer 5'-GACTGACCACTCGACCAGGTTCTG-3' and reverse primer 5'-CTTGTAAGTTGGATTCTCATATCCG-3'); Bace1 (forward primer 5'-CCTACACCCAGGGCAAGT-3' and reverse primer 5'-GGGCAGTAGTAAC TTTGCAGT-3'); and Gapdh (forward primer 5'-CTTCCAGGAGCGAGACC-3' and reverse primer 5'-CGGAGATGATGACCCTTTT-3'). The expression levels of each sample were normalized against Gadph (internal control) and calculated using the comparative CT method (2^−ΔΔCT).

Immunochemical analysis of protein level and phosphorylation status was performed by Western blotting in standard conditions. Briefly, equal amounts of samples were separated by 10~12% SDS-PAGE and transferred to PVDF membranes. The membranes were blocked in Tris-buffered saline mixed with Tween-20 (TBST, pH 7.4) containing 5% skim milk for 60 min. Then the membranes were incubated with primary antibodies for desired proteins in blocking buffer at 4°C overnight. β-Actin, β-Tubulin and Lamin B were used as loading control for total, cytosolic and nuclear proteins, respectively. For densitometric analysis, the blots were scanned with Amersham Imager 600 and the pixel intensities of each band of interest were quantified using Image Quant TL (7.0 version, GE Healthcare, USA). Immunoblots shown are representative of three independent biological replicates.

All statistical analyses were performed using the SPSS software (version 17.0) and GraphPad Prism 5 for Windows was used for graph fitting. The escape latency data in the MWM test were analyzed by a two-way analysis of variance (ANOVA). Probe trail and other data among multiple groups were determined using one-way ANOVA followed by Tukey's post-hoc test. The correlation between the two variables was analyzed by line correlation analysis. All data in the text and figures were expressed as mean ± standard error of the mean (SEM), with n representing the number of animals used in each experiment. Statistical significance was defined at the level of P < 0.05.

MWM test was performed to evaluate the effects of resveratrol gavage on impairment of spatial learning and memory in Pb-treated mice and the results were summarized in Figure 2. There was a significant difference in mean latency between training days[F_{(4, 220)} = 29.47, P < 0.001] and between treatments [F_{(3, 220)} = 4.72, P < 0.05], but there was no interaction between the factors day and treatment [F_{(12, 220)} = 0.38, P > 0.05].

The Pb-group mice performed longer latency to reach the hidden platform on day 4 and 5 of acquisition period (P < 0.01, Figure 2A) and cumulative path length of 5th day (P < 0.001, Figure 2B) compared to control group. Meanwhile, the mice subjected to resveratrol treatment showed a significant decrease in the latency (P < 0.05) and the path length (P < 0.01) relative to Pb group on day 5.

On day 6 of probe trail, time spent in target quadrant was remarkably reduced in Pb-treated group when compared to the control group (P < 0.001, Figure 2C), whereas the time to cross the first non-exit platform was prolonged (P < 0.01, Figure 2D). However, these indexes were significantly improved in mice that received resveratrol compared to those that only received Pb (P < 0.05 respectively, Figures 2C,D). PbR group had effects similar to those of control group. No significant difference was found in crossing numbers and swimming speed between all groups (P > 0.05 respectively, Figures 2E,F).

To confirm our findings from MWM test, we further measured the level of hippocampal BDNF. As shown in Figure 3, RT-qPCR results demonstrated that the Bdnf expression at transcription level had an appreciable reduction after Pb exposure while treatment with resveratrol substantially promoted Bdnf expression (P < 0.05 respectively, n = 6, Figure 3A). The level of mature BDNF (mBDNF) protein by Western blotting had a similar trend (P < 0.05 respectively, Figures 3B,C). In comparison, the level of proBDNF in both PbR and CR group gradually increased compared to control (P < 0.001 respectively, Figure 3C), but it did not display a marked increase in the mice of Pb group. The activation of TrkB, known as cognate receptor of BDNF, was also determined by western blotting (Figure 3B). Exposure to Pb resulted in significant declines in pTrkB/TrkB ratio in the hippocampus when compared to control mice (P < 0.001, n = 6, Figure 3D). Additionally, resveratrol demonstrated the ability to normalize the decline of TrkB activation induced by Pb (P < 0.001).

Substantial evidences indicate oxidative stress is lightly linked with neurotoxic action of Pb. Hence, we firstly assessed cortical oxidation by measuring pro-oxidant and anti-oxidant biomarker profiles. As shown in Figure 4A, the mice displayed an increase in MDA after Pb exposure (P < 0.01, n = 8). Meanwhile, there was a significant reduction in cortical activities of GSH-Px and CAT as well as GSH content in Pb group compared to control group (P < 0.001 for GSH-Px and GSH, P < 0.01 for CAT, Figures 4B–D). In comparison, resveratrol administration led to reversal of high level of MDA and reduction of the above anti-oxidant biomarkers induced by Pb. As a marker of oxidative damage to DNA, 8-OHdG immunoreactivity was also detected and representative images for quantitative analysis of fluorescence value in the sub-region of brain were shown in Figure 5A. After Pb exposure, there were significant increases in the 8-OHdG content in the cortex as well as CA1 and CA3 while fail to alter its content in the DG region (P < 0.001 in cortex, P < 0.05 in CA1 and DG, P < 0.01 in CA3, Figure 5B). Treatment with resveratrol partly depressed the enhancement of 8-OHdG content in the cortex and hippocampus (P < 0.05 in cortex, P < 0.01 in CA1 and CA3).Taken together, the results indicated that Pb-induced oxidant damage were partly restored following resveratrol treatment.

Given that resveratrol was reported to be a potential activator of SIRT1 and could protect mice against Pb-induced oxidative stress, we detected the involvement of SIRT1 in mounting the protective effect of chronic resveratrol using Western blotting and RT-qPCR. As shown in Figure 6, the level of SIRT1 in nucleus was found reduced in Pb group compared to control group (P < 0.001, n = 6, Figures 6A,B). The rate of SIRT1 level in nucleus/cytosol had a significant difference (P < 0.001, Figure 6E) while no alternation of total SIRT1 protein and mRNA level between two groups (Figures 6D,F). In contrast, chronic resveratrol treatment could induce SIRT1 expression in the cell nuclei of PbR group mice (P < 0.05) indicating that resveratrol can repress the nucleo-cytoplasmic translocation of SIRT1 induced by Pb exposure (Figures 6A,G).

Previous studies suggested that phosphorylation of human SIRT1 can affect its subcellular distribution. In view of this, we further investigated whether phosphorylation of SIRT1 was altered during chronic Pb-induced nuclear SIRT1 by Western blotting. As shown in Figure 7, the phosphorylated SIRT1 at Ser47 site was found to be down-regulated in Pb group compared to control or PbR group (P < 0.001 or P < 0.01, Figure 7A). Additionally, we analyzed the effects of Pb and resveratrol on the phenotypic expression of hippocampal SIRT1 phosphorylation using IHF staining. As shown in Figure 7B, decreased pSIRT1 expression was found in Pb group rather than control group and the presence of resveratrol can induce upregulation of pSIRT1.

We further assessed the relationship between pSIRT1 and spatial learning and memory performance by correlation analysis conducted with GraphPad Prism 5. As shown in Figure 7C, pSIRT1 have positive correlation with probe time in target area. In addition, the hippocampus pSIRT1 expression was significantly and negatively related to cortical MDA content by correlation analysis (Figure 7D).

Some evidence has shown that SIRT1 is a crucial regulator of AMPK activation in mitochondrial function, thus we explored whether AMPK was involved in Pb-induced SIRT1 alteration. As shown in Figure 8, compared to control mice, the ratio of pAMPK/AMPK was declined in mice of Pb group(P < 0.001, Figures 8A,B). In contrast, resveratrol partly restored the Pb-induced reduction of AMPK phosphorylation in the hippocampus (P < 0.05, n = 6). Moreover, the expression of nuclear PGC-1α was decreased in mice of Pb group (P < 0.01, Figure 8C), whereas no change were determined in total PGC-1α protein. Also, a recuperative increase of nuclear PGC-1α in PbR group was observed (P < 0.01, n = 6). In addition, the level of pAMPK/ AMPK had positive correlation with probe time in target area by correlation analysis (Figure 8D).

Aβ is known to play a critical pathogen role in AD and is involved in Pb-induced cognitive impairment; we therefore measured Aβ₁₋₋₄₀ in the cortex by ELISA. As shown in Figure 9A, the Aβ₁₋₋₄₀ level of mice from Pb group was significantly higher than that of control group (P < 0.01, n = 8). In contrast, resveratrol could suppress Aβ₁₋₋₄₀ overproduction when compared to the Pb group(P < 0.05). Furthermore, there was a negative correlation of cortical Aβ₁₋₋₄₀ content with probe time in target area of Pb-exposed mice(P < 0.01, Figure 9B). As shown in Figures 9C–E, chronic Pb exposure increased the mRNA of App (P < 0.01, n = 8) and protein level of BACE1 (P < 0.001, n = 8) whereas no change was determined in mRNA of Bace1. In contrast, resveratrol demonstrated the ability to normalize the increase of BACE1 level following Pb exposure (P < 0.001, Figure 9E). The data suggested that resveratrol could partly abolish abnormal production of Aβ₁₋₋₄₀ and the amyloidogenic processing induced by Pb.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.