Introduction

The pathognomonic signs of Alzheimer’s disease (AD) are extracellular senile plaques with amyloid-beta (Aβ) peptide deposits and intracellular neurofibrillary tangles with hyperphosphorylated microtubule-associated tau-protein aggregates which cause neuronal stress and neuronal circuit disruption, eventually culminating in neurodegeneration and brain atrophy (Praticò 2013). Although the “primum movens” of AD is remaining obscure, many reports underscore the causal link between perturbed neuronal death-survival equilibrium and neuro-behavioral dysfunction in AD (Fiocchetti et al. 2013; Godoy et al. 2014).

The “Aβ oligomer hypothesis” has supplanted the “amyloid cascade hypothesis” as the former seems to fill the lacuna in correlating amyloid plaque deposits and memory impairment or neuronal injury (Benilova et al. 2012). According to the widely accepted “Aβ oligomer hypothesis”, soluble neurotoxic Aβ oligomers play a pivotal role in AD pathogenesis and the Aβ oligomers accelerate a plethora of noxious cellular and sub-cellular events in the brain including calcium dyshomeostasis, excitotoxicity, neuroinflammation, ER stress, cytoskeletal collapse, mitochondrial damage and fragmentation, neuro-synaptic dysfunction, and accelerated apoptosis—ultimately leading to nerve cell demise and cognitive dysfunction (Benilova et al. 2012; Kayed and Lasagna-Reeves 2013).

Most researchers advocate that extracellular signal-regulated kinase-mitogen-activated protein kinase (ERK-MAPK), phosphoinositide-3-kinase/Akt (PI3K/Akt) and Janus kinase 2/signal transducer and activator of transcription 3 (JAK2/STAT3) signaling are the crucial survival pathways impaired in the Alzheimer’s disease (Kawamata and Shimohama 2011; Marwarha and Ghribi 2012). Thus, activation of these pathways might rescue the AD brain from learning and memory impairments. Recently, Denner et al. (2012) reported that hippocampal ERK2 is a central regulator of learning and memory functions. This study is in sync with the report of Satoh et al. (2007) on ERK2 knock-down mice, which displayed long-term memory deficits. Besides, there are substantial reports on the amyloid β-induced impairment of PI3K/Akt1 and JAK2/STAT3 axes which in turn caused hippocampus-linked memory dysfunctions (Chiba et al. 2009; Liao and Xu 2009).

Based on the recent reviews and research studies on both clinical and experimental AD models, we conjecture that synchronized regulation of various neuromolecular switches—involved in the neuronal apoptosis and survival signaling pathways—might confer protection against the pathogenesis of beta-amyloid-induced brain cell damage and memory deficiency (Fiocchetti et al. 2013; Godoy et al. 2014; Nehlig 2013; Tan et al. 2012). Thus, an exemplar anti-Alzheimer’s drug must possess dual multipotency to counteract the toxic episodes and enhance the declined neuro-behavioral functions in AD.

3,4-Dihydroxyphenylethanol (DOPET; hydroxytyrosol) is a biophenol present in olive oil, wine, grape juice and notably, it is an endogenous metabolite of dopamine (de la Torre et al. 2006). Beyond its high safety index (Auñon-Calles et al. 2013), DOPET has been reported to exhibit central and peripheral neuroprotective (González-Correa et al. 2008; Ristagno et al. 2012), cardioprotective (Granados-Principal et al. 2014), uroprotective (Rouissi et al. 2011), nephroprotective (Capasso et al. 2008), hepatoprotective (Pan et al. 2013), anti-diabetic and anti-obesity (Cao et al. 2014), anti-osteoporotic (Hagiwara et al. 2011), anti-inflammatory (de la Puerta et al. 1999), anti-atherosclerotic (González-Santiago et al. 2006), anti-cancer (Zhao et al. 2014), and anti-HIV effects (Lee-Huang et al. 2007). Fascinatingly, St-Laurent-Thibault et al. (2011) showed that DOPET offers protection against amyloid β-induced neurotoxicity in N2a cells. Moreover, DOPET was reported to exert Tau anti-fibrillization effect in vitro (Daccache et al. 2011). With this backdrop of research reports, we hypothesized that DOPET may be a putative multipotent anti-Alzheimer’s agent. As far as we are aware, this is the first report to elucidate the molecular mechanisms involved in the neuroprotective and cognition-enhancing effect of DOPET in an in vivo AD model.

Materials and Methods

Drugs and Animals

DOPET (≥98 %) (Fig. 1a) and synthetic Aβ-(1–42) peptide (Fig. 1b) were purchased from Extrasynthese (Genay Cedex, France) and American Peptide Co., (Sunnyvale, CA) respectively. Ibotenic acid (Fig. 1c) was purchased from Abcam (Cambridge, UK). Thirty adult male C57BL/6 mice, aged 6–8 weeks and weighing 20–25 g, were housed in temperature- and humidity-controlled animal quarters under 12-h light/12-h dark cycles. The light/dark cycle corresponded to the timing of lights on at 07:00 a.m. (IST) and to the timing of lights off at 07:00 p.m. (IST), respectively. The animals had access to food and water ad libitum. Each mouse was handled for 5 min/day on three consecutive days, starting 2 days after arrival, before being subjected to the study. All experiments were conducted in accordance with animal care guidelines approved by the Institutional Animal Ethics Committee (IAEC) of the Vels University.

Fig. 1
figure 1

Structure/amino acid sequence of compounds

Full size image

Surgery and Treatment Protocol

Neurodegeneration was induced in mice, based on the model proposed by Morimoto et al. (1998) with minor modifications. Briefly, mice were deeply anesthetized by injecting ketamine (100 mg/kg, i.p.) and xylazine (10 mg/kg, i.p.) and the surgical site was shaved and sterilized. An incision roughly 1.5 cm in length was made on the cranium to reveal the bregma. On day I (by 7 a.m.), each anesthetized mouse was mounted in a stereotaxic apparatus and one microliter of each neurotoxicant solution (oA42i; soluble oAβ1–42, and ibotenic acid-in-PBS) was injected over 5 min through a 26-gauge Hamilton syringe into the lateral ventricle [−0.6 mm anterior–posterior (AP), 1.2 mm medial–lateral (ML), and −2 mm dorsal–ventral (DV)], relative to the bregma (Paxinos and Franklin 2001). Mice in the sham-operated group were administered with the same volume of sterile saline in the similar fashion. The needle was slowly withdrawn from the site of injection only after 5 min to prevent any backflow. One hour after the i.c.v. injection of sterile saline or oA42i (oligomeric Aβ1–42 plus ibotenic acid), all the thirty mice were randomly allocated to one of the three groups and treated as follows (Fig. 2): saline-injected sham-operated group, oA42i-injected group, and oA42i-injected DOPET treatment group. After 4 days of recovery period (day I-IV), the animals in DOPET group were administered sub-chronically with DOPET [10 mg/(kg day), p.o.] for 14 days by oral gavage, regularly by 7 a.m. Also, 6 h after DOPET administration, animals in all the three groups were subjected to RAM exposure from day 1 to day 14 as portrayed in Fig. 2.

Fig. 2
figure 2

Study design

Full size image

Preparation of Aβ Oligomers

Amyloid beta oligomers were prepared from Aβ1–42 peptide as described by Klein (2002). In brief, the Aβ1–42 peptide, previously chilled on ice, was solubilized in ice-cold hexafluoroisopropanol (HFIP, Sigma–Aldrich Pvt. Ltd., India) to disintegrate any pre-existing aggregates and a final concentration of 1 mM was made. Then, the Aβ-HFIP solution was incubated at room temperature for 1 h in a closed vial and again the solution was placed back on ice for about 10 min. The solution was vortexed and aliquoted into Eppendorf microcentrifuge tubes and left overnight in the hood at room temperature to allow HFIP evaporation. After lyophilization, the obtained dry monomeric films of Aβ1–42 peptide were stored over desiccant at −80 °C. Aβ monomer preparations were prepared in situ to curtail potential aggregation of peptide. For the oligomer preparation, the dried peptide was re-suspended in anhydrous dimethyl sulfoxide (DMSO, Merck India Ltd., India) to make a 5 mM stock. The resultant solution was additionally diluted to 100 µM in phenol red-free Ham’s F-12 medium, immediately vortexed for 30 s, and incubated for 24 h at 5 °C. The preparation was centrifuged at 14,000×g for 10 min at 4 °C to pellet out insoluble aggregates (protofibrils and fibrils), and the supernatants containing soluble neurotoxic Aβ-oligomers (also called ADDLs; Aβ-derived diffusible ligands) were transferred to a new tube and stored at 4 °C for further use.

Behavioral Assessment

Apparatus

Spatial reference and working memories were assessed using a radial arm maze (RAM), as reported by Brown et al. (2002) with minor modifications. Briefly, the apparatus was designed with eight identical and equally spaced arms (25 cm long × 8.5 cm wide) radiating from a central octagonal platform (25 cm in diameter) with a recessed food cup (1 cm from the end and 1.5 cm deep). The RAM apparatus was elevated to a height of 30 cm from the ground and placed in a small, well-lit room that contained a number of visual cues. At the entrance to each arm was a door that could be controlled using strings.

Habituation Phase

The behavioral assessment was carried out as shown in Fig. 2. Every exposure of a mouse to the maze at any phase is designated as a trial. During habituation phase, food baits (Kellogg’s Chocos Duet® Snacks) were scattered throughout the maze to encourage exploration and to lessen fear and anxiety of mice from exposure to the strange maze setting. The residence time of the mice was restricted to 5 min during each exploration and there were two trials on each day. Throughout the behavioral assessment, the maze was cleaned using paper towels dampened in ethanol (70 %) to remove the traces of urine and feces and to exclude any olfactory cue, at the end of every trial.

Training Phase

During training period, only single food bait was placed down in each of the 8 arms of RAM. Mice were released on the central platform and allowed to travel around all the 8 arms. Mice remained on the maze for 5 min during each training trial. The observations in both habituation and training phases were not recorded as these phases are mainly meant to lessen the experimental errors in testing phase.

Testing Phase

In the testing phase, mice were tested for the spatial reference and working memory performances for 10 consecutive days with one trial per day. Four arms (2, 3, 5, and 7) were randomly baited with food reward so that no more than one pair of adjacent arms was baited or unbaited. This paradigm was maintained steadily till the end of the assessment. The trial was terminated when all of the baits were consumed or 5 min had passed, whichever was earlier.

Error Recording

Three types of errors were scored: reference memory errors (RME), defined as the number of first entries into an unbaited arm; “correct” working memory errors (CWME), defined as the number of reentries into a baited arm and “incorrect” working memory errors (IWME), defined as the number of reentries into an unbaited arm (Schmitt et al. 2003).

Tissue Harvesting

On the 15th day, all the mice were sacrificed by cervical dislocation, their brains were rapidly excised, and the hippocampi were dissected out on ice, then snap frozen in liquid nitrogen, and kept at −80 °C until analyzed.

Western Blot Analysis

The hippocampi were chopped into small pieces and about 5 mg of the tissue was transferred to microcentrifuge tubes. Then, 300 µl of the cell lysis buffer—cold RIPA (radio-immunoprecipitation assay) buffer (Abcam, Cambridge, UK) containing 0.22 % beta glycerophosphate, 10 % tergitol-NP40, 0.18 % sodium orthovanadate, 5 % sodium deoxycholate, 0.38 % EGTA, 1 % SDS, 6.1 % Tris, 0.29 % EDTA, 8.8 % sodium chloride, 1.12 % sodium pyrophosphate decahydrate (pH 7.5)—was added to the tubes and homogenized with an electric homogenizer and the blades were rinsed twice with the same volume of lysis buffer. The lysate was placed in an orbital shaker for 2 h at 4 °C. Then, the solubilized proteins were collected from the supernatant after centrifugation at 16,000×g for 20 min at 4 °C and the debris was discarded. A small volume (50 μl) of the lysate was taken to determine protein concentration using protein assay reagent (Bio-Rad Laboratories, Hercules, CA). Equal amount of proteins (30 μg) were separated on SDS–polyacrylamide gels with a Tris–glycine/SDS running buffer system and then transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, Billerica, MA). After blocking in 5 % non-fat milk in Tris-buffered saline containing 0.1 % Tween-20 (TBST), the membranes were incubated overnight at 4 °C with specific primary antibodies: anti-Akt1 and anti-phospho-Ser473-Akt1 (1:1,000, Cell Signaling Technology, Inc. Danvers, MA), anti-ERK1/2 and anti-phospho-Thr202/Tyr204-ERK1/2 (1:1,000, Santa Cruz Biotechnology, Inc., CA), anti-RSK2 and anti-phospho-Thr577-RSK2 (1:1,000, Santa Cruz Biotechnology, Inc., CA), anti-JAK2 and anti-phospho-Tyr1007/1008-JAK2 (1:1,000, Cell Signaling Technology, Inc. Danvers, MA), anti-STAT3 and anti-phospho-Tyr705-STAT3 (1:1,000, Cell Signaling Technology, Inc. Danvers, MA), anti-JNK and anti-phospho-Thr183-JNK (1:1,000, Santa Cruz Biotechnology, Inc., CA), anti-p38 and anti-phospho-Thr180/Tyr182-p38 (1: 2,000, Cell Signaling Technology, Inc. Danvers, MA), anti-CREB and anti-phospho-Ser133-CREB (1:1,000, Millipore, Billerica, MA), anti-SIRT1 (1:1,000, Santa Cruz Biotechnology, Inc., CA) and anti-ADAM10 (1:1,000, Abcam, Cambridge, UK). After washing, the membranes were then incubated with secondary horseradish peroxidase-conjugated antibodies (Santa Cruz Biotechnology, Inc., CA) and visualized with chemiluminescence reagents provided with the ECL (enhanced chemiluminescent) kit (Abcam, Cambridge, UK). The intensity of chemiluminescent bands was measured using Quantity One® software (Bio-Rad).

RNA Isolation and RT-PCR Assay

For the RT-PCR assay, total RNA was extracted by using RNeasy® total RNA isolation kit (Qiagen, Valencia, CA). Briefly, 10 mg of the chopped hippocampal tissues was disrupted and homogenized by using lysis buffer (350 μl) and the lysate was centrifuged at 16,000×g for 3 min and processed further to yield 10 μg RNA. The cDNA was synthesized from RNA using the SuperScript® One-Step RT-PCR kit (Invitrogen). The Thermal Cycler 2400 (Perkin Elmer) was programed so that cDNA synthesis follows immediately with PCR amplification, automatically. The amplified products were separated by electrophoresis on a 2 % agarose gel containing ethidium bromide and photographed. Band intensities were determined by an image analysis system (Alpha Imager gel documentation system, CA). The following are the primers used in the RT-PCR assay: β-actin: sense 5′-TCTTGGGTATGGAATCCTGTG-3′ and antisense 5′-ATCTCCTTCTGCATCCTGTCA-3′; cytochrome c: sense 5′-ATAGGGGCATGTCACCTCAAAC-3′ and antisense 5′-GTGGTTAGCCATGACCTGAAAG-3′; Caspase-9: sense 5′-GTGGACATTGGTTCTGGC-3′ and antisense 5′-GTTGATGATGAGGCAGTGG-3′; Apaf-1: sense 5′-TTGATGCTGTCATTATGTAGGC-3′ and antisense 5′-AGGTAAAAGGGGAAGTATGTGTT-3′; Caspase 3: sense 5′-TCTTCATCATTCAGGCCTG-3′ and antisense 5′-TGAATTTCTCCAGGAATAGTAACC-3′; Bcl-2: sense 5′-CTGGGATGCCTTTGTGGAAC-3′ and antisense 5′-TCAAACAGAGGTCGCATGCT-3′; Bad: sense 5′-CCAGTGATCTTCTGCTCCACATCCC-3′ and antisense 5′-CAACTTAGCACAGGCACCCGAGGG-3′; BDNF: sense 5′-AGGCACTGGAACTCGCAATG-3′ and antisense 5′-AAGGGCCCGAACATACGATT-3′; c-Fos: sense 5′-GATGTTCTCGGGTTTCAACG-3′ and antisense 5′-GGAGAAGGAGTCGGCTGG-3′; Nurr1: sense 5′-AGTACCTTTATGGACAACTACAGCA-3′ and antisense 5′-CGTAGTGGCCACGTAGTTCTGGT-3′; Egr1: sense 5′-CCCTGTGCCCACTTCCTACTCCT-3′ and antisense 5′-CGGCCATCTCTTCCCTCCTGT-3′.

Transmission Electron Microscopy (TEM)

The TEM study was performed according to the method of Hajibagheri (1999). Briefly, the hippocampi were chopped and fixed with 2.5 % glutaraldehyde in 0.1 M cacodylate buffer. The fixed tissues were post-fixed in 1 % osmium tetroxide for 1 h at RT and then dehydrated in an ethanol series. Then, the slices were infiltrated and embedded in epoxy resins. Ultrathin sections of 800 Å were cut using ultramicrotome (LKB Ultratome, NOVA). The grids containing sections were stained with 2 % uranyl acetate followed by Reynolds lead citrate. Grids were examined and photographed under JEM-2010 transmission electron microscope (JEOL Ltd, Tokyo, Japan).

Statistical Analysis

The data were represented as the mean ± standard deviation (SD). Two-way analysis of variance (ANOVA) was applied for behavioral assessment (RAM tasks). One-way ANOVA was applied for biochemical evaluations. In this study, Tukey’s post-hoc test was used as a measure of between-group significance. All analyses were performed using the Statistical Package for the Social Sciences (SPSS) software (Version 13.0; SPSS, Inc., Chicago, IL, USA). P value <0.05 was considered statistically significant.

Results

Effect of Sub-chronic DOPET Treatment on oA42i-Induced Spatial Cognitive Decrements in Mice

To explore the potential of DOPET in reversing the spatial reference and working memory impairments in oA42i-induced cognitive dysfunction, the mice were treated with DOPET and examined by using RAM. Mice in the oA42i group showed significantly (P < 0.05) increased memory errors (RME—2.1-fold; correct working memory error—1.6 fold; incorrect working memory error—1.9-fold) in locating the food bait, when compared to the sham-operated group (Fig. 3a–c). We observed a significant (P < 0.05) improvement in the correct choice and error reduction in the DOPET treated mice, compared to the oA42i group (Fig. 3a–c).

Fig. 3
figure 3

Effect of DOPET on the spatial reference and working memory performances, after injecting oA42i (i.c.v.) in mice. a RME calculated as a function of day/group; b CWME calculated as a function of day/group; c IWME calculated as a function of day/group. Data were expressed as mean ± SD (n = 10 per group). *P < 0.05 (oA42i vs control); # P < 0.05 (DOPET plus oA42i vs oA42i alone)

Full size image

Effect of DOPET Administration on the Upstream MAPK Signaling Pathways in oA42i-Induced Neurotoxicity in Mice

MAPK signaling is one of the most studied molecular mechanisms in the field of Alzheimer’s disease research, due to the critical role of these pathways in regulating cytoprotective/cytotoxic processes. In our study, oA42i-intoxicated mice depicted about 1.6- and 1.7-fold decrease in the relative protein levels of p-ERK1/2/ERK1/2 and p-RSK2/RSK2, respectively, compared to the sham-operated group (P < 0.05) (Fig. 4a–c). On assessment of the death signaling kinases, JNK and p38, we found that the relative protein levels of p-JNK/JNK and p-p38/p38 were significantly increased to about 2.1- and 2.7-fold, respectively, in the oA42i group, when compared to the sham-operated group (Fig. 4d–f). In contrary, DOPET treated mice depicted about 51 % and 57 % increase and about 42 % and 43 % decrease in the relative phosphorylated/total protein levels of ERK1/2, RSK2, JNK, and p38, respectively, compared to the oA42i group (P < 0.05).

Fig. 4
figure 4

Effect of DOPET on the relative phosphorylated/total MAPK protein levels in the hippocampi, 14 days after injecting oA42i (i.c.v.) in mice. a Representative Western blots of phosphorylated- and total-ERK1/2 and RSK2 expressions. b Relative protein level of p-ERK1/2 (Thr202/Tyr204)/ERK1/2. c Relative protein level of p-RSK2 (Thr577)/RSK2. d Representative Western blots of phosphorylated- and total-JNK and -p38 expressions. e Relative protein level of p-JNK (Thr183)/JNK. f Relative protein level of p-p38 (Thr180/Tyr182)/p38. Data were expressed as mean ± SD from six independent experiments with the similar results. *P < 0.05 (oA42i vs control); # P < 0.05 (DOPET plus oA42i vs oA42i alone)

Full size image

Effect of DOPET Administration on the Upstream PI3K/Akt1 and JAK2/STAT3 Signaling Pathways in oA42i-Induced Neurotoxicity in Mice

PI3K/Akt1 and JAK2/STAT3 pathways play a vital role in the neuronal survival mechanisms and thus stimulation of these signaling pathways might have a beneficial role in the AD therapy. Surprisingly, we observed a marked (P < 0.05) 76 %, 74 %, and 59 % increase in the relative protein expression levels of p-Akt1/Akt1, p-JAK2/JAK2, and p-STAT3/STAT3, respectively, against oA42i-induced survival challenge in the hippocampal neurons. oA42i-provoked neurotoxicity caused about 2.4-, 2-, and 1.7-fold decrease (P < 0.05) in the relative protein expression levels of p-Akt1/Akt1, p-JAK2/JAK2, and p-STAT3/STAT3, when compared to the sham-operated group (Fig. 5a–d).

Fig. 5
figure 5

Effect of DOPET on the relative phosphorylated/total protein levels of Akt1, JAK2, and STAT3 in the hippocampi, 14 days after injecting oA42i (i.c.v.) in mice. a Representative Western blot of phosphorylated- and total-Akt1, JAK2, and STAT3 expressions. b Relative protein level of p-Akt1 (Ser473)/Akt1. c Relative protein level of p-JAK2 (Tyr1007/1008)/JAK2. d Relative protein level of p-STAT3 (Tyr705)/STAT3. Data were expressed as mean ± SD from six independent experiments with the similar results. *P < 0.05 (oA42i vs control); # P < 0.05 (DOPET plus oA42i vs oA42i alone)

Full size image

Regulatory Effect of DOPET on oA42i-Induced Hippocampal Neuronal Apoptotic Makers in Mice

With an interest to investigate the caspase-dependent mitochondria-mediated (intrinsic) apoptosis in the hippocampus, the relative mRNA expression level of Bcl-2/Bad and the mRNA expression levels of cytochrome c, caspase-9, Apaf-1, and caspase-3 were measured (Fig. 6a–c). oA42i group displayed about 1.8-fold decrease in the relative mRNA expression level of Bcl2/Bad and about 2.3-, 1.9-, 2.1-, and 2.9-fold increase in the mRNA expression levels of cytochrome c, caspase-9, Apaf-1, and caspase-3 vs sham-operated group (P < 0.05). On the other side, DOPET treatment caused about 60 % increase in the relative mRNA expression level of Bcl-2/Bad and about 48 %, 41 %, 43 %, and 54 % decrease in the mRNA expression levels of cytochrome c, caspase-9, Apaf-1, and caspase-3, respectively, vs oA42i group (P < 0.05).

Fig. 6
figure 6

Effect of DOPET on the apoptotic markers in the hippocampi, 14 days after injecting oA42i (i.c.v.) in mice. a Representative RT-PCR images of Bcl-2, Bad, cytochrome c, caspase-9, Apaf-1, and caspase-3 mRNA expressions. b Relative mRNA expression level of Bcl-2/Bad. c mRNA expression levels of cytochrome c, caspase-9, Apaf-1, and caspase-3. Data were expressed as mean ± SD from six independent experiments with the similar results. *P < 0.05 (oA42i vs control); # P < 0.05 (DOPET plus oA42i vs oA42i alone)

Full size image

Effect of DOPET in Modulating oA42i-Induced Dysregulation in SIRT1- and CREB-associated-Protein/mRNA Expression Levels in the Hippocampi of Mice

SIRT1 and CREB have a well-documented profile in the positive regulation of survival and memory functions in Alzheimer’s disease. In our study, oA42i-induced hippocampal injury was substantiated (P < 0.05) by the decrease in protein levels of SIRT1 and ADAM10 as well as the relative protein levels of p-CREB/CREB to about 1.9-, 1.8-, and 2-fold, respectively, in comparison with the sham-operated group. Parallel evaluation of DOPET treated mice showed significantly (P < 0.05) increased protein levels of SIRT1 and ADAM10 as well as the relative protein levels of p-CREB/CREB to about 67 %, 53 %, and 65 % (P < 0.05), respectively, compared to the oA42i group (Fig. 7a–d).

Fig. 7
figure 7

Effect of DOPET on the protein expressions of SIRT1, ADAM10, and CREB in the hippocampi, 14 days after injecting oA42i (i.c.v.) in mice. a Representative Western blots of SIRT1, ADAM10, p-CREB, and CREB. b Protein level of SIRT1, as % of control. c Protein level of ADAM10, as % of control. d Relative protein level of p-CREB (Ser133)/CREB. Data were expressed as mean ± SD from three independent experiments. *P < 0.05 (oA42i vs control); # P < 0.05 (DOPET plus oA42i vs oA42i alone)

Full size image

We also evaluated the mRNA expression levels of CREB-associated genes (BDNF, c-Fos, Nurr1, and Egr1) involved in the memory formation to unearth the link between oA42i-induced spatial memory deficits and the CREB-associated gene functions (Fig. 8a, b). Obviously, the mRNA expression levels of BDNF, c-Fos, Nurr1, and Egr1 were decreased in the oA42i group to about 2.2-, 1.8-, 2.0-, and 1.8-fold, respectively, when compared to the sham-operated group (P < 0.05). In contrary, we found that the mRNA expression levels of BDNF, c-Fos, Nurr1, and Egr1 were enhanced (P < 0.05) to about 79, 43, 78, and 51 %, respectively, in the DOPET treated mice against oA42i-induced neuro-cognitive dysfunction.

Fig. 8
figure 8

Effect of DOPET on the expression of CREB-associated genes (BDNF, c-Fos, Nurr1, and Egr1) in the hippocampi, 14 days after injecting oA42i (i.c.v.) in mice. a Representative RT-PCR images of BDNF, c-Fos, Nurr1, and Egr1. b mRNA expression levels of BDNF, c-Fos, Nurr1, and Egr1. Data were expressed as mean ± SD from six independent experiments with the similar results. *P < 0.05 (oA42i vs control); # P < 0.05 (DOPET plus oA42i vs oA42i alone)

Full size image

Evaluation of the Mitochondrial Protective Potential of DOPET Against oA42i-Induced Mitochondrial Aberrations by Using Transmission Electron Microscopy

Ultra-structural examination of mitochondria using transmission electron microscopy revealed a great deal of fragmentation, vacuoles, and cristae disruption in the oA42i-intoxicated mice. Besides, we observed abnormally enlarged mitochondria due to drastic mitochondrial swelling in the oA42i group. However, DOPET treatment revealed intact, healthy mitochondrial structures with clear cristae, in harmony with the ultra-architecture of the mitochondria in the hippocampal neurons of sham-operated group (Fig. 9).

Fig. 9
figure 9

Transmission electron microscopic study on the effect of DOPET (10 mg/kg/day for 14 days, p.o.) on the neuronal mitochondrial ultra-architecture, 14 days after injecting oA42i (i.c.v.) in mice

Full size image

Discussion

The consistency and reproducibility of amyloid beta-provoked neurotoxicity and cognitive dysfunction in the experimental models are still remaining a matter of debate in the avenue of Alzheimer’s disease research. The in vivo Alzheimer’s disease rodent model produced by Morimoto et al. (1998), using neurotoxic Aβ peptide (25–35) plus ibotenic acid, seems to resolve this predicament. Nevertheless, emergence of toxic ‘oligomeric Aβ hypothesis’ questions the aforesaid model, as the latter is based on the outmoded ‘amyloid cascade hypothesis’ which fails to explain the link between deposited amyloid-β (Aβ) in plaques and progressive neuro-cognitive dysfunction (Benilova et al. 2012). In this context, we have made two simple yet influential modifications in Morimoto group’s paradigm to mimic a more reliable and reproducible in vivo Alzheimer’s disease model: (i) we used soluble oligomeric Aβ (oAβ) peptides, but not fibrillar Aβ (fAβ), for two key reasons—oAβ is a better neurotoxicant than fAβ or Aβ cocktail (Benilova et al. 2012); unlike fAβ, oAβ complies with the modern oligomeric Aβ hypothesis. (ii) Aβ1–42 fragment was used in our study instead of Aβ25–35, as the former more potently disrupts the activity of hippocampal network and associated spatial memory functions (Gutiérrez-Lerma et al. 2013; Yue et al. 2014).

In our study, mice challenged with oligomeric Aβ1–42 plus ibotenic acid (oA42i) depicted severe impairment in the spatial reference and working memories as assessed using RAM. This outcome is in line with many earlier studies which employed Aβ1–42 plus ibotenic acid (Hruska and Dohanich 2007), Aβ40 plus ibotenic acid (Hu et al. 2005), or Aβ25–35 plus ibotenic acid (Feng et al. 2012) to incite cognitive dysfunction. However, DOPET treatment at a dosage of 10 mg/(kg day) for 14 days significantly improved the spatio-cognitive performances in mice. To unravel the enigma behind the cognition-enhancing effect of DOPET, we investigated the activities of various neuromolecular signals involved in the apoptotic and survival pathways inside the hippocampal neurons.

A wealth of work has shown that three MAPK signaling pathways, namely ERK, JNK, and p38 play pivotal role in the prevention/pathogenesis of AD (Kawamata and Shimohama 2011; Kim and Choi 2010; Marwarha and Ghribi 2012). Parallel evidence on the effect of ibotenic acid revealed that ERK is inhibited in the hippocampi of intoxicated rats (Yu et al. 2012). In sync with the studies on oligomeric Aβ- or ibotenic acid-induced ERK1/2 down-regulation, our study showed that ERK1/2 activity (phosphorylation) is significantly decreased in the oA42i group, whereas treatment with DOPET increased the ERK1/2 activity similar to the level of control group.

Recent studies in the hippocampal culture (He et al. 2013; Ma et al. 2009; Zhang et al. 2014) and transgenic mouse (Sclip et al. 2011) models showed that oligomeric Aβ-induced activation of p38-MAPK and JNK-MAPK impaired the neuronal and memory functions in simulation with the pathogenic mechanisms in AD. In addition, a study showed that ibotenic acid challenge significantly upregulated the phosphorylation of p38 and JNK leading to brain cell damage (Yu et al. 2012). Our study demonstrated that DOPET treatment effectively inhibited oA42i-evoked phosphorylation of p38 and JNK, thus underscoring its positive effect against detrimental neurosignaling pathways (Fig. 10). A study in a neuroblastoma cell line demonstrated that DOPET offers neuroprotection by mitigating Aβ-induced nuclear translocation of the nuclear factor-kappaB (NF-κB) subunits (St-Laurent-Thibault et al. 2011). This piece of evidence is tempting us to speculate that down-regulation of p38 and JNK by DOPET might have a predominant role in the decrease of both NF-κB activation and subsequent brain cell demise in the oA42i-intoxicated mice hippocampi. Of note, NF-κB was down-regulated by DOPET treatment in our study (data not shown).

Fig. 10
figure 10

Graphical representation of molecular signaling pathways involved in the neuronal survival-apoptotic programs. DOPET 3,4-Dihydroxyphenylethanol, oAβ soluble oligomeric amyloid β1–42 peptide, IBO ibotenic acid

Full size image

Interestingly, Watson and Fan reported that amyloid β1–42 provoked hippocampal neuronal apoptosis is mediated by the inactivation of PI3K/Akt and ERK/RSK pathways with concomitantly activated Bad-mediated mitochondrial pore opening, followed by apoptotic neuronal death (Watson and Fan 2005). Along the lines of these findings, observations of our study revealed that oA42i-induced neuro-cognitive dysfunction is a consequence of the down-regulated PI3K/Akt and ERK1/2-RSK2/CREB survival pathways with associated Bad activation- and Bcl-2 inactivation-arbitrated apoptosis (Fig. 10). Moreover, oA42i-intoxication has triggered the caspase-dependent mitochondrial (intrinsic) apoptotic pathway that involved—the release of cytochrome c from the mitochondria into the cytosol as well as the activation of Apaf-1 followed by the activation of downstream initiator and executioner caspases, caspase-9, and caspase-3, respectively—finally leading to the inevitable demise of the neurons. Corresponding to this mitochondria-mediated apoptosis, a repercussion of mitochondrial failure, examination under transmission electron microscope (TEM) revealed swollen mitochondria and loss of cristae in the oA42i mice. Nevertheless, DOPET treatment significantly antagonized the negative effects of oA42i by augmenting the mitochondrial integrity, bolstering the survival signaling, and deactivating Bad-mediated apoptosis. Notably, DOPET has a well-documented mitochondrial protection profile tested across various pathological models (Schaffer et al. 2007; Signorile et al. 2014; Zhu et al. 2010).

Strategies focused on the up-regulation of genes involved in memory functions are becoming one of the most attractive areas in AD research (España et al. 2010; Gao et al. 2010). In our study, we analyzed some of the vital survival and memory genes including SIRT1 (sirtuin1; NAD+-dependent deacetylase), CREB, and CREB-target genes (BDNF, c-Fos, Nurr1, and Egr1) that regulate cognitive functions. As expected, cognitive decline in the oA42i group mirrored the down-regulation of these genes in the hippocampal neurons of mice (España et al. 2010). An interesting report by Gao et al. (2010) disentangled the CREB-mediated beneficial role of SIRT1, a pro-survival gene in the spatial learning and memory functions.

We further assessed the contribution of ADAM10, a SIRT1-regulated metalloprotease gene in protecting against oA42i-induced neuro-behavioral impairment. In concurrence with the earlier reports of other researchers in neuroblastoma cell line, primary hippocampal cultures and transgenic mouse models of beta-amyloid-mediated toxicity, we found that ADAM10 activity is compromised in the oA42i-intoxicated group (Epis et al. 2008; Wang et al. 2012). Surprisingly, DOPET treatment efficiently alleviated the attenuation of ADAM10 expression, thereby advocating its robust neuroprotective effects. The increased expression of SIRT1 and CREB by DOPET alone or DOPET-rich extract was already reported in previous studies (Bayram et al. 2012; Signorile et al. 2014). As SIRT1 activation further elicited the α-secretase gene ADAM10, it could be inferred that DOPET might stimulate the non-amyloidogenic cleavage of amyloid-β protein precursor (AβPP) and enhance the clearance of neurotoxic Aβ peptide, so as to protect the hippocampal neurons. Many studies suggested that Aβ peptides induce tau phosporylation which in turn produces cytoskeletal disruption and neuritic dystrophy in the hippocampal neurons (Jin et al. 2011; Sáez et al. 2006). In this context, the ability of DOPET to inhibit tau aggregation (Daccache et al. 2011) is an evidence for the direct anti-tau effect of DOPET.

Conclusion

On the whole, our study outcome demonstrates the versatile potential of DOPET in the management of Alzheimer’s disease through its potential to alleviate aberrant apoptosis-survival mechanisms that underlie neurodegeneration and cognitive impairment in AD. Nonetheless, further studies on DOPET are imperative to explicate the comprehensive mechanisms involved in the anti-amyloidogenic and anti-tauopathic effects in AD. More to the point, we deem that future research on the potential of DOPET in mitigating the expression of AD-linked genes and stimulating the longevity- and cognition-enhancing genes to enhance neuroregeneration is highly inevitable to validate the therapeutic efficacy of this putative anti-AD drug candidate.