Oleuropein Promotes Neural Plasticity and Neuroprotection via PPARα-Dependent and Independent Pathways

Oleuropein (OLE), a main constituent of olives, displays a pleiotropic beneficial dynamic in health and disease; the effects are based mainly on its antioxidant and hypolipidemic properties, and its capacity to protect the myocardium during ischemia. Furthermore, OLE activates the peroxisome proliferator-activated receptor (PPARα) in neurons and astrocytes, providing neuroprotection against noxious biological reactions that are induced following cerebral ischemia. The current study investigated the effect of OLE in the regulation of various neural plasticity indices, emphasizing the role of PPARα. For this purpose, 129/Sv wild-type (WT) and Pparα-null mice were treated with OLE for three weeks. The findings revealed that chronic treatment with OLE up-regulated the brain-derived neurotrophic factor (BDNF) and its receptor TrkB in the prefrontal cortex (PFC) of mice via activation of the ERK1/2, AKT and PKA/CREB signaling pathways. No similar effects were observed in the hippocampus. The OLE-induced effects on BDNF and TrkB appear to be mediated by PPARα, because no similar alterations were observed in the PFC of Pparα-null mice. Notably, OLE did not affect the neurotrophic factors NT3 and NT4/5 in both brain tissues. However, fenofibrate, a selective PPARα agonist, up-regulated BDNF and NT3 in the PFC of mice, whereas the drug induced NT4/5 in both brain sites tested. Interestingly, OLE provided neuroprotection in differentiated human SH-SY5Y cells against β-amyloid and H2O2 toxicity independently from PPARα activation. In conclusion, OLE and similar drugs, acting either as PPARα agonists or via PPARα independent mechanisms, could improve synaptic function/plasticity mainly in the PFC and to a lesser extent in the hippocampus, thus beneficially affecting cognitive functions.


Introduction
Over the last twenty years, intensive research indicated that the revolution in adult brain functionality largely depended on neural plasticity, a property describing the ability of the brain to adapt to various intrinsic and extrinsic stimuli by reorganizing its structure, function and connections [1][2][3].
Accumulating evidence suggests that neurotrophins (NTs) along with their cognate tyrosine kinase receptors (Trks) hold key roles in neural plasticity, thus determining the increased the synthesis of NT4/5. FEN also triggered a strong induction of BDNF, NT3 and TrkB in the PFC of mice, and increased the synthesis of NT4/5 in their hippocampus and PFC. The OLE-and FEN-induced effects on these important neural plasticity factors were PPARα-dependent, because they did not occur in the PPARα-deficient mice.

Animals
Adult male 129/Sv WT and Ppara-null mice [51,52] were used in this study. The WT and Pparα-null mice received ad libitum the standard rodent chow diet (diet 1324 TPF, Altromin Spezialfutter GmbH & Co., KG, Lage, Germany). All of the animals were housed up to five per cage under a standard 12-h light, 12-h dark cycle, and had continuous access to drinking water. The mice were monitored on a daily basis for outward signs of distress or adverse health effects. All procedures involving the animals were reviewed and approved by the ethics committee of the Medical School of the University of Ioannina. They conformed to the International European Ethical Standards (86/609-EEC) for the care and use of laboratory animals.

Drugs and Treatment
OLE (100 mg/kg) was administered daily in the food pellets for three consecutive weeks. The dosing regime of OLE was designed using findings from previous doseresponse experiments [40]. At the end of the experiment, the mice were killed by CO 2 asphyxiation, and trunk blood was collected in BD microtainer serum separator tubes (Becton, Dickinson and Company, Franklin Lakes, NJ, USA) for biochemical analyses. The hippocampus and prefrontal cortex were dissected from the brain for total RNA and total cellular protein extraction. All of the brain tissues and serum samples were kept at −80 • C until assayed.

In Vivo Experiments
Adult male (129/Sv) WT and Ppara-null mice were randomly assigned into groups of 8-10 mice. OLE (100 mg/kg) was administered daily in the food pellets for three consecutive weeks in both WT and Ppara-deficient mice. The controls received the normal rodent diet.
The dose of OLE was based on our previous findings [40,53] and on the literature [54]. OLE was administered in the food pellets, because it has been shown that even under normal iso-osmotic luminal conditions, OLE is poorly absorbed. Its absorption can be significantly improved by solvent flux through paracellular junctions, made possible by hypotonic conditions in the intestinal lumen [55]. The presence of glucose or amino acids in the intestinal lumen that follows a meal stimulates water flux via the opening of paracellular junctions. It is possible that this mechanism has a similar effect on OLE absorption as a hypotonic solution [56]. Although the pharmacokinetic profile of OLE has not been determined in mice, Boccio and colleagues indicated that a single oral dose of OLE (100 mg/kg) is absorbed in rats, reaching 200 ng/mL in tmax of 2 h [57]. The experiment was terminated when the mice of all groups were killed with CO 2 asphyxiation.

Quantitative Real-Time PCR
The total RNA from the hippocampus and PFC was isolated using Trizol reagent (Invitrogen, Carlsbad, CA, USA), following the manufacturer's protocol. The concentration of total RNA in each sample was determined spectrophotometrically. Quantitative real-time PCR (qPCR) was performed with cDNA, which was generated from 1 µg of total RNA using a SuperScript II reverse transcriptase kit (Invitrogen). The sequences of the forward and reverse gene-specific primers that were used in this study are shown in Table 1. The SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA) was used for the real-time reactions, which were performed employing a C1000 Touch thermal cycler with a real-time detection system (Bio-Rad Laboratories, Hercules, CA, USA). The relative mRNA expression was normalized to β-actin levels (QuantiTect primer assay; QIAGEN, Valencia, CA, USA), and the values were quantified using the comparative threshold cycle method. Table 1. Oligonucleotide sequences for quantitation of gene mRNA concentration using quantitative PCR assays.

Gene
Sequences of Primers

Western Blot Analysis
Immunoblot analysis of BDNF, TrkB, phospho-ERK, phospho-CREB and phospho-AKT protein levels was performed using total cellular extracts from the hippocampus, PFC and differentiated into neurons of human SH-SY5Y neuroblastoma cells. The total cellular proteins were extracted using the RIPA buffer supplemented with protease inhibitors, phenylmethylsulfonyl fluoride (1 mM), β-glycerophosphate (5 mM), NaF (5 mM), Na 2 MoO 4 (2 mM) and NaVO 3 (1 mM). The BCA protein assay kit (Pierce, IL, USA) was used for the determination of the protein concentration in the samples. The proteins were subjected to SDS-polyacrylamide gel electrophoresis and immunoblotting using the following antibodies: rabbit polyclonal BDNF-specific IgG (Santa Cruz Biotechnology, Dallas, TX, USA), rabbit polyclonal phosphorylated (Ser133) CREB-1-specific IgG (cell signaling, Danvers, MA, USA), rabbit monoclonal TrkB (cell signaling), rabbit monoclonal phospho-ERK (cell signaling) and rabbit monoclonal phospho-AKT (cell signaling). Secondary antibodies conjugated with horseradish peroxidase (Santa Cruz Biotechnology) were used, and the proteins were detected using an enhanced chemiluminescence detection kit (GE Healthcare, Chalfont St. Giles, Buckinghamshire, UK). Immunoblotting with either α-tubulinor β-actin-specific antibodies (Santa Cruz Biotechnology) and anti-goat IgG horseradish peroxidase-conjugated secondary antibody was used as a loading control.

Cell Differentiation and Viability
Following 24 h of incubation, the human neuroblastoma SH-SY5Y cells were differentiated into cholinergic neurons using retinoic acid (RA, 5 µM), which was added to culture medium that did not contain FBS. The cell incubation lasted for 5 days, and the culture medium supplemented with fresh RA was changed every three days.
Following 5 days of incubation, differentiated human SH-SY5Y cells were treated for 4 h with the culture medium without FBS, containing only either OLE (10 µM) or fenofi-brate (FEN, Sigma-Aldrich, Burghausen, Germany, 10 µM), or the highly selective PPARα agonist, 4-Chloro-6-(2,3-xylidino)-2-pyrimidinylthioacetic acid (Wy-14643, Sigma-Aldrich, Germany, 10 µM). Subsequently, either H 2 O 2 (final concentration 750 µM) or 7PA2-CHO supernatant (final concentration 75% v/v) from a CHO (Chinese hamster ovary) cell line stably expressing a mutant form of the human amyloid precursor protein (APP) [58] was added in the cell culture, in order to induce β-amyloid toxicity. CHO-K1 supernatant at a final concentration of 75% v/v was used in the control cell cultures. Following incubation of the differentiated cells for an additional 48 h, MTS reagent [20 µL of MTS (1.90 mg/mL)] was added, and the plates were incubated at 37 • C and 5% CO 2 for 120 min. The absorbance was measured using a spectrophotometer at 492 nm, and the results were expressed as % cell viability compared to DMSO-or medium-treated cells, which represented 100% of cell viability.

Statistical Analysis
The present data are presented as the mean ± SE. They were analyzed using the oneway analysis of variance (ANOVA) program, followed by multiple comparisons employing the Bonferonni's and Tuckey's list honest significant difference methods. The significance level for all of the analyses was set at probability of less than 0.05.

Assessment of Chronic OLE Treatment in Neural Plasticity Indices in PFC
The treatment of WT mice with OLE, a PPARα agonist [39], for 21 days increased BDNF mRNA and protein expression in their PFCs. This effect was Pparα-dependent, because OLE did not increase BDNF in Ppara-null mice ( Figure 1A). Similarly, chronic treatment of WT mice with OLE increased TrkB mRNA and protein expression in their PFCs, an effect apparently involving PPARα, because no change in TrkB expression was observed in this brain region of Pparα-null mice ( Figure 1B). Notably, OLE repressed BDNF mRNA and protein expression in the PFC of Pparα-deficient mice ( Figure 1A), an effect that underscores the distinct role of PPARα in the effects of OLE on BDNF in this brain area. Chronic treatment of WT mice with OLE had no effect on NT3 and NT4/5 mRNA expression in their PFCs ( Figures 1C and 1D, respectively). Interestingly, baseline NT3 mRNA levels were higher in Pparα-null than in WT mice, and OLE repressed them ( Figure 1C). Constitutive NT4/5 mRNA expression raged at lower levels in the PFC of Pparα-null mice than in WT mice, and OLE further repressed it ( Figure 1D).  (BDNF) mRNA levels were analyzed in 129/Sv wild-type (WT) and Pparα-null mice by qPCR and protein levels using Western blot. (B) TrkB receptor mRNA and protein levels were also analyzed using qPCR and Western blot analysis, respectively. (C) Neurotrophin NT3 and (D) NT4/5 mRNA levels were also analyzed with qPCR. Values were normalized to β-actin, and are expressed as mean ± SE (n = 8-10). Comparisons were between controls C and OLE-treated mice. Treatment group differences were calculated using one-way ANOVA, followed by Bonferroni's test. * p < 0.05, ** p < 0.01, *** p < 0.001.

Assessment of Chronic OLE Treatment on Neural Plasticity Indices in the Hippocampus
Chronic treatment of WT mice with OLE did not affect BDNF and TrkB mRNA and protein expression in their hippocampuses (Figures 2A and 2B, respectively). Importantly, however, OLE significantly repressed BDNF mRNA and protein expression in the hippocampus of Pparα-deficient mice (Figure 2A), an effect that indicates the preventive role of PPARα in the OLE-mediated down-regulation of BDNF in the hippocampus. In WT mice following chronic OLE treatment, no alteration was observed in NT3 mRNA expression in their hippocampuses compared to controls ( Figure 2C). Notably, constitutive NT3 mRNA expression ranged at higher levels in the hippocampus of Pparα-null mice than in WT mice ( Figure 2C). OLE did not affect NT4/5 mRNA expression in the hippocampus of WT and Pparα-null mice ( Figure 2D). Following treatment with OLE, brain-derived neurotrophic factor (BDNF) mRNA levels were analyzed in 129/Sv wild-type and Pparα-null mice using qPCR, and protein levels using Western blot. (B) TrkB receptor mRNA and protein levels were also analyzed using qPCR and Western blot analysis, respectively. (C) Neurotrophin NT3 and (D) NT4/5 mRNA levels were also analyzed with qPCR. Values were normalized to β-actin, and are expressed as mean ± SE (n = 8-10). Comparisons were between controls C and OLE-treated mice. Treatment group differences were calculated using one-way ANOVA, followed by Bonferroni's test. ** p < 0.01, *** p < 0.001.

OLE-Induced ERK, AKT and PKA/CREB Activation
Chronic treatment of WT mice with OLE increased the phosphorylation of ERK1/2, CREB and AKT in their PFCs compared to controls ( Figure 3A). The role of PPARα in ERK1/2, CREB and AKT activation by OLE appears to be determinant, because the drug did not affect the activation of these signaling pathways in Pparα-null mice ( Figure 3A). Interestingly, pCREB protein levels were markedly lower in the PFCs of Pparα-deficient mice, and OLE did not affect them ( Figure 3A). OLE had no similar activating effects on ERK1/2-, AKT-and PKA/CREB-linked signaling pathways in the hippocampus of WT mice ( Figure 3B). Following treatment with OLE, brain-derived neurotrophic factor (BDNF) mRNA levels were analyzed in 129/Sv wild-type and Pparα-null mice using qPCR, and protein levels using Western blot. (B) TrkB receptor mRNA and protein levels were also analyzed using qPCR and Western blot analysis, respectively. (C) Neurotrophin NT3 and (D) NT4/5 mRNA levels were also analyzed with qPCR. Values were normalized to β-actin, and are expressed as mean ± SE (n = 8-10). Comparisons were between controls C and OLE-treated mice. Treatment group differences were calculated using one-way ANOVA, followed by Bonferroni's test. ** p < 0.01, *** p < 0.001.

OLE-Induced ERK, AKT and PKA/CREB Activation
Chronic treatment of WT mice with OLE increased the phosphorylation of ERK1/2, CREB and AKT in their PFCs compared to controls ( Figure 3A). The role of PPARα in ERK1/2, CREB and AKT activation by OLE appears to be determinant, because the drug did not affect the activation of these signaling pathways in Pparα-null mice ( Figure 3A). Interestingly, pCREB protein levels were markedly lower in the PFCs of Pparα-deficient mice, and OLE did not affect them ( Figure 3A). OLE had no similar activating effects on ERK1/2-, AKT-and PKA/CREB-linked signaling pathways in the hippocampus of WT mice ( Figure 3B).

Assessment of Subacute OLE Treatment in Neural Plasticity Indices in PFC
The treatment of WT mice with OLE for 4 days had no effect on BDNF mRNA and protein expression in their PFC compared to controls ( Figure 4A); however, FEN, a more potent PPARα agonist, up-regulated BDNF in this brain area ( Figure 4A). The inducing effect of FEN on BDNF expression in the PFC was PPARα-dependent, because the drug did not affect BDNF expression in the PFC of Pparα-null mice ( Figure 4A).

Assessment of Subacute OLE Treatment in Neural Plasticity Indices in PFC
The treatment of WT mice with OLE for 4 days had no effect on BDNF mRNA and protein expression in their PFC compared to controls ( Figure 4A); however, FEN, a more potent PPARα agonist, up-regulated BDNF in this brain area ( Figure 4A). The inducing effect of FEN on BDNF expression in the PFC was PPARα-dependent, because the drug did not affect BDNF expression in the PFC of Pparα-null mice ( Figure 4A).
As in the case of BDNF, subacute treatment of WT mice with OLE had no effect on TrkB mRNA expression in their PFCs when compared to controls ( Figure 4B). Interestingly though, OLE markedly repressed NT3 mRNA expression in the PFC of WT mice potentially via PPRAα activation, because no similar effect was observed in Pparα-deficient mice ( Figure 4C). FEN up-regulated both NT3 and NT4/5 in the PFC of WT mice, and this effect appears to be PPARα-dependent, because it was not observed in Pparα-null mice ( Figure 4C,D). and Pparα-null mice using qPCR, and protein levels using Western blot. (B) TrkB receptor mRNA and protein levels were also analyzed using qPCR and Western blot analysis, respectively. (C) Neurotrophin NT3 and (D) NT4/5 mRNA levels were also analyzed with qPCR. Values were normalized to β-actin and are expressed as mean ± SE (n = 8-10). Comparisons were between controls C and OLE-treated mice. Treatment group differences were calculated using one-way ANOVA, followed by Bonferroni's test. * p < 0.05, ** p < 0.01, *** p < 0.001.
As in the case of BDNF, subacute treatment of WT mice with OLE had no effect on TrkB mRNA expression in their PFCs when compared to controls ( Figure 4B). Interestingly though, OLE markedly repressed NT3 mRNA expression in the PFC of WT mice potentially via PPRAα activation, because no similar effect was observed in Pparα-deficient mice ( Figure 4C). FEN up-regulated both NT3 and NT4/5 in the PFC of WT mice, and this effect appears to be PPARα-dependent, because it was not observed in Pparα-null mice ( Figure 4C,D).

Assessment of Subacute OLE Treatment in Neural Plasticity Indices in Hippocampus
Treatment of mice with either OLE or FEN for 4 days had no effect on BDNF, TrkB and NT3 expression in their hippocampus compared to controls ( Figure 5A-C, respectively). Nonetheless, both drugs up-regulated NT4/5 in the hippocampus, and this up-regulation appears to be PPARα-mediated, because no NT4/5 up-regulation was induced by FEN and OLE in Pparα-null mice ( Figure 5D). Following treatment with either OLE or fenofibrate (FEN), brainderived neurotrophic factor (BDNF) mRNA levels were analyzed in 129/Sv wild-type (WT) and Pparαnull mice using qPCR, and protein levels using Western blot. (B) TrkB receptor mRNA and protein levels were also analyzed using qPCR and Western blot analysis, respectively. (C) Neurotrophin NT3 and (D) NT4/5 mRNA levels were also analyzed with qPCR. Values were normalized to β-actin and are expressed as mean ± SE (n = 8-10). Comparisons were between controls C and OLE-treated mice. Treatment group differences were calculated using one-way ANOVA, followed by Bonferroni's test. * p < 0.05, ** p < 0.01, *** p < 0.001.

Assessment of Subacute OLE Treatment in Neural Plasticity Indices in Hippocampus
Treatment of mice with either OLE or FEN for 4 days had no effect on BDNF, TrkB and NT3 expression in their hippocampus compared to controls ( Figure 5A-C, respectively). Nonetheless, both drugs up-regulated NT4/5 in the hippocampus, and this up-regulation appears to be PPARα-mediated, because no NT4/5 up-regulation was induced by FEN and OLE in Pparα-null mice ( Figure 5D). Values were normalized to β-actin, and are expressed as mean ± SE (n = 8-10). Comparisons were between controls C and OLE-treated mice. Treatment group differences were calculated using one-way ANOVA, followed by Bonferroni's test. * p < 0.05, ** p < 0.01.

Effect of OLE on Differentiated Human SH-SY5Y Neuroblastoma Cells
An in vitro investigation using human SH-SY5Y neuroblastoma cells differentiated into cholinergic neurons indicated that OLE and FEN activated both the ERK1/2 and PKA/CREB signaling pathways ( Figure 6A), whereas Wy-14643 did not affect these signaling pathways ( Figure 6A). Moreover, all three substances induced PPARα expression in these cells. Interestingly, OLE increased the phosphorylation of GSK3β, whereas FEN and Wy-14643 had a weaker effect on it ( Figure 6A). Following treatment with either OLE or fenofibrate (FEN), brain-derived neurotrophic factor (BDNF) mRNA levels were analyzed in 129/Sv wild-type (WT) and Pparα-null mice via qPCR, and protein levels using Western blot. (B) TrkB receptor mRNA and protein levels were also analyzed using qPCR and Western blot analysis, respectively. (C) Neurotrophin NT3 and (D) NT4/5 mRNA levels were also analyzed with qPCR. Values were normalized to β-actin, and are expressed as mean ± SE (n = 8-10). Comparisons were between controls C and OLE-treated mice. Treatment group differences were calculated using one-way ANOVA, followed by Bonferroni's test. * p < 0.05, ** p < 0.01.

Effect of OLE on Differentiated Human SH-SY5Y Neuroblastoma Cells
An in vitro investigation using human SH-SY5Y neuroblastoma cells differentiated into cholinergic neurons indicated that OLE and FEN activated both the ERK1/2 and PKA/CREB signaling pathways ( Figure 6A), whereas Wy-14643 did not affect these signaling pathways ( Figure 6A). Moreover, all three substances induced PPARα expression in these cells. Interestingly, OLE increased the phosphorylation of GSK3β, whereas FEN and Wy-14643 had a weaker effect on it ( Figure 6A).
It is noteworthy that preincubation of the differentiated SH-SY5Y cells with OLE (but not with FEN or Wy-14643) protected them most prominently from natural amyloid β (Aβ) peptides ( Figure 6B), and to some extent from the H 2 O 2 -induced cell toxicity ( Figure 6C). In particular, OLE at a concentration of 5-10 µM provided 50% neuroprotection against Aβ-induced toxicity. These OLE-induced neuroprotective effects appear to be PPARαindependent, because two other more selective PPARα agonists, Wy-14643 and FEN, either did not protect or even dose-dependently exaggerated the toxic effects of Aβ amyloid peptides, respectively ( Figure 6B). , ERK1/2 and CREB expression levels were analyzed in proteins extracted from differentiated human SH-SY5Y cells using Western blot analysis. The numbers underneath the lanes represent the relative protein expressions that are defined as the ratio between the drug-treated and control expression, which is set at 1. Beta-amyloid-and H2O2-induced neurotoxicity was assessed using a spectrophotometric analysis of the samples at 492 nm to determine the percentage of cell viability. C: control (DMSO-treated cells). * p < 0.05, ** p < 0.01, *** p < 0.001.
It is noteworthy that preincubation of the differentiated SH-SY5Y cells with OLE (but not with FEN or Wy-14643) protected them most prominently from natural amyloid β (Aβ) peptides ( Figure 6B), and to some extent from the H2O2-induced cell toxicity ( Figure  6C). In particular, OLE at a concentration of 5-10 µM provided 50% neuroprotection against Aβ-induced toxicity. These OLE-induced neuroprotective effects appear to be PPARα-independent, because two other more selective PPARα agonists, Wy-14643 and FEN, either did not protect or even dose-dependently exaggerated the toxic effects of Αβ amyloid peptides, respectively ( Figure 6B).

Discussion
The current study investigated the impact of OLE on neural plasticity in the hippocampus and PFC of mice, emphasizing the role of the nuclear receptor and transcription factor, PPARα. The findings indicated that chronic treatment of WT mice with OLE increased the synthesis of BDNF, as previously reported [59], and its receptor, TrkB, in their PFC as compared to controls; however, the drug had no effect on them in the PFC of Pparα-null mice. This finding underscores the crucial role of PPARα in the OLEinduced up-regulation of these important indices of neural plasticity in the PFC, which is potentially triggered by activation of the ERK1/2, AKT and PKA/CREB signaling pathways that possess crucial roles in the regulation of neurotrophins [4], neural plasticity [60,61] and survival [61][62][63][64]. The present findings are in line with those of a previous study reporting that plant compounds, including resveratrol and OLE among others, improve synaptic plasticity by activating neuronal signaling pathways, which control the memory and long-term potentiation (LTP) of synapses. The OLE-induced LTP in the hippocampus

Discussion
The current study investigated the impact of OLE on neural plasticity in the hippocampus and PFC of mice, emphasizing the role of the nuclear receptor and transcription factor, PPARα. The findings indicated that chronic treatment of WT mice with OLE increased the synthesis of BDNF, as previously reported [59], and its receptor, TrkB, in their PFC as compared to controls; however, the drug had no effect on them in the PFC of Pparα-null mice. This finding underscores the crucial role of PPARα in the OLE-induced up-regulation of these important indices of neural plasticity in the PFC, which is potentially triggered by activation of the ERK1/2, AKT and PKA/CREB signaling pathways that possess crucial roles in the regulation of neurotrophins [4], neural plasticity [60,61] and survival [61][62][63][64]. The present findings are in line with those of a previous study reporting that plant compounds, including resveratrol and OLE among others, improve synaptic plasticity by activating neuronal signaling pathways, which control the memory and long-term potentiation (LTP) of synapses. The OLE-induced LTP in the hippocampus indicates increased synaptic activity, which is usually followed by a long-lasting increase in signal transmission among neurons, and is triggered by activation of signaling pathways including the PKA/CREB [29,65,66].
Unlike long-term OLE treatment, the subacute administration of WT mice with the drug at the given dose did not manage to increase BDNF and TrkB synthesis in their PFCs. Nonetheless, subacute treatment of WT mice with FEN, a selective PPARα agonist [67], up-regulated BDNF in their PFCs, but it did not affect TrkB expression in this brain tissue. The FEN-induced BDNF up-regulation in the PFC is PPARα-dependent, because the drug did not increase this neurotrophic factor in the PFC of Pparα-null mice. Interestingly, FEN also increased the synthesis of NT3 in the PFC of WT mice, and NT4/5 in both the PFC and hippocampus via PPARα activation, as no similar up-regulating effects were detected in Pparα-deficient mice. Both the long-term and subacute treatments with OLE had no up-regulating effect on NT3 expression in the PFC and hippocampus of WT mice. Although subacute OLE administration had a down-regulating effect on NT3 in the PFC of WT mice, the drug up-regulated NT4/5 in their hippocampus via a mechanism potentially involving PPARα activation, because it did not affect NT4/5 expression in Pparα-null mice. Unlike previous studies indicating that OLE can improve synaptic plasticity in the dentate gyrus of the rat hippocampus, thus attenuating Alzheimer's disease-like pathology [68], the present study indicated that either chronic or subacute treatment of WT mice with OLE at the given dose did not up-regulate the effect on BDNF and TrkB in their hippocampus; however, subacute OLE increased the synthesis of NT4/5 in this brain region. Apparently, the effect of OLE on neuronal plasticity indices is species-, dose-and time-dependent.
It is also noteworthy that OLE protected differentiated human SH-SY5Y neuroblastoma cells from β-amyloidand H 2 O 2 -induced toxicity. This neuroprotective effect of OLE against β-amyloid-induced toxicity was unrelated to PPARα activation, because both selective PPARα agonists, FEN and Wy-14643, did not prevent neurotoxicity in this in vitro neuronal model. It appears that OLE may exploit both PPARα-dependent and independent pathways to promote neural plasticity and protect against oxidative stress and β-amyloid neurotoxicity. This hypothesis is supported by the findings of a previous study reporting that the neuroprotective effect of PPARα agonists do not necessarily directly depend on PPARα-regulated pathways [69]. The present findings, along with those from previous studies reporting that OLE prevents the aggregation of β-amyloids, tau, amylin, α-synuclein and ubiquitin proteins in the brain, reducing neuronal apoptosis and activating several antioxidant pathways [70], indicate that OLE and similar drugs such as hydroxytyrosol provide neuroprotection and could be used to prevent or delay the onset of neurodegenerative disorders, a subject that should be thoroughly investigated in the framework of clinical studies.

Conclusions
The present findings indicate that neuroprotection against oxidative stress and βamyloid toxicity, as well as the induction of neural plasticity in several brain sites, belongs to the broad spectrum of the beneficial effects of OLE, the main constituent of olive products, a basic constituent of the Mediterranean diet. In this concept, OLE and similar drugs acting predominantly as PPARα agonists could modulate a diverse repertoire of functions in the central and peripheral nervous systems, as well as in non-neuronal tissues. Therefore, it is of particular interest to further investigate the potential beneficial effects of PPARα agonists on synaptic plasticity/function and dendritic outgrowth, which are critical parameters, among others, in the regulation of cognitive functions.