AD is the most common form of neurodegenerative dementias.¹ Amyloid-beta (Abeta) deposition in senile plaques, neurofibrillary tangle formation and neuro-inflammation are the major pathogenic mechanisms that remain to be clarified. Data from animal models and human autopsy revealed that reactive gliosis might exert a key pathogenetic role in AD.² On the basis of these considerations, it is reasonable to assume that early combination of neuroprotective and anti-inflammatory treatments aimed at restoring astrocyte functions may represent an appropriate approach to treat AD.³ PEA is a naturally occurring amide of ethanolamine and palmitic acid, abundant in the CNS, and produced by glial cells,⁴ studied for its anti-inflammatory and neuroprotective effects.^3,5-8In the present work we exploited the availability of a transgenic model of AD (3xTg-AD) harboring three mutant human genes (betaAPPSwe, PS1M146V, tauP301L) to investigate if chronic PEA treatment might modulate the onset and the progression of Abeta and tau pathologies. The 3xTg-AD mice develop amyloid plaques and neurofibrillary pathology in AD-relevant brain regions (Hip, cortex) and mimic the disease progression in humans.^9-11An integrated approach, involving histological and spectroscopic studies were used. The main experimental strategy involved a longitudinal ageing study examining metabolic alteration and relating any changes with indices of brain pathology taken from a subset of mice killed at 6 and 12 months of age. The experiment compared 6 (n=11) and 12 months of age (n=16) 3xTg-AD mice and wildtype controls (wt n=13 and n=14, for 6 and 12 months, respectively). Chronic PEA treatment were given to evaluate its impact on the onset of the neuropathology. In particular, 3-month-old mice were continuously infused (subcutaneously) up to 90 days via osmotic mini-pumps with 10mg/kg of PEA and vehicle. Moreover, in order to elucidate whether the progression of Abeta and tau pathologies are impacted by the PEA administration, another set of animals at 9 months of age were chronically treated as previously described and tested until they reach 12 months of age. At the end of treatment animals undergo MRI and MRS scanning to evaluate genotype- and treatment- induced differences in brain metabolism. MR examinations were performed on VARIAN Inova MRI/MRS system (4.7T) and a combination of volume and surface coil (RAPID Biomedical). Multislice fast spin-echo (TR/TEeff=3000/70ms, ns=2, slice thickness 1mm, matrix 128x256) sagittal images were acquired to localise the regions of interest. Single-voxel localised ¹H MR spectra (PRESS,TR/TE=4000/23ms, ns=256) were collected from: prefrontal cortex (PFC), 10.8ml and hippocampus (Hip), 9.5ml. Spectra were analysed by using LCModel program and unsuppressed water signal as reference.¹² Statistical analysis was performed by using ANOVA (2x2 genotype and treatment).

Quantitative MRS analyses and their significant differences due to genotype or treatment are summarized in Fig.1. In PFC at 6 months post-hoc analyses showed increased Ins in 3xTg-AD with respect to wt in both PEA- and vehicle-treated mice, an increase of tCr in vehicle-treated mice and of Tau in PEA-treated mice. At 12 months PEA effect was observed in all mice as an increase of tCr, and as an increase of Ins and Tau in 3xTG-AD mice only. In Hip at 6 months post-hoc analyses revealed a tCho increase, as PEA effect, in 3xTg-AD only mice. At 12 month, an increase of Glu was observed in 3xTg-AD with respect to wt mice.

Our studies show that PEA treatment increase tCr in the PFC in the 3xTg-AD mice suggesting a change in energy metabolism. The apparent paradox of Ins decrease at 12 months in 3xTg-AD with respect to 6 months, could be not directly associated with the occurrence of Abeta, but associated with a change in the balance of neuronal/glial contributions. In fact, at 6 months the higher level of Ins of 3xTg-AD with respect to WT mice could be due to an increase of microglia and/or astrocytes activity, while at 12 months PEA could help to restore the original neuronal/glial balance.¹³

Our findings are supported by parallel RT-PCR and Western blot analyses. In 6 months old mice, an increase of markers related to glia activation and neuroinflammation (e.g. S100B, GFAP, iNOS) in Hip homogenates was found. On the contrary, data from 12 months old mice revealed a reduction of some of these factors, suggesting a decrease of glia activity.

Our data indicate that PEA treatment affects brain metabolism. The current investigation provides evidence that PEA rescues altered molecular pathways that can mimic some traits of AD. Considering the safety and tolerability of PEA in humans, our findings offer new opportunity in AD treatment.