No abstract

Alzheimer disease (AD) is a complex neurodegenerative disorder with no definite treatment. The expression of miR-29 family is significantly reduced in AD, suggesting a part for the family members in pathogenesis of the disease. The recent emergence of microRNA (miRNA)-based therapeutic approaches is emphasized on the efficiency of miRNA transfer to target cells. The endogenously made secretory vesicles could provide a biological vehicle for drug delivery. Characteristics such as small sizes, the ability to cross the blood-brain barrier, the specificity in binding to the right target cells, and most importantly the capacity to be engineered as drug carriers have made exosomes desirable vehicles to deliver genetic materials to the central nervous system. Here, we transfected rat bone marrow mesenchymal stem cells and HEK-293T cells (human embryonic kidney 293 cells) with recombinant expression vectors, carrying either mir-29a or mir-29b precursor sequences. A significant overexpression of miR-29 and downregulation of their targets genes, BACE1 (β-site amyloid precursor protein cleaving enzyme 1) and BIM [Bcl−2 interacting mediator of cell death (BCL2-like 11)], were confirmed in the transfected cells. Then, we confirmed the packaging of miR-29 in exosomes secreted from the transfected cells. Finally, we investigated a possible therapeutic effect of the engineered exosomes to reduce the pathological effects of amyloid-β (Aβ) peptide in a rat model of AD. Aβ-treated model rats showed some deficits in spatial learning and memory. However, in animals injected with miR-29containing exosomes at CA1 (cornu ammonis area), the aforementioned impairments were prevented. In conclusion, our findings provide a new approach for the packaging of miR-29 in exosomes and that the engineered exosomes might have a therapeutic potential in AD.

No abstract

No abstract

No abstract

No abstract

Prediction of solvation effects on the kinetics of elementary reactions occurring at metal−water interfaces is of high importance for the rational design of catalysts for the biomass and electrocatalysis communities. A lack of knowledge of the reliability of various computational solvation schemes for processes on metal surfaces is currently a limiting factor. Using a multilevel quantum mechanical/molecular mechanical (QM/MM) description of the potential energy surface, we determined characteristic time and length scales for typical free-energy perturbation (FEP) calculations of bond cleavages in ethylene glycol, a sugar surrogate molecule, over Pt(111). Our approach is based on our explicit solvation model for metal surfaces and the repetition of FEP calculations to estimate confidence intervals. Results indicate that aqueous phase effects on the free energies of elementary processes can be determined with 95% confidence intervals from limited configuration space sampling and the fixed charge approximation used in the QM/MM-FEP methodology of smaller 0.1 eV. Next, we computed the initial O−H, C−H, and C− OH bond cleavages in ethylene glycol over Pt(111) in liquid water utilizing two different metal−water interaction potentials. Our calculations predict that aqueous phase effects are small (<0.1 eV) for the C−H bond cleavage and the activation barrier of the C−OH bond cleavage. In contrast, solvation effects are large (>0.35 eV) for the O−H bond cleavage and the reaction free energy of the C−OH bond scission. While the choice of a different Pt−water force field can lead to differences in predicted solvation effects of up to 0.2 eV, the differences are usually smaller (<0.1 eV), and the trends are always the same. In contrast, implicit solvation methods appear to currently not be able to reliably describe solvation effects originating from hydrogen bonding for metal surfaces even qualitatively.

Computational catalyst discovery involves the development of microkinetic reactor models based on estimated parameters determined from density functional theory (DFT). For complex surface chemistries, the cost of calculating the adsorption energies by DFT for a large number of reaction intermediates can become prohibitive. Here, we have identified appropriate descriptors and machine learning models that can be used to predict part of these adsorption energies given data on the rest of them. Our investigations also included the case when the species data used to train the predictive model is of different size relative to the species the model tries to predict -an extrapolation in the data space which is typically difficult with regular machine learning models.We have developed a neural network based predictive model that combines an established model with the concepts of a convolutional neural network that, 1 arXiv:1910.00623v1 [physics.chem-ph] 1 Oct 2019 when extrapolating, achieves significant improvement over the previous models.