Although mesenchymal stem cells (MSCs) have been increasingly trialed to treat a variety of diseases, the underlying mechanisms remain still elusive. In this study, human umbilical cord (UC)-derived MSCs were stimulated by hypoxia, and the membrane microvesicles (MVs) in the supernatants were collected by ultracentrifugation, observed under an electron microscope, and the origin was identified with the flow cytometric technique. The results showed that upon hypoxic stimulus, MSCs released a large quantity of MVs of *100 nm in diameter. The MVs were phenotypically similar to the parent MSCs, except that the majority of them were negative for the receptor of platelet-derived growth factor. DiI-labeling assay revealed that MSC-MVs could be internalized into human UC endothelial cells (UC-ECs) within 8 h after they were added into the culture medium. Carboxyfluorescein succinimidyl ester-labeling technique and MTT test showed that MSC-MVs promoted the proliferation of UC-ECs in a dose-dependent manner. Further, MVs could enhance in vitro capillary network formation of UC-ECs in a Matrigel matrix. In a rat hindlimb ischemia model, both MSCs and MSC-MVs were shown to improve significantly the blood flow recovery compared with the control medium (P < 0.0001), as assessed by laser Doppler imaging analysis. These data indicate that MV releasing is one of the major mechanisms underlying the effectiveness of MSC therapy by promoting angiogenesis.

Extracellular vesicles (EVs) are increasingly tested as therapeutic vehicles and biomarkers, but still EV subtypes are not fully characterised. To isolate EVs with few co-isolated entities, a combination of methods is needed. However, this is timeconsuming and requires large sample volumes, often not feasible in most clinical studies or in studies where small sample volumes are available. Therefore, we compared EVs rendered by five commonly used methods based on different principles from conditioned cell medium and 250 µl or 3 ml plasma, that is, precipitation (Exo-Quick ULTRA), membrane affinity (exoEasy Maxi Kit), size-exclusion chromatography (qEVoriginal), iodixanol gradient (OptiPrep), and phosphatidylserine affinity (MagCapture). EVs were characterised by electron microscopy, Nanoparticle Tracking Analysis, Bioanalyzer, flow cytometry, and LC-MS/MS. The different methods yielded samples of different morphology, particle size, and proteomic profile. For the conditioned medium, Izon 35 isolated the highest number of EV proteins followed by exoEasy, which also isolated fewer non-EV proteins. For the plasma samples, exoEasy isolated a high number of EV proteins and few non-EV proteins, while Izon 70 isolated the most EV proteins. We conclude that no method is perfect for all studies, rather, different methods are suited depending on sample type and interest in EV subtype, in addition to sample volume and budget.

Human maltase-glucoamylase (MGAM) hydrolyzes linear alpha-1,4-linked oligosaccharide substrates, playing a crucial role in the production of glucose in the human lumen and acting as an efficient drug target for type 2 diabetes and obesity. The amino-and carboxyl-terminal portions of MGAM (MGAM-N and MGAM-C) carry out the same catalytic reaction but have different substrate specificities. In this study, we report crystal structures of MGAM-C alone at a resolution of 3.1 Å, and in complex with its inhibitor acarbose at a resolution of 2.9 Å. Structural studies, combined with biochemical analysis, revealed that a segment of 21 amino acids in the active site of MGAM-C forms additional sugar subsites (+ 2 and + 3 subsites), accounting for the preference for longer substrates of MAGM-C compared with that of MGAM-N. Moreover, we discovered that a single mutation of Trp1251 to tyrosine in MGAM-C imparts a novel catalytic ability to digest branched alpha-1,6-linked oligosaccharides. These results provide important information for understanding the substrate specificity of alphaglucosidases during the process of terminal starch digestion, and for designing more efficient drugs to control type 2 diabetes or obesity.

Molecular acidity is one of the important physiochemical properties of a molecular system, yet its accurate calculation and prediction are still an unresolved problem in the literature. In this work, we propose to make use of the quantities from the information-theoretic (IT) approach in density functional reactivity theory and provide an accurate description of molecular acidity from a completely new perspective. To illustrate our point, five different categories of acidic series, singly and doubly substituted benzoic acids, singly substituted benzenesulfinic acids, benzeneseleninic acids, phenols, and alkyl carboxylic acids, have been thoroughly examined. We show that using IT quantities such as Shannon entropy, Fisher information, Ghosh-Berkowitz-Parr entropy, information gain, Onicescu information energy, and relative Rényi entropy, one is able to simultaneously predict experimental pKa values of these different categories of compounds. Because of the universality of the quantities employed in this work, which are all density dependent, our approach should be general and be applicable to other systems as well. © 2017 Wiley Periodicals, Inc.