br Acknowledgments This study was

Acknowledgments This study was funded by a grant provided by the Royan Institute. We thank Mr. Ehsan Janzamin for his unlimited assistance in the Flow Cytometry Laboratory, Dr. Abass Piryaei and Leila Sattarian in the Histology Laboratory, Mr. Chehrazi for statistical analysis, and Miss Saloomeh Gooyan for schematic illustration.
Introduction Myocardial infarction (MI) and related complications (e.g. THZ531 failure) are a great socio-economic burden to society and healthcare systems. Recent advances in (interventional) cardiology have resulted in timely and optimized coronary flow restoration through the culprit artery. Subsequently, more patients survive the initial infarction, but are susceptible to heart failure or other infarct-related complications (International Cardiovascular Disease Statistics, 2004). The major determinant for heart failure and other complications related to MI is infarct size (Schwartz et al., 2011). The increase in morbidity after MI triggered the search for adjunctive therapeutics to further limit excessive tissue loss and enhance cardiac performance. Pre-clinical studies show a great potential for engineered heart tissue for replacement therapy (Naito et al., 2006; Zimmermann et al., 2006) while stem cell injections may be promising in the treatment of patients with acute MI (Martin-Rendon et al., 2008). Interestingly, there is increasing evidence showing that the observed therapeutic effects are partly mediated by stem cell secretion. This so called ‘paracrine hypothesis’ has gained much attention and is supported by recent experimental data (Gnecchi et al., 2008). It has been shown that mesenchymal stem cell-conditioned medium (MSC-CM) enhances cardiomyocyte and/or progenitor survival after hypoxia-induced injury (Chimenti et al., 2010; Deuse et al., 2009; Gnecchi et al., 2006; Matsuura et al., 2009; Rogers et al., 2011). Furthermore, MSC-CM induces angiogenesis in the infarcted myocardium (Chimenti et al., 2010; Deuse et al., 2009; Li et al., 2010). We have shown in both murine and porcine models of myocardial ischemia/reperfusion (I/R) injury that MSC-CM reduces infarct size (Timmers et al., 2007). High performance liquid chromatography (HPLC) and dynamic light scatter (DLS) analyses revealed that MSCs secrete cardioprotective microparticles with a hydrodynamic radius ranging from 50 to 65ηm (Chen et al., 2011; Lai et al., 2010b)). Furthermore, the therapeutic efficacy of MSC-derived exosomes was independent of the tissue source for the MSCs. For example, exosomes from human embryonic stem cell-derived MSCs were similar to those derived from other fetal tissue sources (e.g. limb, kidney). This suggests that secretion of therapeutic exosomes may be a general property of all MSCs (Lai et al., 2010a). Exosomes are bi-lipid membrane vesicles with a diameter of 50–100ηm. They are secreted by various cell types through the fusion of multivesicular bodies with the plasma membrane. Exosomes carry a complex cargo load of proteins and RNA with the potential to effect many cellular processes and pathways. They are involved in complex cellular interactions such as immune responses, intercellular communication and antigen presentation (Thery et al., 2009). Multivesicular bodies store the exosomes within the cell and release them upon fusion with the plasma membrane. The identification of exosomes as the cardioprotective factor in MSC secretion makes it a potential therapeutic tool in myocardial infarction. In contrast to cell-based therapy, MSC-derived exosomes provide an ‘off-the-shelf’ therapeutic. Furthermore, exosomes are potentially safer as they are non-viable and will incur less manufacturing and storage costs. Although we previously described that exosomes reduce infarct size in mice (Lai et al., 2010a) the functional consequence and the molecular alterations in vivo responsible for its cardioprotective actions remain unknown. We have recently profiled the proteome of ischemic/reperfused mouse hearts (Li et al., 2012) and MSC-derived exosomes (Lai et al., 2013), to elucidate a molecular mechanism for the cardioprotective action of MSC-derived exosomes. Briefly, we found that the most significant proteomic changes in ischemic/reperfused mouse hearts were the depletion of proteins in fatty acid (FA) oxidation, glycolysis, tricarboxylic acid (TCA) cycle, redox homeostasis, gluthatione S-transferase and heat shock proteins. We also observed accumulation of proteins involved in apoptosis and the electron transport chain (ETC) (Li et al., 2012). Taken together, the observed proteomic changes predicted reduced ATP production, increased oxidative stress and apoptosis; all are key features of cell death after myocardial I/R injury (Ovize et al., 2010; Yellon and Hausenloy, 2007). Interestingly, the predicted biochemical changes as a result of the depletion or accumulation could be reversed or circumvented by the >800 proteins found in our MSC-derived exosomes (data available at www.exocarta.org; Lai et al., 2013. MSC exosomes were found to contain all five enzymes in the ATP generating stage of glycolysis namely glyceraldehyde 3-phosphate dehydrogenase (GAPDH), phosphoglycerate kinase (PGK), phosphoglucomutase (PGM), enolase (ENO) and pyruvate kinase m2 isoform(PKm2) as well as the phosphorylated PFKFB3 which upregulates phosphofructose kinase, the rate limiting glycolytic enzyme. This glycolytic potential was evident when exposure to MSC exosomes increased ATP production in oligomycin-treated H9C2 cardiomyocytes in which oxidative phosphorylation was inhibited. Furthermore, oxidative stress was reduced via peroxiredoxins and glutathione S-transferases in MSC-derived exosomes. Finally, we found that MSC-derived exosomes have enzymatically active CD73. CD73 is the major enzyme responsible for the formation of extracellular adenosine from released adenine nucleotides (Zimmermann, 2000). We have shown that exosomes activate adenosine receptors and induce adenosine-induced phosphorylation of ERK1/2 and Akt in H9C2 cardiomyocytes (Li et al., 2012).