Exosomes are nanovesicles secreted by cells and contain various molecules including protein, lipid, and DNA/RNA. They are crucial mediators of the intercellular communication and serve as promising vehicles for drug delivery and gene therapy. Recently, accumulating evidence suggests that microRNAs (miRNAs) may serve as new and potentially powerful targets for therapeutic interventions against various human diseases. However, steadily and effectively delivering miRNA mimics or inhibitors to target cells remains a major obstacle. To enhance the efficacy of exosome-mediated delivery of miRNA molecules, it is crucial to develop a convenient and efficient method to enrich specific miRNAs or antisense oligos in isolated exosomes. Here we report a novel method to prepare specific miRNA molecule-loaded exosomes. Using a modified calcium chloride-mediated transfection method, we successfully enhanced the designated miRNA mimics or inhibitors in isolated exosomes directly, instead of transfecting their mother cells. We also compared this method with direct transfection of exosomes using electroporation. Both methods confirmed that exosomes can serve as cargos to deliver a robustly increased amount of selected miRNA mimic(s) or inhibitor(s) to the recipient cells. Delivery of these miRNA molecule enriched-exosomes subsequently results in highly efficient overexpression or deletion of the designated miRNAs in the recipient cells both in vivo and in vitro. Additionally, we confirmed that exosome-delivered miRNA mimics or inhibitors are functional in the recipient cells. Collectively, we developed a novel protocol to conveniently manipulate exosomal miRNAs with high efficiency and successfully deliver the exosomal miRNA molecules to recipient cells.

In this paper, we formally define the problem of topic modeling with network structure (TMN). We propose a novel solution to this problem, which regularizes a statistical topic model with a harmonic regularizer based on a graph structure in the data. The proposed method combines topic modeling and social network analysis, and leverages the power of both statistical topic models and discrete regularization. The output of this model can summarize well topics in text, map a topic onto the network, and discover topical communities. With appropriate instantiations of the topic model and the graph-based regularizer, our model can be applied to a wide range of text mining problems such as authortopic analysis, community discovery, and spatial text mining. Empirical experiments on two data sets with different genres show that our approach is effective and outperforms both text-oriented methods and network-oriented methods alone. The proposed model is general; it can be applied to any text collections with a mixture of topics and an associated network structure.

Intercellular communications between lung epithelial cells and alveolar macrophages play an essential role in host defense against acute lung injury. Hyperoxia-induced oxidative stress is an established model to mimic human lung injury. We show that after hyperoxia-associated oxidative stress, a large amount of extracellular vesicles (EVs) are detectable in bronchoalveolar lavage fluid (BALF) and culture medium of lung epithelial cells. Microvesicles (MVs), but not exosomes (Exos) or apoptotic bodies (Abs), are the main type of EVs found in the early stages after hyperoxia. Among all the MV compositions, small RNAs are altered the most significantly after hyperoxia-associated oxidative stress. We further confirmed that hyperoxia up-regulates the levels of certain specific miRNAs in the epithelial cell-derived MVs, such as the miR-320a and miR-221. Functionally, the hyperoxia-induced epithelial MVs promote macrophage activation in vitro and facilitate the recruitment of immunomodulatory cells in vivo detected in BALF. Using MV as a cargo, delivery of the specific miRNA-enriched epithelial MVs (miR-221 and/or miR-320a) also triggers macrophage-mediated pro-inflammatory effects. Collectively, epithelial cell-derived MVs promote macrophage-regulated lung inflammatory responses via MV-shuttling miRNAs.

Exosomes (EXOs) are a type of extracellular nanovesicles released from living cells. Accumulating evidence suggests that EXOs are involved in the pathogenesis of human diseases, including lung conditions. In recent years, the potential of EXO-mediated drug delivery has gained increasing interest. In this report, we investigated whether inhaled EXOs serve as an efficient and practical delivery vehicle to activate or inhibit alveolar macrophages (AMs), subsequently modulating pulmonary immune responses. We first identified the recipient cells of the inhaled EXOs, which were labeled with PKH26. We found that only lung macrophages efficiently take up intratracheally instilled EXOs in vivo. Using modified calcium chloride-mediated transformation, we manipulated small RNA molecules in serum-derived EXOs, including siRNAs, microRNA (miRNA) mimics, and miRNA inhibitors. Via intratracheal instillation, we successfully delivered siRNA and miRNA mimics or inhibitors into lung macrophages using the serum-derived EXOs as vehicles. Furthermore, EXO siRNA or miRNA molecules are functional in modulating LPS-induced lung inflammation in vivo. Beneficially, serum-derived EXOs themselves do not trigger lung immune responses, adding more favorable features to serve as drug delivery agents. Collectively, we developed a novel protocol using serum-derived EXOs to deliver designated small RNA molecules into lung macrophages in vivo, potentially shedding light on future gene therapy of human lung diseases.