Atherosclerosis, a chronic inflammatory disease, is the major cause of life-threatening complications such as myocardial infarction and stroke. Endothelial apoptosis plays a vital role in the initiation and progression of atherosclerotic lesions. Although a subset of microRNAs (miRs) have been identified as critical regulators of atherosclerosis, studies on their participation in endothelial apoptosis in atherosclerosis have been limited. In our study, we found that miR-26a expression was substantially reduced in the aortic intima of ApoE−/− mice fed with a high-fat diet (HFD). Treatment of human aortic endothelial cells (HAECs) with oxidized low-density lipoprotein (ox-LDL) suppressed miR-26a expression. Forced expression of miR-26a inhibited endothelial apoptosis as evidenced by MTT assay and TUNEL staining results. Further analysis identified TRPC6 as a target of miR-26a, and TRPC6 overexpression abolished the anti-apoptotic effect of miR-26a. Moreover, the cytosolic calcium and the mitochondrial apoptotic pathway were found to mediate the beneficial effects of miR-26a on endothelial apoptosis. Taken together, our study reveals a novel role of miR-26a in endothelial apoptosis and indicates a therapeutic potential of miR-26a for atherosclerosis associated with apoptotic cell death.

Document-level machine translation (MT) remains challenging due to the difficulty in efficiently using document context for translation. In this paper, we propose a hierarchical model to learn the global context for documentlevel neural machine translation (NMT). This is done through a sentence encoder to capture intra-sentence dependencies and a document encoder to model document-level intersentence consistency and coherence. With this hierarchical architecture, we feedback the extracted global document context to each word in a top-down fashion to distinguish different translations of a word according to its specific surrounding context. In addition, since largescale in-domain document-level parallel corpora are usually unavailable, we use a twostep training strategy to take advantage of a large-scale corpus with out-of-domain parallel sentence pairs and a small-scale corpus with in-domain parallel document pairs to achieve the domain adaptability. Experimental results on several benchmark corpora show that our proposed model can significantly improve document-level translation performance over several strong NMT baselines.

Due to its great importance in deep natural language understanding and various down-stream applications, text-level parsing of discourse rhetorical structure (DRS) has been drawing more and more attention in recent years. However, all the previous studies on text-level discourse parsing adopt bottom-up approaches, which much limit the DRS determination on local information and fail to well benefit from global information of the overall discourse. In this paper, we justify from both computational and perceptive points-of-view that the top-down architecture is more suitable for textlevel DRS parsing. On the basis, we propose a top-down neural architecture toward text-level DRS parsing. In particular, we cast discourse parsing as a recursive split point ranking task, where a split point is classified to different levels according to its rank and the elementary discourse units (EDUs) associated with it are arranged accordingly. In this way, we can determine the complete DRS as a hierarchical tree structure via an encoder-decoder with an internal stack. Experimentation on both the English RST-DT corpus and the Chinese CDTB corpus shows the great effectiveness of our proposed top-down approach towards textlevel DRS parsing.

Background and Purpose Pressure overload‐induced cardiac interstitial fibrosis is viewed as a major cause of heart failure in patients with hypertension or aorta atherosclerosis. The purpose of this study was to investigate the effects and the underlying mechanisms of genistein, a natural phytoestrogen found in soy bean extract, on pressure overload‐induced cardiac fibrosis. Experimental Approach Genisten was administered to mice with pressure overload induced by transverse aortic constriction. Eight weeks later, its effects on cardiac dysfunction, hypertrophy and fibrosis were determined. Its effects on proliferation, collagen production and myofibroblast transformation of cardiac fibroblasts (CFs) and the signalling pathways were also assessed in vitro. Key Results Pressure overload‐induced cardiac dysfunction, hypertrophy and fibrosis were markedly attenuated by genistein. In cultured CFs, genistein inhibited TGFβ1‐induced proliferation, collagen production and myofibroblast transformation. Genistein suppressed TGFβ‐activated kinase 1 (TAK1) expression and produced anti‐fibrotic effects by blocking the TAK1/MKK4/JNK pathway. Further analysis indicated that it up‐regulated oestrogen‐dependent expression of metastasis‐associated gene 3 (MTA3), which was found to be a negative regulator of TAK1. Silencing MTA3 by siRNA, or inhibiting the activity of the MTA3‐NuRD complex with trichostatin A, abolished genistein's anti‐fibrotic effects. Conclusions and Implications Genistein improved cardiac function and inhibited cardiac fibrosis in response to pressure overload. The underlying mechanism may involve regulation of the MTA3/TAK1/MKK4/JNK signalling pathway. Genistein may have potential as a novel agent for prevention and therapy of cardiac disorders associated with fibrosis. Linked Articles This article is part of a themed section on Chinese Innovation in Cardiovascular Drug Discovery. To view the other articles in this section visit http://dx.doi.org/10.1111/bph.2015.172.issue-23

Emerging evidence has suggested the critical role of endothelial to mesenchymal transition (EndMT) in fibrotic diseases. The present study was designed to examine whether EndMT is involved in arsenic trioxide (As2O3)-induced cardiac fibrosis and to explore the underlying mechanisms. Cardiac dysfunction was observed in rats after exposure to As2O3 for 15 days using echocardiography, and the deposition of collagen was detected by Masson’s trichrome staining and electron microscope. EndMT was indicated by the loss of endothelial cell markers (VE-cadherin and CD31) and the acquisition of mesenchymal cell markers (α-SMA and FSP1) determined by RT-PCR at the mRNA level and Western blot and immunofluorescence analysis at the protein level. In the in-vitro experiments, endothelial cells acquired a spindle-shaped morphology accompanying downregulation of the endothelial cell markers and upregulation of the mesenchymal cell markers when exposed to As2O3. As2O3 activated the AKT/GSK-3β/Snail signaling pathway, and blocking this pathway with PI3K inhibitor (LY294002) abolished EndMT in As2O3-treated endothelial cells. Our results highlight that As2O3 is an EndMT-promoting factor during cardiac fibrosis, suggesting that targeting EndMT is beneficial for preventing As2O3-induced cardiac toxicity.