Search

Microwave ablation is a first-line treatment of small hepatocellular carcinoma (HCC), while incomplete ablation induces recurrence and metastasis. However, its underlying mechanism remains largely unexplored. Here we reported that sublethal heat treatment (46 °C) strongly promoted migration and EMT transition in HCC cells. Mechanistic investigation revealed that compared with 37 °C, HCC cells treated with 46 °C expressed higher level of CD47. Knockdown of CD47 significantly attenuated sublethal heat treatment stimulated migration and EMT transition. In addition, METTL3 which is the key enzyme of m6A modification was also induced by 46 °C treatment and triggered CD47 expression in HCC cells. Moreover, CD47 mRNA degradation was further proved to be stabled in the IGF2BP1-dependent manner. Importantly, sublethal heat treatment stimulated CD47 expression and EMT transition were also confirmed in patient-derived organoid. Taken together, our study suggests that METTL3/IGF2BP1/CD47 mediated EMT transition contributes to the incomplete ablation induced metastasis in HCC cells. Moreover, these findings identify the METTL3/IGF2BP1/CD47 axis as a potential therapeutic target for the microwave ablation and shed new lights on the crosstalk between incomplete heat ablation and RNA methylation. • Sublethal heat treatment stimulates CD47 and METTL3 expression in HCC. • METTL3/IGF2BP1 positively regulates CD47 expression in an m6A-dependent manner. • CD47 mediated EMT contributes to the microwave ablation induced metastasis in HCC. [ABSTRACT FROM AUTHOR]

• The structure of proposed network is designed via a view of iterative reconstruction. • Better performance in artifacts removal and detail preservation. • More robustness to the samplings than the state-of-the-art methods. Magnetic resonance imaging has been widely applied in clinical diagnosis. However, it is limited by its long data acquisition time. Although the imaging can be accelerated by sparse sampling and parallel imaging, achieving promising reconstructed images with a fast computation speed remains a challenge. Recently, deep learning methods have attracted a lot of attention for encouraging reconstruction results, but they are lack of proper interpretability for neural networks. In this work, in order to enable high-quality image reconstruction for the parallel magnetic resonance imaging, we design the network structure from the perspective of sparse iterative reconstruction and enhance it with the residual structure. Experimental results on a public knee dataset indicate that, as compared with the state-of-the-art deep learning-based and optimization-based methods, the proposed network achieves lower error in reconstruction and is more robust under different samplings. [ABSTRACT FROM AUTHOR]