Background. Excessive proliferation and activation of B cells, resulting in the production of various autoantibodies, is a crucial link and significant feature of the pathogenesis of systemic lupus erythematosus (SLE), as well as the pathological basis of systemic multiorgan damage. However, whether exosomes derived from human umbilical cord mesenchymal stem cells (hucMSCs-Exo) are involved in the immune regulation of SLE has not been clarified. Objectives. Therefore, our study aimed to investigate the efficacy of hucMSCs-Exo for treating SLE. Methods. hucMSCs-Exo and peripheral blood mononuclear cells (PBMCs) of SLE patients were cocultured in vitro, and B cell apoptosis, activation, proliferation, and inflammation levels were detected by flow cytometry. Subsequently, the expression level of miR-155 in B lymphocytes of SLE patients was detected by qRT-PCR, and the target gene relationship between miR-155 and SHIP-1 was found through bioinformatics and dual luciferase activity experiments, which verified the inhibition of miR-155 in B lymphocytes of SLE patients to regulate immunity. Results. We found that hucMSCs-Exo promoted B cell apoptosis, prevented B cell overactivation, and reduced inflammation. MicroRNA-155 (miR-155) has a powerful regulatory function in B cells. It was demonstrated that hucMSCs-Exo acts synergistically with miR-155 inhibitors to target SHIP-1 to B cells more effectively than exosomes alone. Conclusion. Our results provide insight into how hucMSCs-Exo regulates autoimmunity in patients with lupus and suggest targeting miR-155 for autoimmunity while protecting immunity.

For the health monitoring of the offshore drilling platform in the salt-fog environment, three nonmetallic materials, i.e., silica, FR-4 epoxy board and sheet molding compound (SMC), with the good anti-corrosion ability were chosen as the packaging materials for the fiber Bragg grating (FBG) strain sensors. By selecting a highly sensitive structure of the fiber Bragg grating strain sensor, the performances of the sensors with three materials were investigated both numerically and experimentally. The strain sensitivities were 3.76 pm/μɛ, 3.02 pm/μɛ and 3.03 pm/μɛ, respectively. The linearity was better than 0.998. It provides useful information for developing sensors for the marine engineering.

Background : Glioma is the most common human central nervous system tumour with a high degree of malignancy. Some Rab GTPases have significant effects on glioma. Objective: This study aimed to investigate the effect of Rab3b (Rab GTPase3b) on human glioma cell proliferation and apoptosis by silencing Rab3b and to initially verify the value of Rab3b expression for the diagnosis and progression in human glioma. Methods: Rab3b was silenced by siRNA transfection. Human glioma tissues and normal brain tissues adjacent to glioma were obtained by surgery. Rab3b, P53, Caspase 7, Bax, and Bim mRNA and protein expression levels were detected by quantitative real-time polymerase chain reaction (qRT-PCR) and Western blotting. Cell proliferation was detected by the cell counting kit-8 assay, and the cell cycle and apoptosis were analysed using flow cytometry. Results: Rab3b mRNA and protein expression in human glioma U251 and U87 cells were significantly downregulated after Rab3b silencing. Rab3b silencing inhibited glioma cell proliferation by promoting cell cycle arrest and induced apoptosis by upregulating the expression of apoptosis-related proteins. Rab3b expression in human glioma (n = 33) was significantly higher than that in normal brain tissues adjacent to glioma (n = 15). In addition, Rab3b expression levels in high-grade gliomas (WHO III-IV, n = 19) were also significantly higher than those in low-grade gliomas (WHO I-II, n = 14). Conclusion: Rab3b expression levels are significantly related to the progression of gliomas. Moreover, Rab3b silencing not only significantly inhibits cell proliferation in gliomas via cell cycle arrest but also promotes cell apoptosis by upregulating the expression levels of apoptosis-related proteins; however these preliminary in vitro results warrant validation on in vivo studies.

Exosomes are membrane‐enclosed nanovesicles that shuttle active cargoes, such as circular RNAs (circRNAs) and microRNAs (miRNAs), between different cells. Human umbilical cord‐derived mesenchymal stem cells (Hu‐MSCs) can migrate to tumor sites and exert complex functions throughout tumor progression. In this study, we successfully isolated Hu‐MSCs from human umbilical cords based on their surface marker expression. Hu‐MSC‐derived exosomes significantly reduced the invasion, migration, and proliferation of cholangiocarcinoma (CCA) cells. Furthermore, circ_0037104 was downregulated in CCA and inhibited the proliferation and metastasis of CCA cells. Then, we investigated the effect of Hu‐MSC‐derived exosomal circ_0037104 on CCA. Circ_0037104 mainly regulates miR‐620 and enhances APAF1 expression, inhibiting CCA cell proliferation and metastasis. Overall, Hu‐MSC exosomal circ_0037104 contributes to the progression and stemness of CCA cells via miR‐620/APAF1. In conclusion, Hu‐MSC‐derived exosomal circ_0037104 sponges miR‐620 directly and negatively targets APAF1 to suppress CCA.

Text classification is a well-established task in NLP, but it has two major limitations. Firstly, text classification is heavily reliant on domain-specific knowledge, meaning that a classifier that is trained on a given corpus may not perform well when presented with text from another domain. Secondly, text classification models require substantial amounts of annotated data for training, and in certain domains, there may be an insufficient quantity of labeled data available. Consequently, it is essential to explore methods for efficiently utilizing text data from various domains to improve the performance of models across a range of domains. One approach for achieving this is through the use of multi-domain text classification models that leverage adversarial training to extract domain-shared features among all domains as well as the specific features of each domain. After observing the varying distinctness of domain-specific features, our paper introduces a curriculum learning approach using a ranking system based on keyword weight to enhance the effectiveness of multi-domain text classification models. The experimental data from Amazon reviews and FDU-MTL datasets show that our method significantly improves the efficacy of multi-domain text classification models adopting adversarial learning and reaching state-of-the-art outcomes on these two datasets.