RCMNet: A deep learning model assists CAR-T therapy for leukemia.

Acute leukemia is a type of blood cancer with a high mortality rate. Current therapeutic methods include bone marrow transplantation, supportive therapy, and chemotherapy. Although a satisfactory remission of the disease can be achieved, the risk of recurrence is still high. Therefore, novel treatments are demanding. Chimeric antigen receptor-T (CAR-T) therapy has emerged as a promising approach to treating and curing acute leukemia. To harness the therapeutic potential of CAR-T cell therapy for blood diseases, reliable cell morphological identification is crucial. Nevertheless, the identification of CAR-T cells is a big challenge posed by their phenotypic similarity with other blood cells. To address this substantial clinical challenge, herein we first construct a CAR-T dataset with 500 original microscopy images after staining. Following that, we create a novel integrated model called RCMNet (ResNet18 with Convolutional Block Attention Module and Multi-Head Self-Attention) that combines the convolutional neural network (CNN) and Transformer. The model shows 99.63% top-1 accuracy on the public dataset. Compared with previous reports, our model obtains satisfactory results for image classification. Although testing on the CAR-T cell dataset, a decent performance is observed, which is attributed to the limited size of the dataset. Transfer learning is adapted for RCMNet and a maximum of 83.36% accuracy is achieved, which is higher than that of other state-of-the-art models. This study evaluates the effectiveness of RCMNet on a big public dataset and translates it to a clinical dataset for diagnostic applications.

MERS-CoV nsp1 regulates autophagic flux via mTOR signalling and dysfunctional lysosomes.

Autophagy, a cellular surveillance mechanism, plays an important role in combating invading pathogens. However, viruses have evolved various strategies to disrupt autophagy and even hijack it for replication and release. Here, we demonstrated that Middle East respiratory syndrome coronavirus (MERS-CoV) non-structural protein 1(nsp1) induces autophagy but inhibits autophagic activity. MERS-CoV nsp1 expression increased ROS and reduced ATP levels in cells, which activated AMPK and inhibited the mTOR signalling pathway, resulting in autophagy induction. Meanwhile, as an endonuclease, MERS-CoV nsp1 downregulated the mRNA of lysosome-related genes that were enriched in nsp1-located granules, which diminished lysosomal biogenesis and acidification, and inhibited autophagic flux. Importantly, MERS-CoV nsp1-induced autophagy can lead to cell death in vitro and in vivo. These findings clarify the mechanism by which MERS-CoV nsp1-mediated autophagy regulation, providing new insights for the prevention and treatment of the coronavirus.

Targeted inhibition of STAT3 induces immunogenic cell death of hepatocellular carcinoma cells via glycolysis.

In hepatocellular carcinoma (HCC), the signal transducer and activator of transcription 3 (STAT3) is present in an overactive state that is closely related to tumour development and immune escape. STAT3 inhibition reshapes the tumour immune microenvironment, but the underlying mechanisms have not been fully clarified. We found that STAT3 inhibition could induce immunogenic cell death (ICD) of HCC cells via translocation of the "eat me" molecule calreticulin to the cell surface and a significant reduction in the expression of the "don't eat me" molecule leucocyte surface antigen CD47. STAT3 inhibition promoted dendritic cell (DC) activation and enhanced the recognition and phagocytosis of HCC cells by macrophages. Furthermore, STAT3 inhibition prevented the expression of key glycolytic enzymes, facilitating the induction of ICD in HCC. Interestingly, STAT3 directly regulated the transcription of CD47 and solute carrier family 2 member 1 (SLC2A1; also known as GLUT1). In subcutaneous and orthotopic transplantation mouse tumour models, the STAT3 inhibitor napabucasin prevented tumour growth and induced the expression of calreticulin and the protein disulfide isomerase family A member 3 (PDIA3; also known as ERp57) but suppressed that of CD47 and GLUT1. Meanwhile, the amount of tumour-infiltrated DCs and macrophages increased, along with the expression of costimulatory molecules. More CD4+ and CD8+ T cells accumulated in tumour tissues, and CD8+ T cells had lower expression of checkpoint molecules such as lymphocyte activation gene 3 protein (LAG-3) and programmed cell death protein 1 (PD-1). Significantly, the antitumour immune memory response was induced by treatment targeting STAT3. These findings provide a new mechanism for targeting STAT3-induced ICD in HCC, and confirms STAT3 as a potential target for the treatment of HCC via reshaping the tumour immune microenvironment.

Exosome-mediated remodeling of the tumor microenvironment: From local to distant intercellular communication.

Extracellular vesicles (EVs) are membrane-enveloped nanoscale particles that carry various bioactive signaling molecules secreted by cells. Their biological roles depend on the original cell type from which they are derived and their inclusions. Exosomes, a class of EVs, are released by almost all eukaryotic cell types, including tumor cells. Tumor cell-derived exosomes mediate signal transduction between tumor cells and surrounding non-tumor cells. This intercellular communication actively contributes to the remodeling of the tumor microenvironment (TME) to enable tumor growth, invasion, and metastasis. This review summarizes the latest progress in the exploration of exosome-mediated cell-cell communication implicated in TME remodeling and underlying mechanisms. We focus on the role of cell-cell interactions mediated by tumor cell-derived exosomes in promoting invasion and metastasis, and their potential as a therapeutic intervention target against distant metastasis. We also discuss the clinical translational significance of tumor-derived exosomes for early diagnosis, efficacy and progression evaluations.