Osteosarcoma (OS) is a common type of bone tumor for which there has been limited therapeutic progress over the past three decades. The prevalence of transcriptional addiction in cancer cells emphasizes the biological significance and clinical relevance of super-enhancers. In this study, we found that Max-like protein X (MLX), a member of the Myc-MLX network, is driven by super-enhancers. Upregulation of MLX predicts a poor prognosis in osteosarcoma. Knockdown of MLX impairs growth and metastasis of osteosarcoma in vivo and in vitro. Transcriptomic sequencing has revealed that MLX is involved in various metabolic pathways (e.g., lipid metabolism) and can induce metabolic reprogramming. Furthermore, knockdown of MLX results in disturbed transport and storage of ferrous iron, leading to an increase in the level of cellular ferrous iron and subsequent induction of ferroptosis. Mechanistically, MLX regulates the glutamate/cystine antiporter SLC7A11 to promote extracellular cysteine uptake required for the biosynthesis of the essential antioxidant GSH, thereby detoxifying reactive oxygen species (ROS) and maintaining the redox balance of osteosarcoma cells. Importantly, sulfasalazine, an FDA-approved anti-inflammatory drug, can inhibit SLC7A11, disrupt redox balance, and induce massive ferroptosis, leading to impaired tumor growth in vivo. Taken together, this study reveals a novel mechanism in which super-enhancer-driven MLX positively regulates SLC7A11 to meet the alleviated demand for cystine and maintain the redox balance, highlighting the feasibility and clinical promise of targeting SLC7A11 in osteosarcoma.

Denosumab (DMAB), a human monoclonal antibody against the receptor activator of the nuclear factor-kappa B ligand, is used for the treatment for unresectable giant cell tumor of bone (GCTB). However, little is known about the molecular and functional characteristics of GCTB-infiltrating lymphocytes after DMAB treatment. Here, we performed single-cell RNA sequencing and immunostaining assays to delineate the immune landscape of GCTB in the presence and absence of DMAB. We found that exhausted CD8+ T cells were preferentially enriched in DMAB-treated GCTB. A distinct M2-skewed type of tumor-associated macrophages (TAMs) comprises the majority of GCTB TAMs. We identified cytokines, including interleukin-10, and inhibitory receptors of M2 TAMs as important mediators of CD8+ T cell exhaustion. We further revealed that DMAB treatment notably increased the expression levels of periostin (POSTN) in GCTB cells. Furthermore, POSTN expression was transcriptionally regulated by c-FOS signaling and correlated with GCTB recurrence in patients after DMAB treatment. Collectively, our findings reveal that CD8+ T-cells undergo unappreciated exhaustion during DMAB therapy and that GCTB cell-derived POSTN educates TAMs and establishes a microenvironmental niche that facilitates GCTB recurrence.

No abstract

Named entity recognition aims to identify entities from unstructured text and is an important subtask for natural language processing and building knowledge graphs. Most of the existing entity recognition methods use conditional random fields as label decoders or use pointer networks for entity recognition. However, when the number of tags is large, the computational cost of method based on conditional random fields is high and the problem of nested entities cannot be solved. The pointer network uses two modules to identify the first and the last of the entities separately, and a single module can only focus on the information of the first or the last of the entities, but cannot pay attention to the global information of the entities. In addition, the neural network model has the problem of local instability. To solve mentioned problems, a named entity recognition model based on global pointer and adversarial training is proposed. To obtain global entity information, global pointer is used to decode entity information, and rotary relative position information is considered in the model designing to improve the model’s perception of position; to solve the model’s local instability problem, adversarial training is used to improve the robustness and generalization of the model. The experimental results show that the F1 score of the model are improved on several public datasets of OntoNotes5, MSRA, Resume, and Weibo compared with the existing mainstream models.

No abstract