TET2 regulates extranodal NK/T cell lymphoma progression through regulation of DNA methylation.

The biological role of Ten-11 translocation 2 (TET2) and the conversion of 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC) in the development of extra-nodal natural killer/T-cell lymphoma (ENKTL) remains unclear. The level of 5mC and 5hmC was detected in 112 cases of ENKTL tissue specimens by immunohistochemical (IHC) staining. Subsequently, TET2 knockdown and the overexpression cell models were constructed in ENKTL cell lines. Biochemical analyses were used to assess proliferation, apoptosis, cell cycle and monoclonal formation in cells treated or untreated with L-Ascorbic acid sodium salt (LAASS). Dot-Blots were used to detect levels of genome 5mC and 5hmC. Additionally, the ILLUMINA 850k methylation chip was used to analyze the changes of TET2 regulatory genes. RNA-Seq was used to profile differentially expressed genes regulated by TET2. The global level of 5hmC was significantly decreased, while 5mC was highly expressed in ENKTL tissue. TET2 protein expression was negatively correlated with the ratio of 5mC/5hmC (p < 0.0001). The 5mC/5hmC status were related to the site of disease, clinical stage, PINK score and Ki-67 index, as well as the 5-year OS. TET2 knockdown prolonged the DNA synthesis period, increased the cloning ability of tumor cells, increased the level of 5mC and decreased the level of 5hmC in ENKTL cells. While overexpression of TET2 presented the opposite effect. Furthermore, treatment of ENKTL cells with LAASS significantly induced ENKTL cell apoptosis. These results suggest that TET2 plays an important role in ENKTL development via regulation of 5mC and 5hmC and may serve as a novel therapeutic target for ENKTL.

Development and Validation of an Explainable Deep Learning Model to Predict In-Hospital Mortality for Patients With Acute Myocardial Infarction: Algorithm Development and Validation Study.

BACKGROUND: Acute myocardial infarction (AMI) is one of the most severe cardiovascular diseases and is associated with a high risk of in-hospital mortality. However, the current deep learning models for in-hospital mortality prediction lack interpretability. OBJECTIVE: This study aims to establish an explainable deep learning model to provide individualized in-hospital mortality prediction and risk factor assessment for patients with AMI. METHODS: In this retrospective multicenter study, we used data for consecutive patients hospitalized with AMI from the Chongqing University Central Hospital between July 2016 and December 2022 and the Electronic Intensive Care Unit Collaborative Research Database. These patients were randomly divided into training (7668/10,955, 70%) and internal test (3287/10,955, 30%) data sets. In addition, data of patients with AMI from the Medical Information Mart for Intensive Care database were used for external validation. Deep learning models were used to predict in-hospital mortality in patients with AMI, and they were compared with linear and tree-based models. The Shapley Additive Explanations method was used to explain the model with the highest area under the receiver operating characteristic curve in both the internal test and external validation data sets to quantify and visualize the features that drive predictions. RESULTS: A total of 10,955 patients with AMI who were admitted to Chongqing University Central Hospital or included in the Electronic Intensive Care Unit Collaborative Research Database were randomly divided into a training data set of 7668 (70%) patients and an internal test data set of 3287 (30%) patients. A total of 9355 patients from the Medical Information Mart for Intensive Care database were included for independent external validation. In-hospital mortality occurred in 8.74% (670/7668), 8.73% (287/3287), and 9.12% (853/9355) of the patients in the training, internal test, and external validation cohorts, respectively. The Self-Attention and Intersample Attention Transformer model performed best in both the internal test data set and the external validation data set among the 9 prediction models, with the highest area under the receiver operating characteristic curve of 0.86 (95% CI 0.84-0.88) and 0.85 (95% CI 0.84-0.87), respectively. Older age, high heart rate, and low body temperature were the 3 most important predictors of increased mortality, according to the explanations of the Self-Attention and Intersample Attention Transformer model. CONCLUSIONS: The explainable deep learning model that we developed could provide estimates of mortality and visual contribution of the features to the prediction for a patient with AMI. The explanations suggested that older age, unstable vital signs, and metabolic disorders may increase the risk of mortality in patients with AMI.

Semi-supervised model based on implicit neural representation and mutual learning (SIMN) for multi-center nasopharyngeal carcinoma segmentation on MRI.

BACKGROUND: The issue of using deep learning to obtain accurate gross tumor volume (GTV) and metastatic lymph nodes (MLN) segmentation for nasopharyngeal carcinoma (NPC) on heterogeneous magnetic resonance imaging (MRI) images with limited labeling remains unsolved. METHOD: We collected 918 patients with MRI images from three hospitals to develop and validate models and proposed a semi-supervised framework for the fine delineation of multi-center NPC boundaries by integrating uncertainty-based implicit neural representations named SIMN. The framework utilizes the deep mutual learning approach with CNN and Transformer, incorporating dynamic thresholds. Additionally, domain adaptive algorithms are employed to enhance the performance. RESULTS: SIMN predictions have a high overlap ratio with the ground truth. Under the 20 % labeled cases, for the internal test cohorts, the average DSC in GTV and MLN are 0.7981 and 0.7804, respectively; for external test cohort Wu Zhou Red Cross Hospital, the average DSC in GTV and MLN are 0.7217 and 0.7581, respectively; for external test cohorts First People Hospital of Foshan, the average DSC in GTV and MLN are 0.7004 and 0.7692, respectively. No significant differences are found in DSC, HD95, ASD, and Recall for patients with different clinical categories. Moreover, SIMN outperformed existing classical semi-supervised methods. CONCLUSIONS: SIMN showed a highly accurate GTV and MLN segmentation for NPC on multi-center MRI images under Semi-Supervised Learning (SSL), which can easily transfer to other centers without fine-tuning. It suggests that it has the potential to act as a generalized delineation solution for heterogeneous MRI images with limited labels in clinical deployment.

Physical activity trajectory during pregnancy and associations with maternal fatigue using a growth mixture modeling approach.

The purpose of this study was to investigate the associations of physical activity trajectories with maternal fatigue. Pregnant women provided objectively assessed physical activity data by Pregnancy Physical Activity Questionnaire four times. Fatigue scale-14 was used to assess fatigue during pregnancy. Growth mixture modelling characterized physical activity trajectories across pregnancy. The generalized estimating equations was used to analyze the relationship between different physical activity profiles and fatigue in pregnant women. A total of 626 pregnant women were included in analysis in a teaching hospital in Nantong city. Fatigue (total, mental and physical) was not different between two groups based on total energy expenditure of PA (constantly high vs. constantly low). The pregnant women in "constantly high household PA" group had the higher fatigue compared to "constantly low household PA" (P < 0.05) and "constantly medium household PA" (P < 0.05). The pregnant women in "constantly high sport PA" group had lower fatigue compared to "constantly low sport PA" (P < 0.05). Household PA and sport PA were still an independent influencing factor for fatigue after controlling for confounding variables. Specifically, we observed that higher household PA and lower sport PA were associated with higher fatigue during pregnancy.

Lycium ruthenicum water extract preserves retinal ganglion cells in chronic ocular hypertension mouse models.

Lycium ruthenicum Murray (LR), known as "black goji berry" or "black wolfberry", is widely utilized in chinese herbal medicine. LR fruit showed its antioxidant and/or anti-inflammation activity in treating cardiac injury, experimental colitis, nonalcoholic fatty liver disease, fatigue, and aging. Glaucoma is the leading cause of irreversible blindness. Besides elevated intraocular pressure (IOP), oxidative stress and neuroinflammation were recognized to contribute to the pathogenesis of glaucoma. This study investigated the treatment effects of LR water extract (LRE) on retinal ganglion cells (RGCs) threatened by sustained IOP elevation in a laser-induced chronic ocular hypertension (COH) mouse model and the DBA/2J mouse strain. The antioxidation and anti-inflammation effects of LRE were further tested in the H2O2-challenged immortalized microglial (IMG) cell line in vitro. LRE oral feeding (2 g/kg) preserved the function of RGCs and promoted their survival in both models mimicking glaucoma. LRE decreased 8-hydroxyguanosine (oxidative stress marker) expression in the retina. LRE reduced the number of Iba-1+ microglia in the retina of COH mice, but not in the DBA/2J mice. At the mRNA level, LRE reversed the COH induced HO-1 and SOD-2 overexpressions in the retina of COH mice. Further in vitro study demonstrated that LRE pretreatment to IMG cells could significantly reduce H2O2 induced oxidative stress through upregulation of GPX-4, Prdx-5, HO-1, and SOD-2. Our work demonstrated that daily oral intake of LRE can be used as a preventative/treatment agent to protect RGCs under high IOP stress probably through reducing oxidative stress and inhibiting microglial activation in the retina.

Identification of UNC5B as a novel aggressive biomarker for osteosarcoma based on basement membrane genes.

BACKGROUND: The prognosis of patients with metastatic osteosarcoma is poor, and the variation of basement membrane genes (BMGs) is associated with cancer metastasis. However, the role of BMGs in osteosarcoma has been poorly studied. METHODS: BMGs were collected and differentially expressed BMGs (DE-BMGs) were found through difference analysis. DE-BMGs were further screened by univariate Cox regression and Lasso regression analyses, and six key BMGs were identified and defined as basement membrane genes signatures (BMGS). Then, BMGS was used to construct the osteosarcoma BMGS risk score system, and the osteosarcoma patients were divided into high- and low-risk groups based on the median risk score. Single-sample gene set enrichment analysis (ssGSEA) and ESTIMATE scores were used to investigate the differences in immune infiltration between the two scoring groups. Additionally, we investigated whether UNC5B affects various features in tumors by bioinformatic analysis and whether UNC5B was involved in multiple biological functions of osteosarcoma cells by wound healing assay, transwell assay, and western blot. RESULTS: The osteosarcoma BMGS risk score reliably predicts the risk of metastasis, patient prognosis, and immunity. UNC5B expression was elevated in osteosarcoma, and correlated with various characteristics such as immune infiltration, prognosis, and drug sensitivity. In vitro assays showed that UNC5B knockdown reduced osteosarcoma cells' capacity for migration and invasion, and EMT process. CONCLUSION: A novel BMGS risk score system that can effectively predict the prognosis of osteosarcoma was developed and validated. The UNC5B gene in this system is one of the key aggressive biomarkers of osteosarcoma.

TIMP1 protects against blood-brain barrier disruption after subarachnoid haemorrhage by inhibiting ubiquitination of astrocytic ß1-integrin.

BACKGROUND: Astrocytes regulate blood-brain barrier (BBB) integrity, whereas subarachnoid haemorrhage (SAH) results in astrocyte dysregulation and BBB disruption. Here, we explored the involvement of tissue inhibitor of matrix metalloprotease-1 (TIMP1) in astrocyte-mediated BBB protection during SAH, along with its underlying mechanisms. METHODS: C57BL/6J mice were used to establish a model of SAH. The effects of TIMP1 on SAH outcomes were analysed by intraperitoneal injection of recombinant mouse TIMP1 protein (rm-TIMP1; 250 µg/kg). The roles of TIMP1 and its effector ß1-integrin on astrocytes were observed by in vivo transduction with astrocyte-targeted adeno-associated virus carrying TIMP1 overexpression plasmid or ß1-integrin RNAi. The molecular mechanisms underlying TIMP1 and ß1-integrin interactions were explored in primary cultured astrocytes stimulated with red blood cells (RBCs). RESULTS: TIMP1 was upregulated after SAH. Administration of rm-TIMP1 mitigated SAH-induced early brain injury (EBI) in male and female mice. TIMP1 was primarily expressed in astrocytes; its overexpression in astrocytes led to increased ß1-integrin expression in astrocytes, along with the preservation of astrocytic endfoot attachment to the endothelium and subsequent recovery of endothelial tight junctions. All of these effects were reversed by the knockdown of ß1-integrin in astrocytes. Molecular analysis showed that TIMP1 overexpression decreased the RBC-induced ubiquitination of ß1-integrin; this effect was partially achieved by inhibiting the interaction between ß1-integrin and the E3 ubiquitin ligase Trim21. CONCLUSION: TIMP1 inhibits the interaction between ß1-integrin and Trim21 in astrocytes, thereby rescuing the ubiquitination of astrocytic ß1-integrin. It subsequently restores interactions between astrocytic endfeet and the endothelium, as well as BBB integrity, eventually mitigating SAH-induced EBI.

Identifying PLAUR as a Pivotal Gene of Tumor Microenvironment and Regulating Mesenchymal Phenotype of Glioblastoma.

The mesenchymal (MES) phenotype of glioblastoma (GBM) is the most aggressive and therapy-resistant subtype of GBM. The MES phenotype transition during tumor progression results from both tumor-intrinsic genetic alterations and tumor-extrinsic microenvironmental factors. In this study, we sought to identify genes that can modulate the MES phenotype via both mechanisms. By integrating weighted gene co-expression network analysis (WGCNA) and the differential expression analysis of hypoxia-immunosuppression-related genes, we identified the plasminogen activator, urokinase receptor (PLAUR) as the hub gene. Functional enrichment analysis and GSVA analysis demonstrated that PLAUR was associated with the MES phenotype of glioma and the hypoxia-immunosuppression-related microenvironmental components. Single-cell sequencing analysis revealed that PLAUR mediated the ligand-receptor interaction between tumor-associated macrophages (TAMs) and glioma cells. Functional experiments in vitro with cell lines or primary glioma cells and xenograft models using BALB/c nude mice confirmed the role of PLAUR in promoting the MES phenotype of GBM. Our findings indicate that PLAUR regulates both glioma cells and tumor cell-extrinsic factors that favor the MES phenotype and suggest that PLAUR might be a potential target for GBM therapy.

Neutrophil-like pH-responsive pro-efferocytic nanoparticles improve neurological recovery by promoting erythrophagocytosis after intracerebral hemorrhage.

Rationale: Intracerebral hemorrhage (ICH) is a devastating cerebrovascular disease resulting from blood extravasating into the brain parenchyma. Escalation of erythrophagocytosis (a form of efferocytosis), avoiding the consequent release of the detrimental erythrocyte lysates, may be a promising target of ICH management. The ADAM17 inhibitor and liver X receptor (LXR) agonist could promote efficient efferocytosis and injury repair. Nevertheless, the poor bioavailability and restriction of the blood-brain barrier (BBB) hinder their application. Therefore, it is needed that biocompatible and smart nanoplatforms were designed and synthesized to realize effective therapy targeting erythrophagocytosis. Methods: We first assessed the synergistic effect of therapeutic GW280264X (an ADAM17 inhibitor) and desmosterol (an LXR agonist) on erythrophagocytosis in vitro. Then a pH-responsive neutrophil membrane-based nanoplatform (NPEOz) served as a carrier to accurately deliver therapeutic GW280264X and desmosterol to the damaged brain was prepared via co-extrusion. Afterwards, their pH-responsive performance was valued in vitro and targeting ability was assessed through fluorescence image in vivo. Finally, the pro-erythrophagocytic and anti-neuroinflammatory ability of the nanomedicine and related mechanisms were investigated. Results: After the synergistical effect of the above two drugs on erythrophagocytosis was confirmed, we successfully developed neutrophil-disguised pH-responsive nanoparticles to efficiently co-deliver them. The nanoparticles could responsively release therapeutic agents under acidic environments, and elicit favorable biocompatibility and ability of targeting injury sites. D&G@NPEOz nanoparticles enhanced erythrophagocytosis through inhibiting shedding of the efferocytotic receptors MERTK/AXL mediated by ADAM17 and accelerating ABCA-1/ABCG-1-mediated cholesterol efflux regulated by LXR respectively. In addition, the nano-formulation was able to modulate the inflammatory microenvironment by transforming efferocytes towards a therapeutic phenotype with reducing the release of proinflammatory cytokines while increasing the secretion of anti-inflammatory factors, and improve neurological function. Conclusions: This biomimetic nanomedicine is envisaged to offer an encouraging strategy to effectively promote hematoma and inflammation resolution, consequently alleviate ICH progression.

TRIM25 promotes glioblastoma cell growth and invasion via regulation of the PRMT1/c-MYC pathway by targeting the splicing factor NONO.

BACKGROUND: Ubiquitination plays an important role in proliferating and invasive characteristic of glioblastoma (GBM), similar to many other cancers. Tripartite motif 25 (TRIM25) is a member of the TRIM family of proteins, which are involved in tumorigenesis through substrate ubiquitination. METHODS: Difference in TRIM25 expression levels between nonneoplastic brain tissue samples and primary glioma samples was demonstrated using publicly available glioblastoma database, immunohistochemistry, and western blotting. TRIM25 knockdown GBM cell lines (LN229 and U251) and patient derived GBM stem-like cells (GSCs) GBM#021 were used to investigate the function of TRIM25 in vivo and in vitro. Co-immunoprecipitation (Co-IP) and mass spectrometry analysis were performed to identify NONO as a protein that interacts with TRIM25. The molecular mechanisms underlying the promotion of GBM development by TRIM25 through NONO were investigated by RNA-seq and validated by qRT-PCR and western blotting. RESULTS: We observed upregulation of TRIM25 in GBM, correlating with enhanced glioblastoma cell growth and invasion, both in vitro and in vivo. Subsequently, we screened a panel of proteins interacting with TRIM25; mass spectrometry and co-immunoprecipitation revealed that NONO was a potential substrate of TRIM25. TRIM25 knockdown reduced the K63-linked ubiquitination of NONO, thereby suppressing the splicing function of NONO. Dysfunctional NONO resulted in the retention of the second intron in the pre-mRNA of PRMT1, inhibiting the activation of the PRMT1/c-MYC pathway. CONCLUSIONS: Our study demonstrates that TRIM25 promotes glioblastoma cell growth and invasion by regulating the PRMT1/c-MYC pathway through mediation of the splicing factor NONO. Targeting the E3 ligase activity of TRIM25 or the complex interactions between TRIM25 and NONO may prove beneficial in the treatment of GBM.

Simultaneously Modulating HIF-1α and HIF-2α and Optimizing Macrophage Polarization through the Biomimetic Gene Vector toward the Treatment of Osteoarthritis.

In osteoarthritis (OA), articular cartilage is continuously submerged in a hypoxic environment throughout life, and hypoxia-inducible factors (HIFs) play a crucial role in OA progression. Among the various HIF phenotypes, HIF-1α positively contributes to maintaining the stability of the articular cartilage matrix. In contrast, HIF-2α has a detrimental effect, leading to chondrocyte apoptosis and exacerbating inflammation. Notably, there is currently no simultaneous regulation of HIF-1α and HIF-2α for OA treatment. Thus, the biomimetic gene vector (MENP) was developed for co-delivery of siHIF-2α and Mg2+ to the inflamed regions in OA joints, comprising an inner core consisting of siHIF-2α and Mg2+ and an outer M2 macrophage membrane. In vitro and in vivo studies demonstrate that MENP effectively targets inflamed areas, efficiently silences HIF-2α, and facilitates HIF-1α-mediated cartilage restoration through Mg2+. Furthermore, it indirectly promotes the polarization of macrophages toward an anti-inflammatory M2 phenotype through its action on inflamed synoviocytes. Overall, MENP is an efficient biomimetic vehicle for alleviating inflammation and promoting cartilage repair, representing an appealing approach for OA treatment.

Impaired microglial glycolysis promotes inflammatory responses after intracerebral haemorrhage via HK2-dependent mitochondrial dysfunction.

INTRODUCTION: Intracerebral haemorrhage (ICH) is a devastating disease that leads to severe neurological deficits. Microglia are the first line of defence in the brain and play a crucial role in neurological recovery after ICH, whose activities are primarily driven by glucose metabolism. However, little is known regarding the status of glucose metabolism in microglia and its interactions with inflammatory responses after ICH. OBJECTIVES: This study investigated microglial glycolysis and its mechanistic effects on microglial inflammation after ICH. METHODS: We explored the status of glucose metabolism in the ipsilateral region and in fluorescence-activated-cell-sorting-isolated (FACS-isolated) microglia via 2-deoxy-[18F]fluoro-D-glucose positron emission tomography (FDG-PET) analyses and gamma emission, respectively. Energy-related targeted metabolomics, along with 13C-glucose isotope tracing, was utilised to analyse glycolytic products in microglia. Mitochondrial membrane potential and mitochondrial reactive oxygen species (MitoROS) accumulation was assessed by flow cytometry. Behavioural, western blotting, gene regulation, and enzymatic activity analyses were conducted with a focus on microglia. RESULTS: Neurological dysfunction was strongly correlated with decreased FDG-PET signals in the perihaematomal region, where microglial uptake of FDG was reduced. The decreased quantity of glucose-6-phosphate (G-6-P) in microglia was attributed to the downregulation of glucose transporter 1 (GLUT1) and hexokinase 2 (HK2). Enhanced inflammatory responses were driven by HK2 suppression via decreased mitochondrial membrane potential, which could be rescued by MitoROS scavengers. HK inhibitors aggravated neurological injury by suppressing FDG uptake and enhancing microglial inflammation in ICH mice. CONCLUSION: These findings indicate an unexpected metabolic status in pro-inflammatory microglia after ICH, consisting of glycolysis impairment caused by the downregulation of GLUT1 and HK2. Additionally, HK2 suppression promotes inflammatory responses by disrupting mitochondrial function, providing insight into the mechanisms by which inflammation may be facilitated after ICH and indicating that metabolic enzymes as potential targets for ICH treatment.