Stroke is one of the leading causes of disability and mortality worldwide. Research into its mechanisms and the development of therapeutic strategies heavily rely on animal models that accurately replicate the pathological features of human disease. An ideal animal model for stroke should not only reproduce the neurological deficits and pathological changes observed in clinical patients but also demonstrate good reproducibility and translational value. This review focuses on the preparation and evaluation methods of ischemic stroke animal models. Firstly, it elaborates on the selection criteria, advantages, and disadvantages of experimental animals, including rodents (rats, mice) and non-rodents (non-human primates, miniature pigs, rabbits, zebrafish). Secondly, it provides a detailed overview of the modeling principles, key procedures, and application scopes for ischemic stroke models and hemorrhagic stroke models. Furthermore, the review summarizes advances in the applications of emerging technologies—including gene editing [e.g., clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9) gene editing], multimodal imaging (e.g., two-photon microscopy, photoacoustic imaging), artificial intelligence, optogenetics, 3D bioprinting, organoid models, and multi-omics–in model optimization, precise assessment, and mechanistic investigation. Finally, based on a systematic analysis of relevant domestic and international literature from 2019 to 2024, this review discusses model selection strategies based on research objectives, a multidimensional evaluation system encompassing behavioral, imaging, and molecular pathological assessments, and envisions future directions involving technological integration to achieve model precision and individualization. This article aims to provide a comprehensive methodological reference to help researchers select appropriate animal models of stroke according to specific scientific questions.

ObjectiveThe malaria parasites remodel the host erythrocyte structure by exporting parasite proteins that interact with the membrane skeleton proteins of red blood cells (RBCs), facilitating their intracellular survival and pathogenicity. Skeleton-binding protein 1 (SBP1) is a conserved exported protein across Plasmodium species. In Plasmodium falciparum, SBP1 has been reported to interact with erythrocyte membrane skeleton proteins 4.1R and spectrin, while its contribution to erythrocyte remodeling and parasite virulence in Plasmodium berghei (Pb) remains unclear. This study aims to determine whether PbSBP1 associates with the host cytoskeletal protein 4.1R and to investigate its role in the remodeling of host RBCs and the pathogenicity of Plasmodium berghei. MethodsIn Plasmodium berghei, the relationship between PbSBP1 and the erythrocyte cytoskeletal protein 4.1R was examined using co-immunoprecipitation. A Pbsbp1 gene knockout mutant of Plasmodium berghei (Pbsbp1∆) was generated based on the principle of double crossover homologous recombination. The deformability of erythrocytes infected with Pbsbp1∆ parasites was assessed using microfluidic methods. Microchannels with an array of cylindrical pillars were used to detect modifications in infected RBC deformability. The infected RBCs were squashed between the rows and recovered between the columns and the transit velocity (μm/s) of infected RBCs travelling through the microchannel was recorded. The component of the erythrocyte membrane skeleton junctional complex, tropomodulin (TMOD), was fluorescently labeled, and the cytoskeletal network of infected erythrocytes was imaged using super-resolution stochastic optical reconstruction microscopy (STORM) to analyze ultrastructural changes in the cytoskeleton of wild-type (WT) and Pbsbp1∆-infected erythrocytes. Actin-based junctional complexes were displayed as individual clusters by the labeled TMOD in the STORM images, and the cluster densities and distances between adjacent clusters of infected RBCs were calculated. Additionally, rodent malaria models (BALB/c mice) and experimental cerebral malaria models (C57BL/6 mice) were employed to monitor the growth of Pbsbp1∆ and WT parasites during the intraerythrocytic stage and their capacity to induce cerebral malaria in mice. ResultsPbSBP1 may participate in the remodeling of infected erythrocytes through direct or indirect interaction with the erythrocyte cytoskeletal protein 4.1R. Microfluidic assays revealed that the deformability of erythrocytes infected with Pbsbp1∆ parasites was significantly enhanced compared to those infected with WT parasites. STORM imaging further demonstrated that the ultrastructure of the erythrocyte cytoskeleton in Pbsbp1∆-infected cells was altered relative to that in WT-infected erythrocytes. The distances between nearest neighbors of clusters had a tendency to increase while the cluster densities were decreased in Pbsbp1∆-infected RBCs compared to WT-infected RBCs. Subsequent phenotypic analysis indicated that the growth rate of Pbsbp1∆ parasites during the intraerythrocytic stage was significantly slower than that of WT parasites, and their ability to induce cerebral malaria in mice was also attenuated. These findings suggest that PbSBP1 is involved in the remodeling of the erythrocyte membrane skeleton, likely through its direct or indirect interaction with protein 4.1R, thereby regulating the deformability of infected erythrocytes and influencing the pathogenicity of the blood-stage parasites. ConclusionThis study establishes a role for PbSBP1 in host erythrocyte remodeling and parasite virulence, providing new research strategies for the prevention and treatment of malaria.

Objective To construct and evaluate a prognostic model for clear cell renal cell carcinoma(ccRCC)based on preoperative lipid ratios, cM0(i+)staging, and other clinical characteristics, so as to provide a precise tool for clinical prognosis assessment. Methods A retrospective analysis was conducted on the clinical data of 215 ccRCC patients treated in our hospital during May 2014 and May 2023. Lipid ratios were calculated using preoperative lipid data, and patients were divided into cM0(i+)stage or cM0 stage according to postoperative circulating tumor cells(CTCs)test results. The optimal lipid ratio and cutoff value were selected using receiver operating characteristic(ROC)curves and the X-tile method. Patients were subdivided into three groups based on the level of the low-density lipoprotein cholesterol to non-low-density lipoprotein cholesterol ratio(LNLR)and cM0(i+)staging:Group A [LNLR＞1.68 and diagnosed with cM0(i+)], Group B [LNLR＞1.68 or diagnosed with cM0(i+)], and Group C(LNLR ≤1.68 and diagnosed with cM0). Kaplan-Meier survival analysis was used to plot survival curves for patients in different groups. The log-rank test was employed to compare differences in recurrence-free survival(RFS)among the subgroups. Multivariate Cox regression analysis was conducted to identify the independent risk factors influencing RFS, and a nomogram prediction model was constructed based on these results. The predictive performance of the model was validated using ROC curves, calibration curves, and decision curves. Results ROC curves were plotted for lipid markers, and LNLR was identified as the most predictive for RFS. Its optimal cutoff point was 1. 68. Patients in Group A experienced a significantly shorter postoperative RFS. Multivariate Cox regression analysis identified preoperative LNLR, cM0(i+)staging, pathological grade, and stage as independent risk factors for RFS. A nomogram model was constructed based on these risk factors. The area under the ROC curve(AUC)for 1-, 3-, and 5-year RFS was 0.896(95% CI:0.8121-0.9627), 0.890(95%CI:0.7879-0.9641), and 0.870(95%CI:0.7697-0.9526), indicating good discriminatory ability and predictive performance. Calibration plots demonstrated good agreement between predicted and actual outcomes. Clinical decision curve analysis showed high clinical net benefit. Conclusion Preoperative LNLR level is an independent risk factor for RFS in ccRCC patients. The prognostic prediction model based on LNLR, cM0(i+)staging, patient pathological grade, and staging demonstrates good predictive performance for RFS and holds potential clinical application value.

Objective To address the need for rapid review of automated organ-at-risk (OAR) contouring for head-and-neck cancer in online adaptive radiotherapy, this study used dosimetric indices as the reference standard to evaluate the suitability of geometric similarity metrics and to develop a rapid screening model that balances the risk of missed errors with review efficiency. Methods This retrospective study included 29 patients with head-and-neck radiotherapy, yielding 243 pairs of OAR contours (auto-generated vs. manual). Geometric similarity metrics, including the Dice similarity coefficient (Dice), the 95th percentile Hausdorff distance (HD95), and the maximum Hausdorff distance (HD100), were computed between the two contour sets. Using the dose distributions from clinical treatment plans, dosimetric differences between the two contour sets were calculated for key indices, including mean dose (D_mean) and maximum dose (D_max). Correlations between geometric metrics and dosimetric differences were analyzed. Dosimetric discrepancy events were defined as Y_3Gy using a threshold of 3 Gy. Within a univariable logistic regression triage framework, each geometric metric was used as an independent variable to estimate the probability of Y_3Gy. Discriminative performance was assessed using receiver operating characteristic (ROC) curves and precision-recall (PR) curves, with the area under the curve (AUC) including ROC-AUC and PR-AUC. Practical operating thresholds were determined via threshold sweeping. Results Geometric similarity metrics showed weak-to-moderate correlations with dosimetric differences; compared with maximum dose, correlations with mean dose differences were more consistent. Among the evaluated metrics, HD95 achieved the best classification performance for Y_3Gy. Threshold sweeping suggested that an HD95 threshold in the range of 4-9 mm can balance the risk of missed discrepancies with the efficiency of automated contour quality assurance (QA). Conclusion Several geometric similarity metrics demonstrated only weak-to-moderate associations with dosimetric differences. For head-and-neck OAR contour QA in this study, HD95 provided the best discriminative performance and can support rapid triage of automated contours, with a practical operating threshold range of 4-9 mm, potentially improving the efficiency of adaptive radiotherapy workflows.

Objective To address the need for rapid review of automated organ-at-risk (OAR) contouring for head-and-neck cancer in online adaptive radiotherapy, this study used dosimetric indices as the reference standard to evaluate the suitability of geometric similarity metrics and to develop a rapid screening model that balances the risk of missed errors with review efficiency. Methods This retrospective study included 29 patients with head-and-neck radiotherapy, yielding 243 pairs of OAR contours (auto-generated vs. manual). Geometric similarity metrics, including the Dice similarity coefficient (Dice), the 95th percentile Hausdorff distance (HD95), and the maximum Hausdorff distance (HD100), were computed between the two contour sets. Using the dose distributions from clinical treatment plans, dosimetric differences between the two contour sets were calculated for key indices, including mean dose (D_mean) and maximum dose (D_max). Correlations between geometric metrics and dosimetric differences were analyzed. Dosimetric discrepancy events were defined as Y_3Gy using a threshold of 3 Gy. Within a univariable logistic regression triage framework, each geometric metric was used as an independent variable to estimate the probability of Y_3Gy. Discriminative performance was assessed using receiver operating characteristic (ROC) curves and precision-recall (PR) curves, with the area under the curve (AUC) including ROC-AUC and PR-AUC. Practical operating thresholds were determined via threshold sweeping. Results Geometric similarity metrics showed weak-to-moderate correlations with dosimetric differences; compared with maximum dose, correlations with mean dose differences were more consistent. Among the evaluated metrics, HD95 achieved the best classification performance for Y_3Gy. Threshold sweeping suggested that an HD95 threshold in the range of 4-9 mm can balance the risk of missed discrepancies with the efficiency of automated contour quality assurance (QA). Conclusion Several geometric similarity metrics demonstrated only weak-to-moderate associations with dosimetric differences. For head-and-neck OAR contour QA in this study, HD95 provided the best discriminative performance and can support rapid triage of automated contours, with a practical operating threshold range of 4-9 mm, potentially improving the efficiency of adaptive radiotherapy workflows.

Objective To address the need for rapid review of automated organ-at-risk (OAR) contouring for head-and-neck cancer in online adaptive radiotherapy, this study used dosimetric indices as the reference standard to evaluate the suitability of geometric similarity metrics and to develop a rapid screening model that balances the risk of missed errors with review efficiency. Methods This retrospective study included 29 patients with head-and-neck radiotherapy, yielding 243 pairs of OAR contours (auto-generated vs. manual). Geometric similarity metrics, including the Dice similarity coefficient (Dice), the 95th percentile Hausdorff distance (HD95), and the maximum Hausdorff distance (HD100), were computed between the two contour sets. Using the dose distributions from clinical treatment plans, dosimetric differences between the two contour sets were calculated for key indices, including mean dose (D_mean) and maximum dose (D_max). Correlations between geometric metrics and dosimetric differences were analyzed. Dosimetric discrepancy events were defined as Y_3Gy using a threshold of 3 Gy. Within a univariable logistic regression triage framework, each geometric metric was used as an independent variable to estimate the probability of Y_3Gy. Discriminative performance was assessed using receiver operating characteristic (ROC) curves and precision-recall (PR) curves, with the area under the curve (AUC) including ROC-AUC and PR-AUC. Practical operating thresholds were determined via threshold sweeping. Results Geometric similarity metrics showed weak-to-moderate correlations with dosimetric differences; compared with maximum dose, correlations with mean dose differences were more consistent. Among the evaluated metrics, HD95 achieved the best classification performance for Y_3Gy. Threshold sweeping suggested that an HD95 threshold in the range of 4-9 mm can balance the risk of missed discrepancies with the efficiency of automated contour quality assurance (QA). Conclusion Several geometric similarity metrics demonstrated only weak-to-moderate associations with dosimetric differences. For head-and-neck OAR contour QA in this study, HD95 provided the best discriminative performance and can support rapid triage of automated contours, with a practical operating threshold range of 4-9 mm, potentially improving the efficiency of adaptive radiotherapy workflows.

Background Lung tissue fibrosis caused by exposure to free silica (SiO₂) dust is the core pathological feature of silicosis. Lipocalin-2 (LCN2), as a secreted glycoprotein, plays an important role in various inflammatory diseases, but its specific role in silicosis-induced inflammatory injury remains unclear. Objective To identify differentially expressed genes (DEGs) and enriched pathways in the lung tissues of rats with silicosis, and to explore the regulatory mechanism of LCN2 in silicosis inflammation. Methods Twenty specific pathogen-free male SD rats were randomly divided into a control group and a silicosis model group (n=10 per group). The silicosis model was established via a single intratracheal instillation of silicon dioxide (SiO₂) suspension (50 mg), while the control group received an equal volume of normal saline. In vitro, NR8383 cells were divided into a control group and a SiO₂ treatment group. Hematoxylin-eosin (HE) staining and Van Gieson (VG) staining were used to evaluate alveolar structure damage, inflammatory infiltration, and collagen deposition in the lung tissues. Transcriptomic sequencing was applied to screen for DEGs and enriched signaling pathways. Western blot was used to detect the protein expression of interleukin-6 (IL-6), E-cadherin (ECA), cluster of differentiation 36 (CD36), liver X receptor (LXR), ATP-binding cassette transporter A1 (ABCA1), and LCN2 in both rat lung tissues and NR8383 cells. Immunohistochemistry was used to further assess LCN2 expression and localization. Additionally, NR8383 cells were treated with the LCN2 inhibitor ZINC00640089 and assigned to four groups: control, ZINC00640089-only, SiO₂, and SiO₂+ZINC00640089. Western blot was used to evaluate the expression of LCN2, IL-6, and interleukin-1β (IL-1β). Results HE and VG staining revealed disorganized lung tissue structure, damaged alveolar walls, nodule formation, and increased collagen deposition in the silicosis model group compared to the controls. Transcriptomic analysis identified 2723 DEGs in the lung tissues of silicosis rats (312 down-regulated and 2411 up-regulated) with LCN2 expression being significantly up-regulated. These DEGs were primarily enriched in pathways related to the innate immune response, extracellular matrix (ECM)-receptor interaction, and inflammatory response. Immunohistochemical staining confirmed elevated LCN2 levels in the lung tissues of the model rats. Western blot demonstrated that the protein levels of IL-6, CD36, and LCN2 were significantly increased (P<0.05), whereas ABCA1 and ECA were significantly decreased in the silicosis group (P<0.05). Similarly, in SiO₂-treated NR8383 cells, IL-6, CD36, and LCN2 expressions significantly increased (P<0.05), while ABCA1 and LXR significantly decreased (P<0.05). Compared with the SiO₂ group, treatment with ZINC00640089 significantly reduced the expression levels of LCN2, IL-6, and IL-1β (P<0.05). Conclusion LCN2 is significantly upregulated in both the lung tissues of silicosis rats and SiO₂-treated NR8383 cells. Inhibiting LCN2 expression effectively reduces the SiO₂-induced inflammatory response, indicating that LCN2 serves as a potential therapeutic target for silicosis.

The recombinant human thyroid-stimulating hormone (rhTSH) can be used for differentiated thyroid cancer (DTC) preparation before ¹³¹I therapy, and follow-up. Since its approval by the American Food and Drug Administration (FDA) in 1998, rhTSH has been clinically used in European countries, the United States, and other countries for nearly 30 years. In April 2024, China’s first independently developed rhTSH agent was approved. Based on clinical experience and evidence from domestic and international studies, an expert consensus on the application of rhTSH has been developed following thorough consultations among experts. The consensus offers evidence-based recommendations to guide the appropriate use of rhTSH in clinical practice.

The Golgi apparatus (GA) is a key membranous organelle in eukaryotic cells, acting as a central component of the endomembrane system. It plays an irreplaceable role in the processing, sorting, trafficking, and modification of proteins and lipids. Under normal conditions, the GA cooperates with other organelles, including the endoplasmic reticulum (ER), lysosomes, mitochondria, and others, to achieve the precise processing and targeted transport of nearly one-third of intracellular proteins, thereby ensuring normal cellular physiological functions and adaptability to environmental changes. This function relies on Golgi protein quality control (PQC) mechanisms, which recognize and handle misfolded or aberrantly modified proteins by retrograde transport to the ER, proteasomal degradation, or lysosomal clearance, thus preventing the accumulation of toxic proteins. In addition, Golgi-specific autophagy (Golgiphagy), as a selective autophagy mechanism, is also crucial for removing damaged or excess Golgi components and maintaining its structural and functional homeostasis. Under pathological conditions such as oxidative stress and infection, the Golgi apparatus suffers damage and stress, and its homeostatic regulatory network may be disrupted, leading to the accumulation of misfolded proteins, membrane disorganization, and trafficking dysfunction. When the capacity and function of the Golgi fail to meet cellular demands, cells activate a series of adaptive signaling pathways to alleviate Golgi stress and enhance Golgi function. This process reflects the dynamic regulation of Golgi capacity to meet physiological needs. To date, 7 signaling pathways related to the Golgi stress response have been identified in mammalian cells. Although these pathways have different mechanisms, they all help restore Golgi homeostasis and function and are vital for maintaining overall cellular homeostasis. It is noteworthy that the regulation of Golgi homeostasis is closely related to multiple programmed cell death pathways, including apoptosis, ferroptosis, and pyroptosis. Once Golgi function is disrupted, these signaling pathways may induce cell death, ultimately participating in the occurrence and progression of diseases. Studies have shown that Golgi homeostatic imbalance plays an important pathological role in various major diseases. For example, in Alzheimer’s disease (AD) and Parkinson’s disease (PD), Golgi fragmentation and dysfunction aggravate the abnormal processing of amyloid β-protein (Aβ) and Tau protein, promoting neuronal loss and advancing neurodegenerative processes. In cancer, Golgi homeostatic imbalance is closely associated with increased genomic instability, enhanced tumor cell proliferation, migration, invasion, and increased resistance to cell death, which are important factors in tumor initiation and progression. In infectious diseases, pathogens such as viruses and bacteria hijack the Golgi trafficking system to promote their replication while inducing host defensive cell death responses. This process is also a key mechanism in host-pathogen interactions. This review focuses on the role of the Golgi apparatus in cell death and major diseases, systematically summarizing the Golgi stress response, regulatory mechanisms, and the role of Golgi-specific autophagy in maintaining homeostasis. It emphasizes the signaling regulatory role of the Golgi apparatus in apoptosis, ferroptosis, and pyroptosis. By integrating the latest research progress, it further clarifies the pathological significance of Golgi homeostatic disruption in neurodegenerative diseases, cancer, and infectious diseases, and reveals its potential mechanisms in cellular signal regulation.

ObjectiveThis study proposes a novel single-cell protein localization method based on a class perception graph convolutional network (CP-GCN) to overcome several critical challenges in protein microscopic image analysis, including the scarcity of cell-level annotations, inadequate feature extraction, and the difficulty in achieving precise protein localization within individual cells. The methodology involves multiple innovative components designed to enhance both feature extraction and localization accuracy. MethodsFirst, a class perception module (CPM) is developed to effectively capture and distinguish semantic features across different subcellular categories, enabling more discriminative feature representation. Building upon this, the CP-GCN network is designed to explore global features of subcellular proteins in multicellular environments. This network incorporates a category feature-aware module to extract protein semantic features aligned with label dimensions and establishes a subcellular relationship mining module to model correlations between different subcellular structures. By doing so, it generates co-occurrence embedding features that encode spatial and contextual relationships among subcellular locations, thereby improving feature representation. To further refine localization, a multi-scale feature analysis approach is employed using the K-means clustering algorithm, which classifies multi-scale features within each subcellular category and generates multi-cell class activation maps (CAMs). These CAMs highlight discriminative regions associated with specific subcellular locations, facilitating more accurate protein localization. Additionally, a pseudo-label generation strategy is introduced to address the lack of annotated single-cell data. This strategy segments multicellular images into single-cell images and assigns reliable pseudo-labels based on the CAM-predicted regions, ensuring high-quality training data for single-cell analysis. Under a transfer learning framework, the model is trained to achieve precise single-cell-level protein localization, leveraging both the extracted features and pseudo-labels for robust performance. ResultsExperimental validation on multiple single-cell test datasets demonstrates that the proposed method significantly outperforms existing approaches in terms of robustness and localization accuracy. Specifically, on the Kaggle 2021 dataset, the method achieves superior mean average precision (mAP) metrics across 18 subcellular categories, highlighting its effectiveness in diverse protein localization tasks. Visualization of the generated CAM results further confirms the model’s capability to accurately localize subcellular proteins within individual cells, even in complex multicellular environments. ConclusionThe integration of the CP-GCN network with a pseudo-labeling strategy enables the proposed method to effectively capture heterogeneous cellular features in protein images and achieve precise single-cell protein localization. This advancement not only addresses key limitations in current protein image analysis but also provides a scalable and accurate solution for subcellular protein studies, with potential applications in biomedical research and diagnostic imaging. The success of this method underscores the importance of combining advanced deep learning architectures with innovative training strategies to overcome data scarcity and improve localization performance in biological image analysis. Future work could explore the extension of this framework to other types of microscopic imaging and its application in large-scale protein interaction studies.