Find the papers that actually matter

The discovery of biologically active compounds from plants is an emerging challenge in natural pharmaceutical development. Here, we report the comprehensive strategy for undescribed compounds discovery targeting p53 pathway from Hypericum lancasteri against chronic myeloid leukemia (CML). The UHPLC-Q-TOF-MS analysis of H. lancasteri obtained 44 compounds information in positive and negative ion modes. Besides, three new compounds (1-3) together with two known derivatives (4 and 5) were obtained and identified by further phytochemical research. The dichloromethane extraction part (DEP) of H. lancasteri as well as compounds 1, 4, and 5 showed cytotoxic activity against CML cell line K562. For the mechanism of action, the potential affinity of DEP with p53 ubiquitinase MDM2 DEP was uncovered in surface plasmon resonance (SPR) experiment, and DEP may induce nucleus damage of K562 cells in fluorescence staining. Furthermore, DEP demonstrated therapeutic efficacy on K562 tumor-bearing mice in vivo. The mitochondria destruction ability of DEP was then indicated by transmission electron microscopy (TEM) analysis, which may further promote the p53 dependent apoptosis. Compounds 1, 4, and 5 may accumulate p53 protein through competitive binding with MDM2 as well as inhibit CML by p53 related cell apoptosis and cell cycle blocking. This study assembled multiple approach to discover bioactive compounds from H. lancasteri and shed new example on novel treatments for CML targeting p53 pathway.

In this study, twelve new triazole substituted pyrazole-pyrimidine hybrid compounds were synthesized through click reaction and evaluated for their cytotoxic activity against pancreatic, breast, and gastric cancer cell lines. Among the tested compounds, 4e demonstrated the most potent cytotoxic activity, with IC₅₀ values below 10 μM across all evaluated cancer cell lines: Breast cancer cell lines MCF-7 (5.6 ± 1.01 μM), MDA-MB-231 (8.18 ± 1.26 μM), and gastric cancer cell line HGC-27 (5.68 ± 0.45 μM). Furthermore, RT-qPCR analysis revealed that 4e significantly modulated the expression of apoptosis-related genes, notably inducing a marked downregulation of BIRC3, implicating the activation of the mitochondria-mediated intrinsic apoptotic pathway. Flow cytometry confirmed apoptosis induction. Furthermore, computational metabolomic pathway analysis indicated that 4e altered glucose metabolism, notably affecting genes and metabolites associated with glycolysis and fatty acid biosynthesis. These results highlight compound 4e as a promising anticancer candidate with dual action on apoptotic signaling and metabolic pathways. Given the potent biological activity of 4e, further optimization was pursued through a structure-based drug design strategy targeting the oncogenic regulator JAB1, 4e was designed as a scaffold for targeting JAB1. A virtual library of analogues was generated, and all derivatives were docked against the JAB1 crystal structure (PDB ID: 5JOG). Several compounds showed higher docking scores than the co-crystallized ligand (CSN5i-3), suggesting enhanced binding affinity. In parallel, binary QSAR models were developed using the MetaCore/MetaDrug platform to predict anticancer activity. Based on the combined docking and QSAR analyses, several promising analogues were identified and proposed for synthesis and subsequent biological evaluation in future studies.

Tumor growth promotion is achieved by overlapping intrinsic pathways of c-Myc and KRAS, and dual-targeting therapies emerge as an encouraging approach for drug discovery. G-quadruplexes (G4s) exist in the promoter regions of c-Myc and KRAS genes, rendering the transcriptional repression. G4 ligands are widely investigated in recent years, but dual-targeting ligands are still in their early stages. Therefore, tris-aryl imidazole analogs were designed and synthesized, with their binding properties to c-Myc and KRAS G4s confirmed firstly. HZ-1 was proven to be the most promising binder with relative selectivity to parallel G4s than non-parallel G4s. Then, the antitumor efficacy of HZ-1 was verified in human breast cancer MDA-MB-231 cells through NRF2-XCT-GPX4 pathway, resulting in the occurrence of ferroptosis, apoptosis and immunogenic cell death (ICD). Finally, HZ-1 exerted potent tumor growth inhibition in vivo in BALB/c mice, without significant adverse effects to the mice. CD8+ cytotoxic T lymphocytes and CD4+ helper T lymphocytes were promoted by HZ-1 both in spleens and tumors. To sum up, the interaction of HZ analogs with multiple G4s formulates a new concept for anticancer strategies.

The development of novel anti-cancer compounds that overcome acquired resistance to third-generation EGFR inhibitors such as osimertinib is of great significance. This study designed and synthesized eighteen methoxyquinazoline sulfonamide derivatives, and evaluated their anti-tumor activity using three EGFR triple-mutant tumor cell lines. Among them, the optimal compound 9f exhibited IC₅₀ values of 33.3-95.3 nM against Baf3-L858R/C797S/T790M, Baf3-Del19/C797S/T790M, and H1975-L858R/C797S/T790M cancer cell lines, which were consistent with the results of colony formation assays and 3D spheroid suspension culture assays. In anti-tumor experiments in mice bearing H1975-L858R/C797S/T790M tumors, compound 9f achieved a tumor growth inhibition rate of 67.3%. Mechanistic studies showed that 9f exerts excellent anti-inflammatory effects, including downregulating the expression of inflammation-related proteins iNOS, COX-2, and NF-κB (p65) in Raw264.7 cells (Western blot assay), increasing NO secretion in LPS-stimulated Raw264.7 cells (fluorescence intensity assay), and reducing IL-6 secretion in Raw264.7 and THP-1 cells (ELISA). Mechanistic studies also indicated that 9f promotes apoptosis of tumor cells and inhibits phosphorylation of the key tumor growth protein EGFR.

Laryngeal cancer is a common head-and-neck malignant tumor with geographically variable incidence. Its lack of specific early clinical symptoms often causes missed diagnosis, leading to most cases being identified at intermediate/advanced stages and thus poor treatment efficacy and prognosis. Early, accurate screening for high-risk populations is therefore critical to improving survival. This study systematically investigates the clinical application value of surface-enhanced Raman spectroscopy (SERS) combined with deep learning models for serum screening in early laryngeal cancer. It specifically examines the discriminative performance of one-dimensional convolutional neural network (1D-CNN) models, including CNN-attention with a fused attention mechanism and CNN-baseline without an attention mechanism, and compares these with traditional machine learning models.Serum samples were collected from three groups: Early laryngeal cancer, vocal fold polyps, and healthy controls. SERS was used to obtain serum molecular vibration fingerprint spectra. After spectral preprocessing, five models (CNN-attention, CNN-baseline, SVM, AdaBoost, KNN) were established to discriminate the three groups and compare performance. Leave-one-patient-out cross-validation (LOPO-CV) results indicate that the CNN-attention and CNN-baseline models significantly outperform other models in discrimination efficacy. Specifically, the CNN-attention model achieves a patient-level accuracy of 89.33%, while the CNN-baseline model reaches 88.67%. SVM, KNN, and AdaBoost had inferior performance. This study confirms that serum SERS-based 1D-CNN provides an efficient, accurate method for early laryngeal cancer screening, laying a foundation for clinical translation and offering new technical support for early screening.

Breast cancer is the leading female malignancy, and CDK4/6, key members the cyclin-dependent kinases (CDK) family, have become validated therapeutic targets. However, small-molecule tools that can track these kinases in living cells and tissues remain scarce. Therefore, guided by the co-crystal structure of the CDK4/6 inhibitor Palbociclib, we carried out two rounds of optimization to develop a mini-library of fluorescent probes (Q1-Q9) and subsequently screened them for selective labeling capability. Docking and biochemical assays confirmed that the new ligands retain the binding mode of Palbociclib. Among them, Q2 exhibited good affinity for CDK4 and CDK6 (IC50 = 151 ± 43 nM and 193 ± 70 nM, respectively) and provided exceptional, higher-resolution images of the kinases in live cells, outperforming antibody staining. The probe may serve as a cost-effective, stable and user-friendly alternative to antibody staining, and a potential visualization tool of drug preliminary screening by showing the changed fluorescent signals when incubated with active compounds. Importantly, Q2 retained bioactivity, inducing cell-cycle arrest and suppressing tumor-cell growth, mirroring the activity of the parent inhibitor Palbociclib. Thus, Q2 is a ready-to-use chemical tool for real-time visualization of CDK4/6 dynamics and holds promise for improving the diagnosis and treatment of breast cancer.

Accurate differentiation of benign and malignant thyroid lesions continues to pose a significant clinical challenge. Raman spectroscopy offers label-free molecular fingerprints of cells, yet the identification of diagnostic spectral patterns remains challenging. While artificial intelligence has been applied to analyze Raman data as one-dimensional (1D) signals, such approaches may overlook subtle nonlinear relationships across wavenumbers, particularly in cases involving spectrally similar constituents. Converting 1D spectral data into two-dimensional (2D) representations can preserve both amplitude and positional correlations, thereby uncovering latent temporal and structural features. However, such transformations risk incurring information loss, the extent of which is contingent upon the encoding strategy employed. To address this, we propose a novel multimodal deep learning framework that synergistically integrates 1D spectral and 2D spatiotemporal features, representing the first application in Raman-based thyroid cancer detection. Our model uniquely combines a Transformer to capture global dependencies in 1D spectra and a 3D-CNN to extract local spatial patterns from multiple 2D spectral transformations. These dual-modality features are adaptively fused through a multi-head cross-attention mechanism, enabling dynamic feature integration. The multimodal model ultimately achieves an accuracy of 94.7% in the identification of thyroid lesions, outperforming the unimodal Transformer and 3D-CNN models, which achieve accuracies of 91.0% and 89.4%, respectively. Notably, the multimodal model enhances interpretability by identifying contributions of key Raman peaks to the classification decision. Thus, the integration of SERS with explainable deep learning establishes a novel method for thyroid cancer diagnosis, achieving both exceptional diagnostic performance and significantly enhanced model interpretability.

Ruxolitinib (RUX), a selective JAK1/JAK2 inhibitor, is considered a therapeutic option for childhood B-cell precursor acute lymphoblastic leukemia (B-ALL) with JAK2 gain-of-function mutations. This study aimed to evaluate whether Raman spectroscopy combined with chemometric analysis can monitor the biochemical effects of RUX treatment in B-ALL cell lines. We employed single-cell confocal Raman imaging, flow cytometry, and Western blotting to assess the response of JAK2-mutated (MUTZ-5 and MHH-CALL-4) and wild-type (SEM) B-ALL cells to 10 μM RUX treatment over 48 h. Dimensionality reduction methods (PCA, t-SNE) and classification approach (o-PLS-DA) were applied to the spectral data to identify treatment-induced changes. RUX selectively reduced STAT5 phosphorylation and induced distinct Raman spectral shifts in JAK2-mutant cells, particularly in DNA- and protein-related bands. No significant changes were observed in JAK2 wild-type cells. The results demonstrate that Raman spectroscopy, when integrated with multivariate analysis, enables the non-destructive tracking of leukemia cell responses to targeted therapy and may support the development of phenotyping tools for drug monitoring in precision oncology.

Accurately identifying the mediastinal regions where metastatic lymph nodes are located is critical for the staging diagnosis of lung cancer. This identification task involves two distinct detection dimensions: mediastinal region identification and lymph node metastasis assessment. Traditional single-task image classification algorithms struggle to manage the interference between different classification dimensions within a single task. Existing multi-task learning methods struggle to balance the relationship between shared and task-specific features, and often fail to effectively fit the underlying data distributions and task characteristics during gradient adjustment. To address these challenges, we propose a Multi-Task Information Decoupling Strategy (MT-IDS). MT-IDS decomposes the main task into multiple auxiliary tasks along different feature dimensions, forming a unified multi-task system to optimize detection performance across tasks. A Dual-control Branch Routing Gate Mechanism (DBR) is employed in MT-IDS to compute the weighting of shared and task-specific features, thereby enabling more precise expert selection and feature extraction for each task. Additionally, a Dual-Dimensional Gradient Balancing Algorithm (DD-GB) is introduced in MT-IDS, whereby gradient balance is achieved through alignment of gradient directions and dynamic scaling of magnitudes, while the distribution of inter-task gradient characteristics is maintained. The significant advantages demonstrated by MT-IDS in both ablation and comparative experiments indicate its potential as an innovative solution for multi-dimensional medical image classification problems.

This study aims to develop and evaluate a fully automated deep learning-driven postprocessing pipeline for multiparametric renal MRI, enabling accurate kidney alignment, segmentation, and quantitative feature extraction within a single efficient workflow. Our method has three main stages. First, a segmentation network delineates renal structures in high-contrast images. Next, a deep learning-based pairwise image registration algorithm maps the multiparametric image series to a common target and transfers the predicted annotations between the multiparametric images. Finally, clinically relevant quantitative parameters are extracted through region-specific assessment of renal structure and function based on the aligned and segmented multiparametric data. We used five-fold cross-validation to compare the segmentation outcomes and extracted features with manual analyses in 24 patients with prostate cancer or neuroendocrine tumors and 10 healthy subjects, each undergoing repeated scans. Our automated pipeline achieved high agreement with expert kidney segmentation while delivering significant alignment improvements through registration. Volumetric analysis showed a strong correlation (r > 0.9) with manual results, and feature extraction demonstrated high intraclass correlation coefficients with minimal bias. The complete processing pipeline, encompassing coregistration, segmentation, and feature extraction, required approximately 15 s per scan from raw input to final quantitative output. The study establishes a reliable automated pipeline for renal multiparametric MRI postprocessing. The achieved accuracy and efficiency can support improved diagnosis and treatment planning for patients with kidney disease.