Extracellular vesicles (EVs) are crucial mediators of intercellular communication and have the potential to serve as biomarkers for disease diagnosis and therapeutic monitoring. However, most EV isolation methods often require large sample volumes and specialized instruments or involve trade-offs between purity, yield, cost, and scalability. We developed MagPEG, a workflow that combines poly(ethylene glycol) (PEG)-mediated EV aggregation with magnetic beads to provide a simple, reproducible alternative to ultracentrifugation, size-exclusion chromatography, and commercial precipitation kits. Our optimization experiments clarified the PEG concentration, ionic strength, and bead surface chemistry that collectively influence EV aggregation, capture efficiency, and contaminant coprecipitation, allowing us to define conditions that improve purity while maintaining high recovery. Compared with commonly used methods, MagPEG produced EVs with comparable size distribution, EV markers, and proteomic profiles while relying only on standard laboratory supplies. A key feature of the platform is that EVs and EV-associated DNA, RNA, and proteins can be sequentially extracted from the same bead-bound material, reducing sample loss and hands-on time and enabling multiomic analysis for limited clinical or small animal samples. MagPEG is compatible with downstream applications including proteomics, bead-based assays, and miRNA quantification. When applied to human serum, the method supported high-throughput EV proteomic profiling and enabled the identification of Alzheimer's disease-associated protein signatures, illustrating its utility for biomarker discovery. Overall, our results establish MagPEG as a powerful, rapid, scalable, and high-throughput solution for translational applications in biomarker discovery.

In response to the growing global population and the unsustainable nature of current livestock agriculture, cultivated meat (CM) offers a promising alternative protein source. CM, produced through advanced tissue engineering techniques, aims to mitigate health, ethical, and environmental issues associated with conventional meat. This review evaluates the safety of CM consumption, comparing its risks and benefits to those of traditional meat. CM presents numerous advantages, including the reduction of zoonotic diseases and antibiotic resistance. However, it also introduces new challenges, such as genetic modification concerns and whether the CM can cause any kind of disease of affluence. The review highlights the necessity for rigorous safety assessments and regulatory frameworks to ensure CM's safe integration into the food supply. Additionally, CM production's lower environmental impact, such as low greenhouse gas emissions or reduced land and water use make it a viable solution to global food security and environmental sustainability.

Early-stage hepatocellular carcinoma diagnosis is hindered by the inability of conventional MRI to concurrently detect tiny tumors and delineate their microvascular networks. To address this, we engineered sequence-programmable iron oxide nanoparticles with precisely tuned magnetic properties that enable dual-phase T1-weighted imaging tailored to specific MRI sequences. Under ultra-short TR/TE angiography sequences, these nanoparticles produce strong T1-positive enhancement, allowing high-resolution visualization of tumor neovasculature down to 0.3 mm (in rabbits) at one-quarter of the standard gadolinium dose. Following selective uptake and aggregation by Kupffer cells, the same nanoparticles induce localized magnetic field inhomogeneity, generating pronounced T1-negative contrast in normal liver parenchyma on conventional sequences and thereby highlighting HCC lesions independent of OATP expression with detection sensitivity down to 0.4 mm (in mouse). This single-agent, dual-phase strategy integrates sensitive lesion detection with detailed vascular phenotyping, and is enabled by a green, scalable synthesis using an FDA-approved biocompatible polymer. By overcoming key limitations of current gadolinium-based agents, our platform provides a versatile and high-performance imaging solution for comprehensive early-stage HCC evaluation.

The three-dimensional (3D) learning model for science teaching and learning introduced in A Framework for K-12 Science Education and the Next-Generation Science Standards (NGSS) provide guidance towards including science practices in biology coursework. Science practices incorporate both content knowledge and skills and describe activities that scientists routinely engage in. The 3D-Learning Assessment Protocol (3D-LAP) provides criteria required for each science practice. We used the 3D-LAP to evaluate the extent to which three linked introductory biology laboratory courses have the potential to engage students in science practices. In the three courses, we found that 24/25 of the lab sessions had the potential to engage students in at least one science practice, with a total of 152 instances over the three courses. There was strong emphasis on four practices: Analyzing and Interpreting Data, Developing and Using Models, and Constructing Explanations and Engaging in Argumentation from Evidence. However, three practices-Planning Investigations, Communicating Information, and Defining Problems/Designing Solutions-were not identified in any of the biology laboratory sessions. In the courses we investigated, we also found that the potential for students to engage in science practices would increase by 89% if activities that met all but one criterion from the 3D-LAP ("near misses") were modified so that all criteria were met. Based on our assessment, if near misses were available in these courses, students would have the opportunity to engage in two science practices that are otherwise absent from the laboratory curriculum. We propose that instructors can align introductory-level, course-based laboratory work with the criteria in the 3D-LAP. A practice-focused curricula can help students gain technical facility and participate in science practices in ways that also support the Core Competencies described in Vision and Change.

Cervical cancer, a human papillomavirus (HPV) -driven malignancy with inherent viral antigenicity, serves as a key model in immuno-oncology. Yet, the variable efficacy of immune checkpoint inhibitors highlights a critical gap: the lack of preclinical models that accurately mirror the patient-specific tumor microenvironment (TME). Patient-derived organoids (PDOs) offer a transformative solution, preserving autologous tumor and immune components for personalized therapeutic testing. This study provides the first bibliometric analysis of this evolving field, mapping its knowledge structure from model development to immuno-oncological application. By analyzing 130 publications (1992-2025) from Web of Science, PubMed, and Scopus using VOSviewer and CiteSpace, we delineate a clear, three-stage evolution: from foundational HPV virology and culture techniques, through pivotal organoid methodology establishment (e.g., Burk RD, 2017), to the current translational phase. This progression is marked by citation bursts in TME modeling (e.g., Lohmussaar K, 2021) and the emergence of keywords like "drug screening" and "immunotherapy". Research is dominated by the U.S., China, and European nations, published within an interdisciplinary core of oncology, immunology, and virology journals. Our analysis highlights the role of organoids as a crucial bridge linking basic virology to clinical immuno-oncology. However, standardization hurdles and suboptimal establishment rates for certain subtypes impede clinical translation. To realize the full potential of PDOs, future work must prioritize standardized immune-competent co-culture systems, integrate AI for high-content data analysis, and promote equitable global collaborations. Our analysis suggests the role of organoids as a crucial bridge linking basic virology to clinical immuno-oncology, highlighting that current advancements in tumor microenvironment modeling are establishing the necessary foundation for future immunotherapeutic applications.

Existing multi-layer perceptron (MLP)-mixer architectures either rely heavily on extensive training data or employ rigid, hand-engineered mixing operations. This study introduces a Basis-Driven Learnable Operator (BDLO) for MLP-mixers, targeting a balance between high-capacity flexible learning and computationally efficient, yet constrained, handcrafted solutions.

Heart failure is a global health challenge with significant impacts on quality of life and healthcare systems. Heart transplantation remains the gold standard for select patients with end-stage heart failure, yet access in Latin America is limited by economic constraints, systemic barriers, and inequities in infrastructure. Despite contributing less than 7% of global Heart transplant procedures, LATAM has made important strides, with countries like Argentina demonstrating success in increasing donor rates through presumed consent legislation. However, disparities in donor availability, centralized heart transplant centers, and survival outcomes persist. This article reviews LATAM's historical contributions, current state, and challenges in heart transplant, identifying critical barriers such as the absence of unified transplant registries, limited access to advanced technologies like LVADs and ECMO, and a shortage of specialized training programs. Cultural attitudes and misconceptions surrounding organ donation further compound these issues. Solutions discussed include establishing a regional transplant registry to improve data transparency, leveraging regional collaboration to expand donor pools, and implementing public awareness campaigns to address cultural barriers. With concerted efforts in education, legislation, cost reduction, and technology integration, LATAM has the potential to significantly improve heart transplant access and outcomes. This transformation would save countless lives and position the region as a key contributor to global advancements in heart transplant.

Efficient coupling between guided optical fiber modes and radiated fields in the surrounding medium remains a fundamental limitation across photonics, sensing, and biophotonics. Micro-lensed fibers offer a promising solution, yet scalable fabrication with predictable geometry and deterministic optical performance has remained elusive. Here, we introduce laser-controlled wet-chemical etching (LCWCE), a single-parameter strategy that directly sculpts micro-lenses, from hyperbolic to parabolic and prolate elliptical profiles, on standard optical fibers. Local laser illumination establishes an axially confined etching-rate gradient, enabling sub-micrometer control of curvature and working distance independent of fiber type or internal structure. A physics-separated multiphysics framework combining wave optics, heat-transfer, and temperature-dependent etching kinetics captures the observed geometry evolution and validates the underlying mechanism. LCWCE enables milliwatt-scale, minimal-power fiber-based optical trapping, as well as minimally invasive in vivo dendritic detection and single-cell neural interrogation beyond 1.5 mm depth in live brains, transforming ubiquitous optical fibers into scalable, high-performance photonic probes.

The treatment of rheumatic diseases (RMD) is still challenged by long-term treatment dependence and refractory relapses. Therapeutic goals are therefore shifting from suppression of inflammation toward deep B-cell depletion and immune resetting. Conventional B-cell-depleting monoclonal antibodies have insufficient capacity for tissue B-cell depletion. Autologous chimeric antigen receptor T-cell (CAR-T) therapy has shown the potential for deep immune reconstitution, but its clinical use is constrained by individualized manufacturing, lymphodepleting conditioning, cost, and safety-management requirements. T-cell engagers (TCE) simultaneously bind cluster of differentiation (CD) 3 on T cells and target antigens on B cells or plasma cells. They redirect endogenous T cells to eliminate pathogenic target cells. TCE can offer the advantages of standardized manufacturing, no requirement for ex vivo cell preparation, and flexible adjustment of dose and treatment duration. Focusing on the latest advances in TCE therapy for rheumatic diseases, this review systematically summarizes studies targeting CD19, CD20, and B-cell maturation antigen (BCMA). Available evidence indicates that TCE can rapidly deplete B cells or plasma cells in systemic lupus erythematosus, systemic sclerosis, idiopathic inflammatory myopathies, and rheumatoid arthritis, and may reduce disease activity and promote remission. The main adverse events include cytokine release syndrome, infections, and hypogammaglobulinemia, most of which appear manageable. Accordingly, the clinical value of TCE in rheumatic diseases is becoming increasingly apparent. However, several key issues remain to be resolved, including the depth of tissue B-cell depletion, the quality of immune reconstitution, endogenous T-cell fitness, mechanisms of inadequate response, and long-term safety. TCE therapy is neither a costly duplication of existing B-cell-depleting therapies nor a definitive solution for immune resetting. Rather, it represents a more accessible and clinically manageable therapeutic approach that may offer a practical route toward immune resetting in RMD.

Objective: To establish a rapid method for 62 psychotropic drugs in plasma using high performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) . Methods: In January 2025, plasma samples were extracted by methanol∶acetonitrile (1∶1, V/V), and then gradient eluted using a C18 column with an aqueous solution containing 5 mmol/L ammonium acetate and 5 mmol/L ammonium acetate in methanol as the mobile phase. Positive and negative ion rapid switching scanning modes were used, and segmented multi-reaction monitoring mode (MRM) was used for analysis. Quantification was performed by the external standard method. Results: There were a good linear relationships for 62 psychotropic drugs in plasma with correlation coefficients ranging from 0.9976 to 0.9999. The recovery rates of spiked samples at low, medium and high concentrations ranged from 90.7% to 107.9%, with relative standard deviations for intra-batch and inter-batch precision ranging from 0.9% to 11.1%. Conclusion: This method is simple, fast, and suitable for clinical poisoning detection, and for the simultaneous determination of 62 psychotropic drugs in plasma.