Browse Articles

Programmable immunoprobiotics orchestrate antitumor immune response with Pin1 inhibition for pancreatic cancer treatment

Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive cancer with limited treatment options due to its desmoplastic and immunosuppressive tumor microenvironment (TME), which impedes drug delivery and limits T cell infiltration. Immune checkpoint blockade (ICB) has shown poor efficacy in PDAC, partly due to the desmoplastic stroma and low immunogenicity. Peptidyl-prolyl cis/trans isomerase NIMA-interacting 1 (Pin1) promotes both fibrosis and immune evasion, making it a compelling target for TME remodeling. Here, we develop a dual-action, programmable immunoprobiotic delivery system (EcN@Nbs-NP@API-1) that combines Pin1 inhibition with PD-L1 blockade to enhance immunotherapy. This system uses Escherichia coli Nissle 1917 (EcN) to selectively deliver nanoparticles encapsulating the Pin1 inhibitor API-1 to PDAC, enabling sustained release to degrade the fibrotic stroma and upregulate PD-L1 on tumor cells, promoting immune infiltration. Engineered EcN also produces anti-PD-L1 nanobodies in situ, synergizing with API-1 to boost CD8 + T cell–mediated immunity. In orthotopic PDAC mouse models, this strategy remodels the TME, enhances immune cell infiltration, and improves antitumor response while minimizing systemic toxicity. Moreover, it shows efficacy in other ECM-rich tumors, such as triple-negative breast cancer, highlighting its broader potential. This work presents a promising platform to overcome immunotherapy resistance in solid tumors.

Polyelectrolyte-coated nanoporous carbon nanoparticles as pH-sensitive nanocontainers for controlled release of corrosion inhibitors

Strokeformer: A novel deep learning paradigm training transformer-based architecture for stroke prognosis prediction

Stroke, a common neurological disorder, is considered one of the leading causes of death and disability worldwide. Stroke prognosis issues involve using clinical characteristics collected from patients presented in tabular form to determine whether they are suitable for thrombolytic therapy. Transformer-based deep learning methods have achieved state-of-the-art performance in various classification tasks, but flaws still exist in dealing with tabular data. These models and algorithms largely tend to overfit and exhibit performance degeneration on small-scale, class-imbalanced datasets. Medical datasets are typically small and imbalanced due to the scarcity of labelled medical data samples. Therefore, this study proposes a novel stroke prognosis prediction model called Strokeformer to address these issues. Specifically, novel intra- and interfeature interaction modules are designed to capture internal and mutual information among individual features for more effective latent representations. In addition, we explore the possibility of performing the training process by pretraining on large-scale, class-balanced datasets and then fine-tuning on small-scale, class-imbalanced downstream datasets. This pretraining and fine-tuning paradigm is dramatically feasible for preventing overfitting. To verify the effectiveness of the proposed model and training method, experiments are conducted on 20 public datasets from OpenML and two private stroke prognosis datasets provided by Shenzhen Fuyong People’s Hospital and The Affiliated Taizhou People’s Hospital of Nanjing Medical University, China, respectively. The results show that Strokeformer performance significantly outperforms that of other comparison models on the introduced datasets. The principal limitation of the model lies in its lack of interpretability from the clinicians’ perspective. Nevertheless, given that the interpretability of deep learning remains an open challenge, the promising empirical results achieved by Strokeformer on real-world stroke prognosis datasets highlight its potential to assist in clinical decision-making.

Amazonian and Andean tree communities are not tracking current climate warming

Climate change is shifting species distributions, leading to changes in community composition and novel species assemblages worldwide. However, the responses of tropical forests to climate change across large-scale environmental gradients remain largely unexplored. Using long-term data over 66,000 trees of more than 2,500 species occurring over 3,500 m elevation along the hyperdiverse Amazon-to-Andes elevational gradients in Peru and Bolivia, we assessed community-level shifts in species composition over a 40+ y time span. We tested the thermophilization hypothesis, which predicts an increase in the relative abundances of species from warmer climates through time. Additionally, we examined the relative contributions of tree mortality, recruitment, and growth to the observed compositional changes. Mean thermophilization rates (TR) across the Amazon-to-Andes gradient were slow relative to regional temperature change. TR were positive and more variable among Andean forest plots compared to Amazonian plots but were highest at midelevations around the cloud base. Across all elevations, TR were driven primarily by tree mortality and decreased growth of highland (cool-adapted) species rather than an influx of lowland species with higher thermal optima. Given the high variability of community-level responses to warming along the elevational gradients, the high tree mortality, and the slower-than-warming rates of compositional change, we conclude that most tropical tree species, and especially lowland Amazonian tree species, will not be able to escape current or future climate change through upward range shifts, causing fundamental changes to composition and function in Earth’s highest diversity forests.

In silico and in vitro antibacterial evaluation of eight Anatolian Salvia species with chemical profiling by LC-HRMS

An integrated suite for strategic urban modelling: Long-term impact assessment of land use and infrastructure development

Integrated land use transport models lie at the heart of the process of strategic level urban planning where the focus is on developing sustainable plans for locating new land uses, geodemographic activities, and transport routes for various modes. Here we develop an integrated suite of models focused on the strategic planning of large metropolitan areas, upwards of one million in population, which dovetail as key parts of a wider package of modules for urban simulation. Each module acts as a plug-in, which defines the links between the strategic, tactical, and operational levels or scales associated with transport planning. Funded by the EU’s Horizon 2020 programme, the wider suite of models from the HARMONY project integrates five modules together around a core Land Use Transport Interaction (LUTI) model. In this paper, we focus on the LUTI model and its integration with a demographic forecasting model (DFM), and a regional economic model (REM). We start by outlining the model, and then we illustrate how it has been applied to the Turin metropolitan area to create different scenarios that project population, employment, and new transport infrastructure into the near to medium term future. The paper concludes by noting that although these kinds of integrated models are difficult to generalise to and build for any large metropolitan area, mainly due to differences in data availability as well as the kinds of ‘what-if’ scenarios that need to be explored through simulation in different urban areas, our focus here is on progress that has been made.

Metabolomics navigates natural variation in pathogen-induced secondary metabolism across soybean cultivar populations

Phytophthora soja e-induced root rot poses a major threat to soybean production. While the molecular mechanisms underlying soybean– P. sojae interactions have been extensively studied, their biochemical basis remains largely unexplored. Previous research has identified key metabolic modules involved in pathogen defense, but structural diversity has largely been constrained by studies on single soybean accessions. Here, we broadened the chemical search space to a diverse soybean germplasm collection using high-throughput metabolomics as a powerful tool for comprehensive metabolic profiling. Chemical classes of lipids and phenylpropanoids again retrieved the most pronounced responses upon P. sojae infection in general. A two-layer analytical strategy further finely resolved metabolites into pathogenesis-, resistance-, and tolerance-type accumulation patterns, leading to the identification of cinnamaldehyde and coumestrol as potent defense metabolites. Bioassays validated cinnamaldehyde directly and strongly inhibited cyst germination and mycelial growth, and coumestrol, a benzofuran-type metabolite, exhibited broad-spectrum activity against spore germination as an identified phytoalexin. Multiomics analyses nailed down the candidate of coumestrol biosynthesis genes, and genetically overexpression of regulatory genes ( Dir2a/4a/4b ) in hairy root systems increased coumestrol accumulation thus positively correlating with improved host resistance. Interestingly, tolerance-type compounds may serve distinct ecological roles, as exemplified by daidzein, which, despite being classified as a tolerance-type metabolite, recruits more zoospores facilitating secondary infection in fact. This study highlights a systematic approach for population-level investigations and emphasizes the necessity of integrating bioinformatics with experimental validation to accurately predict metabolite or gene ecological functions.

High performance grid structured MIMO antenna with regression machine learning for high-speed sub THz and THz 6G IoT applications

Opposing action of photosystem II assembly factors RBD1 and HCF136 underlies light-regulated <i>psbA</i> translation in plant chloroplasts

The D1 subunit of photosystem II (PSII) is subject to light-induced damage. In plants, D1 photodamage activates translation of chloroplast psbA mRNA encoding D1, providing D1 for PSII repair. Three D1 assembly factors have been implicated in the regulatory mechanism: HCF244 and RBD1 activate psbA translation, whereas HCF136 represses psbA translation in the dark. To clarify the regulatory circuit, we analyzed psbA ribosome occupancy in dark-adapted and illuminated rbd1 and rbd1;hcf136 double mutants in Arabidopsis and in Zm- hcf244 and Zm- hcf244; Zm- hcf136 double mutants in maize. The results show that RBD1 is required for light-induced psbA translation but has only a small effect on psbA ribosome occupancy in the dark. RBD1 is not required for psbA translation when HCF136 is absent, indicating that RBD1 activates psbA translation in the light by inhibiting HCF136’s repressive effect. By contrast, HCF244 is required to recruit ribosomes to psbA mRNA in light, dark, and in the absence of HCF136. We demonstrate further that HCF244 is not required for the translational activator HCF173 to bind the psbA 5’UTR. These results show that RBD1 is central to the perception of the D1 photodamage that triggers D1 synthesis and that it activates psbA translation by relieving repression by an HCF136-dependent assembly intermediate. HCF244 activates downstream of those events without impacting HCF173’s binding to psbA mRNA. The results implicate a feature of nascent D1 that is affected by both HCF136 and RBD1 as the signal that reports D1 photodamage to regulate psbA translation rate as needed for PSII repair.

Low-energy shock waves improve the bacterial detection of Staphylococcus aureus biofilms on polyethylene

Abstract With the increasing number of total joint arthroplasties and the associated increase in periprosthetic infections, the further development of non-invasive examination methods to improve bacterial detection is becoming increasingly important. This is particularly important in the case of biofilm-forming bacteria, where false-negative results from joint puncture can lead to a delay in optimal therapy, as the number of planktonic bacteria in the punctate can be low. Extracorporeal shock wave therapy, originally used in the treatment of urolithiasis, has demonstrated promising energy-dependent biofilm-disrupting and even antimicrobial properties against Staphylococcus aureus. High-energy shock waves have been shown to be effective in several studies, but they are often painful and not suitable for all patients. Utilizing shock waves could enhance pathogen detection rates and potentially enable the early initiation of targeted therapy. This study therefore investigates whether low-energy shock waves are suitable for removing bacteria from a Staphylococcus aureus biofilm on polyethylene. The aim of this study is to evaluate the applicability of this method to improve the diagnostic accuracy of periprosthetic infections. In an in vitro model, Staphylococcus aureus biofilms were cultured on polyethylene patellas for 48 h. Biofilm disruption by low-energy shock waves was tested using a ReflecTron hmt device, with shock waves applied in a range of 0–1800 impulses. Colony-forming units (CFU) and XTT assays (to quantify cell viability) were measured. Shock wave treatment with an energy of 0.13 mJ/mm2 proved to be effective in removing bacteria from Staphylococcus aureus biofilms on polyethylene surfaces. A significant increase in CFU within the surrounding solution was observed after just 100 impulses (p = 0.018), and continued to increase until approximately 900 impulses. A linear correlation was identified between the logarithm of the shock wave impulses and both the CFU (r = 0.971, p &lt; 0.001) and the XTT activity (r = 0.94, p &lt; 0.001). This finding suggests that low-energy shock waves detach living bacteria from the biofilm. Consequently, they highlight the potential of low-energy shock waves to effectively disrupt biofilms without compromising bacterial viability, reinforcing their potential diagnostic and therapeutic applications. Low-energy shock waves disrupt Staphylococcus aureus biofilms on polyethylene surfaces in vitro, dislodging bacteria from the biofilm. However, further in vivo studies are required in order to assess the potential of this method for clinical applications. Such studies could determine whether shock waves can enhance periprosthetic infection diagnosis in vivo and facilitate implant-preserving therapies for mature biofilms.

Efficient neural encoding as revealed by bilingualism

The remarkable human capacity for bilingual and multilingual acquisition raises fundamental questions about how the brain develops efficient systems for processing multiple languages. In this study, we used neural network models trained on natural speech input to examine how these efficient representations emerge. Our models show that multiple phonological systems can be organized through parallel representations, preserving the unique aspects of each language while maintaining shared articulatory features. This parallel structure scaled effectively from two to three languages without needing additional neural architecture, highlighting the inherent efficiency in multilingual processing. Furthermore, the development of phonological representations varied based on the timing of language exposure, showing how earlier-learned languages shape the acquisition of subsequent ones. These findings imply that the human ability to speak multiple languages may arise from general principles of neural organization that optimize shared resources while maintaining essential distinctions between languages. This work has important implications for language learning, brain plasticity, and cognitive development.

Improving accuracy of land-use classification through MobileNetV3 and Greedy Osprey Optimization

Action-type mapping principles extend beyond evolutionarily conserved actions, even in people born without hands

How are actions represented in the motor system? Although the sensorimotor system is broadly organized somatotopically, higher-level sensorimotor areas encode action-type information for reaching and grasping actions—regardless of the acting body part. Does the brain similarly support generalization across acting body parts for more evolutionarily recent actions, such as tool-use? We tested whether there is a body-part-independent action-type organization in sensorimotor areas by examining fMRI responses for tool-use actions that participants performed with their hands or feet. We additionally included individuals born without hands to test whether hand sensorimotor experience is necessary for the development of this action-type organization. Across analyses, we found a consistent dissociation in the motor system. The primary sensorimotor cortices encoded concrete, body-part specific information in both groups. In contrast, higher-level motor areas within the tool-use network represent abstract, action-type information independent of the body part for both groups. Together, our results suggest that the hierarchical organization of the motor system is not dependent on a long evolutionary history of an action. Further, this organization is not dependent on an individual’s manual sensorimotor experience. Our results also show that the functional reorganization in congenital handlessness follows the hierarchical organization of the intact cortex, revealing the limitations of brain plasticity. Finally, the results support using a readout of a more abstract code for hierarchical brain–computer interfaces.

Kinetic and thermodynamic investigation of the removal of alizarin red dye using silica-supported nanoscale zero-valent iron particles

Abstract Alizarin red (ARS) dye is a persistent and toxic pollutant in aquatic environments, posing a significant environmental threat. Among various treatment technologies, adsorption offers a practical and highly efficient method for dye removal. This study investigates the application of silica-supported nanoscale zero-valent iron (nZVI) particles as adsorbents for the removal of ARS dye, with a focus on evaluating the adsorption kinetics and thermodynamic behavior of the process. Two nanocomposites were synthesized using iron precursors with different counter ions: chloride (nZVI/Cl⁻) and nitrate (nZVI/NO₃⁻). The choice of counter ion influenced the physicochemical characteristics of the materials, thereby affecting their adsorption efficiency. The effects of key operational parameters, including solution pH, temperature, contact time, and adsorbent dosage, were systematically examined. Optimal adsorption was observed at pH 3, achieving removal efficiencies of 94.9% for nZVI/Cl⁻ and 85.0% for nZVI/NO₃⁻. Adsorption isotherm analysis revealed that the data fit well to the Langmuir model, indicating monolayer adsorption onto a homogeneous surface. Thermodynamic parameters confirmed that the adsorption process is spontaneous and endothermic in nature. These results underscore the potential of silica-supported nZVI nanocomposites as effective, sustainable, and environmentally friendly adsorbents for the remediation of dye-contaminated wastewater.

Profitable third-party punishment destabilizes cooperation

Third-party punishment is theorized by some scholars to be essential to the evolution of large-scale cooperation, but empirically, it often fails to bring about its desired effects. Here, we suggest that third-party punishment destabilizes cooperation when third parties have profit motives to punish. Across nine economic games and judgment experiments (including four preregistered studies), we find that when third-party punishment is profitable, rates of cooperation decrease immediately and remain lower even when punishment outcomes are optimized to support cooperative behavior. Profitable third-party punishment causes targets of punishment to anticipate antisocial punishment and perceive social norms in terms of self-interest, suggesting that the introduction of payment degrades the communicative signals that punishment is meant to convey about punishers’ intentions and social norms. Critically, participants who would benefit from increased cooperation inadvertently reduce their own monetary compensation by opting in to experimental conditions that pay punishers, suggesting that they intuitively fail to consider the signaling consequences of profit motives to punish. Implications for systems of punishment and cooperation in real-world contexts are discussed.

Retraction Note: Inverted U-shaped pattern of green finance influencing the synergistic effect of pollution and carbon reduction

Soil eDNA reflects regionally dominant species rather than local composition of tropical tree communities

Environmental DNA (eDNA) is increasingly used for biodiversity monitoring, but validation of the spatial scale(s) at which eDNA reflects extant communities is scarce, particularly in tropical forests: the terrestrial biome with the most concentrated diversity on earth. We leveraged spatially explicit tree inventory data from the 16-ha Luquillo Forest Dynamics Plot (LFDP) in Puerto Rico to validate soil eDNA as a spatially explicit indicator of tree diversity/composition. Using a comprehensive local chloroplast trnL -P6 reference library, we analyzed soil samples at multiple scales through eDNA trnL -P6 metabarcoding. We compared eDNA taxonomic diversity/composition, considering several bioinformatic thresholds, with inventory data across a range of spatial scales, as well as random points to compare observed correlations with random expectations. Despite considerable fine-scale heterogeneity in soil plant eDNA composition, we detected 53 tree Operational Taxonomic Units (OTUs) across the LFDP, corresponding to 68% of tree OTUs from the census data. Encouragingly, this equated to 98% of the total basal area (and 98% of the total stems). An initial confusion matrix evaluation suggested a highly localized eDNA signal (within 5 m of the sampled locations). However, comparison with random expectations revealed a lack of support for a fine-scale spatial signal due to misclassification (i.e., eDNA false presence or/and false absence) of relatively common taxa. Our study shows that “universal” PCR primer metabarcoding of tree eDNA in tropical soils may be useful for assessing dominant taxa at landscape scales, but not for spatially explicit characterization of rare species and community composition at local scales.

A feasibility study investigating the risk of prediabetes among children in New Zealand

Life’s homochirality: Across a prebiotic network

For centuries, scientists have been puzzled by the mystery of life’s biomolecular homochirality—the single-handedness of biological compounds. Sugars and nucleic acids are right-handed, while amino acids are left-handed in biological systems. Likewise, certain metabolites are homochiral, though their handedness varies. However, efforts to address the homochirality problem have often focused on a single compound, a single molecular class, or invoke an extraterrestrial origin. Here, we emphasize the importance of achieving homochirality across an entire prebiotic chemical network and explore a terrestrial pathway for its emergence. This pathway is supported by recent experimental results from several independent studies, as well as analyses of pristine asteroid materials. Our analysis identifies the genome as a key site for achieving network-scale homochirality on early Earth and addresses the opposite handedness of D -nucleic acids and L -peptides in biology through nonenzymatic, stereoselective coded peptide synthesis.

An enhanced approach to minimum variance unbiased velocity estimation, incorporating horizontal and vertical handoff in HetNets

Abstract This paper presents an enhanced approach to Minimum Variance Unbiased (MVU) velocity estimation in Heterogeneous-Networks (HetNets) by addressing horizontal and vertical handoffs. In HetNets, the abundance of base stations (BSs) results in frequent unnecessary handoffs and service disruptions for mobile users, posing challenges for mobility management. Accurate velocity estimation is crucial for effective mobility management. Our proposed strategy involves tracking vertical and horizontal handoffs over a specified time interval. Through mathematical modeling, we approximate the analytical expression of the handover count probability-mass-function in HetNets as Rayleigh distributed and calculate its scale parameter based on velocity, BS density, and measurement time span. We derive the Cramer-Rao lower bound (CRLB) and utilize the Neyman-Fisher factorization method to obtain the sufficient statistics. Leveraging the Rao-Blackwell-Lehmann-Scheffe (RBLS) theorem, we derive the MVU estimator. Our results demonstrate a close alignment between the proposed estimator’s variance and the CRLB. Furthermore, we observe that increased user velocity leads to higher velocity estimation variance, indicating greater challenges in accurate estimation for faster-moving users. Simulation results show that a higher BS density and longer handover measurement periods can substantially reduce velocity estimation errors, highlighting the benefits of an improved HetNet infrastructure and extended measurement durations for precise velocity estimation.