Biology subjects

Zheng, S.

Publications and source records attributed to Zheng, S..

At least 37 records · Page 2Linked to original sources

bioRxiv2026

Predicting Pre-treatment Resistance or Post-treatment Effect? A Systematic Benchmarking of Single-Cell Drug Response Models

Intratumoral heterogeneity drives variable drug responses in cancer. Single-cell RNA sequencing (scRNA-seq) enables characterization of such heterogeneity and prediction of drug response at single-cell resolution. Accordingly, various computational models have been developed to infer drug response from scRNA-seq data. However, their performance, robustness, and generalizability across different biological contexts remain insufficiently evaluated. To address this gap, we benchmarked representative single-cell drug response prediction models using 26 curated datasets comprising over 760,000 cells across 12 cancer types and 21 therapeutic agents. We constructed balanced and imbalanced scenarios to reflect realistic drug-response label distributions. To address the lack of ground-truth labels in conventional scRNA-seq datasets, we incorporated lineage-tracing data with experimentally validated drug-response annotations, enabling evaluation in a clinically relevant pre-treatment prediction setting. Our results show that prediction performance was markedly higher in cell lines than in tissue samples. Under imbalanced conditions, most methods exhibited sharp performance declines, whereas scDEAL demonstrated the highest robustness. Independent validation using an in-house pancreatic ductal adenocarcinoma dataset further confirmed scDEALs robustness and ability to capture biologically meaningful state transitions. Label-substitution experiment revealed that this robustness was partially driven by the models specific training-label construction. However, benchmarking with lineage-tracing data revealed a fundamental limitation: most models capture drug-induced transcriptional changes but struggled to predict intrinsic resistance before treatment. In summary, our study defines the performance boundaries of current approaches and highlights their limitations in addressing intratumoral heterogeneity, class imbalance, and intrinsic resistance prediction, emphasizing the need for the next-generation single-cell drug response models with stronger clinical relevance.

bioinformatics↗

bioRxiv2026

Midkine Dependency Defines a Therapeutically Actionable Vulnerability in Medulloblastoma

Medulloblastoma (MB), the most common malignant pediatric brain tumor, remains associated with high recurrence rates and substantial treatment-related morbidity despite aggressive multimodal therapy. Here, we identify midkine (MDK), a secreted growth factor/cytokine, as a previously unrecognized central regulator of MB growth and survival. Tissue microarray analysis revealed elevated MDK protein expression in MB specimens, and single-cell RNA sequencing datasets showed tumor cell-specific MDK expression. Genetic depletion or antibody-mediated neutralization of MDK suppressed proliferation and survival and induced apoptosis across multiple MB cell lines, whereas recombinant MDK supplementation exerted the opposite effect. Mechanistically, phosphokinase profiling and immunoblot analyses showed that MDK signals through multiple receptors, including nucleolin (NCL), LRP1, and syndecan-2 (SDC2), to sustain oncogenic ERK1/2 and Akt-mTOR-S6 signaling. Transcriptomic profiling following MDK silencing or depletion revealed marked suppression of ribosome biogenesis, global protein translation, and MYC-driven programs, coupled with activation of inflammatory and apoptotic responses. Consistently, mechanistic studies utilizing RPS6 staining, ribosomal RNA quantification, puromycin SUnSET, and OPP incorporation assays confirmed that loss of MDK impairs ribosome biogenesis and protein synthesis. Notably, MDK suppression also triggered robust activation of the IFN-cGAS-STING pathway, linking translational stress to innate immune signaling. In orthotopic xenograft models, CRISPR-mediated MDK knockout significantly reduced tumor growth and prolonged survival. Together, these findings establish MDK as a key integrator of oncogenic signaling, translational control, and innate immunity in MB, highlighting MDK as a compelling therapeutic target.

cancer biology↗

bioRxiv2026

VeloTrace Reconciles Divergent Velocity and Trajectory in Single-cell Transcriptomics with Deep Neural ODE

Cellular identity and fate transitions are governed by continuous molecular processes that form dynamic trajectories within a high-dimensional transcriptomic landscape. Existing methods attempt to model these dynamics from two complementary perspectives: trajectory inference and velocity modeling. Ideally, velocity and trajectory are dual aspects of transcriptomic dynamics where velocity is tangent to trajectory everywhere. This inherent connection between velocity and trajectory is currently absent in transcriptomic analysis. Splicing velocity are precision-limited to inadequately-sequenced genes, while trajectory inference prioritizes the modeling of global trends while omitting local dynamics. This divergence breaks the geometric continuity between local velocities and global trajectories, hindering the reliable interpretation of developmental dynamics. To reconcile trajectory inference and RNA velocity, we introduce VeloTrace, a framework that unifies them through Neural Ordinary Differential Equations (NeuralODEs). VeloTrace learns a continuous-time velocity field whose integral curves constitute the trajectory itself, while ensuring that velocities are tangent to integral paths everywhere. Leveraging a splicing quality score, VeloTrace incorporates high-quality splicing velocity as partial supervision for velocity orientation and grounding. During optimization, VeloTrace incorporates a Monte Carlo multi-time-frame supervision strategy to ensure coherence between local and global trajectorys and suppress sequencing-induced stochastic diffusion. Through refining the velocity field and cell-specific parameters for pseudo-time, expression, and velocity, VeloTrace reconstructs a smooth, local-and-global-coherent velocity-vector-guided flow in the transcriptomic latent space. This strategy ensures a complementary integration of velocity and trajectory, imputing the transcriptional kinetics for genes of insufficient strength, whose kinetics cannot be accurately portrayed by splicing velocity. In simulation benchmarks, VeloTrace captured the transcriptional dynamics of all expressed genes, even those with inadequate sequencing coverage, producing velocity directions that were most consistent with the true direction and every-where tangential across the entire process, outperforming state-of-the-art methods, including scVelo, UniTVelo, VeloVI and scTour. VeloTrace uniquely reconciles RNA velocity and trajectory inference, creating a velocity field where each cell can infer past and future transitions from its current state. Moreover, VeloTrace extends reliable velocity estimation to a broader set of genes. When applied to mouse neural stem cell differentiation data, it successfully recovers dynamics of driver genes for two developmental lineages, including those with low expression, shedding light on their regulatory roles during differentiation. This unified framework lays the foundation for more accurate modeling of gene regulation and cell fate decisions in complex biological systems.

bioinformatics↗

bioRxiv2026

Degradable porous PLGA/PCL membrane enable a lung alveoli-on-a-chip for modeling particulate-induced alveolar injury

Understanding how airborne particulates disrupt the alveolar barrier requires in vitro systems that recapitulate both the structure and transport properties of the lung air-blood interface. Here, we report a biodegradable lung alveoli-on-a-chip enabled by porous poly(lactic-co-glycolic acid)/polycaprolactone (PLGA/PCL) membranes with an interconnected porous architecture generated via porogen-assisted phase separation process. The membrane exhibits tunable degradation behavior, allowing progressive increases in surface porosity ([~]40%) and reduction in thickness ([~]3 {micro}m) during culture, while PCL maintains mechanical integrity under dynamic conditions. These degradation-driven structural changes regulate membrane transport properties, leading to enhanced permeability and supporting the formation of a functional epithelial-endothelial barrier under air-liquid interface (ALI) culture with breathing-mimetic cycling strain. Primary human alveolar epithelial and microvascular endothelial cells formed confluent, junctional monolayers on opposing membrane surfaces, exhibiting stable barrier function and high viability throughout the culture period. As a functional application, the platform was used to assess diesel particulate matter (DPM)-induced alveolar injury. Apical exposure to DPM induced dose-dependent cytotoxicity, increased barrier permeability, elevated reactive oxygen species, and DNA damage in both epithelial and endothelial layers, demonstrating trans-barrier propagation of particulate-induced injury. Pharmacological modulation with roflumilast-N-oxide (RNO), a phosphodiesterase-4 (PDE4) inhibitor, selectively attenuated oxidative stress and inflammatory responses, with limited effects on barrier integrity. Together, this work establishes degradable PLGA/PCL membranes as tunable interface materials for lung-on-a-chip systems, where structural evolution during degradation directly governs transport and barrier function. The resulting platform provides a physiologically relevant approach for studying particulate toxicity and therapeutic modulation at the alveolar interface.

bioengineering↗

bioRxiv2026

A biodegradable porous membrane-based lung alveoli-on-a-chip for assessing particulate-matter-induced pulmonary toxicity

Understanding how airborne particulates disrupt the human alveolar barrier requires in vitro systems that accurately replicate its composition and function. We present a biodegradable lung alveoli-on-a-chip that reproduces the architecture and physiology of the human air-blood interface using a porous poly(lactic-co-glycolic acid) (PLGA) membrane positioned between epithelium and endothelium under air-liquid interface (ALI) culture. The membrane, fabricated by porogen-assisted nonsolvent-induced phase separation, exhibited >50 % porosity, [~]2 {micro}m thickness, and mechanical compliance over 100-fold higher than conventional Transwell inserts, closely resembling the native interstitium. During co-culture, gradual PLGA degradation was compensated by cell-secreted extracellular-matrix (ECM) proteins such as collagen IV and laminin, forming a self-remodeling barrier that maintained integrity for at least 11 days. The platform supported stable epithelial-endothelial co-culture, high transepithelial electrical resistance, and physiologically relevant permeability. To demonstrate its utility, the chip was used to assess pulmonary toxicity of four types of waste-combustion-derived particulates, including rubber, plastic bags, plastic bottles, and textile fibers, delivered apically under ALI conditions. All combustion products reduced cell viability, increased hydrogen-peroxide release, and elevated {gamma}-H2AX expression, indicating oxidative and genotoxic stress, while disrupting barrier permeability. Rubber combustion particles elicited the most severe toxicity, causing the greatest loss of viability, accumulation of reactive oxygen species, and formation of DNA double-strand breaks. Together, these results establish a biodegradable, ECM-remodeling lung alveoli-on-a-chip as a physiologically relevant platform for investigating source-specific particulate toxicity and alveolar-barrier pathophysiology. By bridging environmental exposure models with human-relevant lung biology, this system provides a quantitative and translatable tool for evaluating respiratory risks and therapeutic interventions.

bioengineering↗

bioRxiv2026

CPS: Mapping Physical Coordinates to High-Fidelity Spatial Transcriptomics via Privileged Multi-Scale Context Distillation

MotivationSpatial transcriptomics enables the dissection of tissue heterogeneity within native contexts, yet current platforms are inherently constrained by high sparsity and low signal-to-noise ratios that obscure fine-grained biological signals. Current efforts to recover these signals are limited by image registration dependencies or the inherent context-blindness of implicit neural representations. ResultsWe introduce the Cell Positioning System (CPS), a context-aware implicit neural representation framework designed to map physical coordinates to high-fidelity spatial transcriptomics via a privileged multi-scale context distillation strategy. CPS treats multi-scale tissue niches as privileged information, employing a teacher network equipped with a multi-scale niche attention mechanism to capture adaptive biological interactions during training. This structural knowledge is explicitly distilled into a student coordinate network, enabling the generation of context-aware expression landscapes solely from spatial coordinates during inference. Benchmarking on the DLPFC dataset demonstrates that CPS achieves state-of-the-art performance in spatial and gene expression imputation and denoising. Furthermore, CPS enables super-resolution to recover high-resolution mouse brain anatomical details and offers interpretability by identifying the scale effective size of biological interactions within human breast cancer tissues. Finally, the framework exhibits superior scalability for large-scale datasets with linear computational complexity. AvailabilitySoftware is available online at https://github.com/tju-zl/CPS.

bioinformatics↗

bioRxiv2026

ELMO1 dependent efferocytosis protects from nephrotoxin induced acute kidney injury

Acute kidney injury (AKI) is a sudden episode of kidney failure linked to a wide range of health conditions. High mortality in AKI highlights the need to identify new therapeutic approaches. Homeostasis in multicellular organisms is exquisitely regulated by phagocytosis of apoptotic cells, also known as efferocytosis. Apoptotic cells are frequently observed at sites of inflammation, including in AKI. Engulfment and cell motility protein-1 (ELMO1) is a regulator of the actin cytoskeleton that promotes apoptotic cell removal by phagocytes during efferocytosis. Mutations in the human ELMO1 gene are linked with diabetic nephropathy and, in animal models of this disease, high ELMO1 levels promote renal dysfunction. However, the role of ELMO1 in AKI was not known. Here, we describe the links between ELMO1 and kidney pathology and test global and tissue-specific ELMO1-deficient mice in models of AKI. While global loss of Elmo1 expression did not impact the immediate loss of renal function after ischemia-reperfusion elicited AKI, ELMO1 deficiency resulted in increased tissue injury in AKI caused by cisplatin injection. Cisplatin induced robust renal cell apoptosis that was significantly elevated in mice with the global loss of ELMO1, but not in mice with the macrophage-specific Elmo1 deletion. Using primary cell culture and immunofluorescence approaches, we highlight the role of ELMO1 in efferocytosis by several renal cell types, suggesting possible additive effects during nephrotoxic injury.

pathology↗

bioRxiv2026

A Long-Context Generative Foundation Model Deciphers RNA Design Principles

Programmable design of RNA sequences with defined functions remains a central challenge in biology. Despite recent advances, existing RNA generative models lack robust controllable design capabilities and are constrained by short context windows, limiting their capacity to model the complex evolutionary manifold of full-length transcripts. Here, we introduce EVA (Evolutionary Versatile Architect), a long-context generative RNA foundation model trained on over 114 million full-length RNA sequences spanning the full breadth of evolutionary diversity. By integrating a Mixture-of-Experts architecture with an 8,192-token context window, EVA learns a unified, evolution-consistent representation of RNA sequence space that captures both global transcript structure and fine-grained functional elements. EVA establishes a unified framework for a broad spectrum of RNA design tasks within a single architecture, ranging from mutational fitness prediction to context-aware sequence engineering. We demonstrate EVAs versatility through the de novo design and targeted optimization of diverse RNA classes, including tRNAs, aptamers, and CRISPR guide RNAs, as well as mRNA and circular RNA therapeutics. In systematic evaluations, EVA achieves state-of-the-art performance across seven of nine established benchmarks, improving structural modeling accuracy by up to an order of magnitude compared to existing approaches. EVA is fully open-source at the following repository: https://github.com/GENTEL-lab/EVA.

molecular biology↗

bioRxiv2026

Automated Cyclic Super-Resolution Microscopy for Nanoscale Protein Mapping

Nanoscale mapping of multiple molecular targets is essential for decoding cellular architecture, but current approaches are limited by low throughput, intensive manual intervention, and signal variability across imaging cycles. Here, we introduce CycSTORM, an integrated super-resolution imaging platform for automated cyclic (d)STORM that enables multiplexed nanoscale protein mapping in single cells. To overcome the instability inherent in multi-day imaging, CycSTORM combines automated fluidic exchange, active 3D drift correction, and an oxygen-excluded imaging environment to stabilize fluorophore blinking and enables sub-5 nm registration across day-long experiments. Furthermore, CycSTORM incorporates a rapid chemical inactivation step using meta-chloroperoxybenzoic acid (mCPBA), eliminating >99.9% residual fluorescence within 10 minutes and minimizing inter-cycle crosstalk while preserving sample integrity. By standardizing labeling to Alexa Fluor 647, a fluorophore with stable and well-characterized blinking behavior, CycSTORM minimizes fluorophore-dependent variation and provides a robust platform for consistent localization precision across cycles, enabling reliable mapping of protein organization in single cells. Using CycSTORM, we simultaneously map multiple protein targets within the same cells with nanometer precision. Together, these advances transform cyclic super-resolution imaging into a scalable approach for quantitative nanoscale mapping of molecular organization in single cells.

cell biology↗

bioRxiv2026

Rewiring of 3D Enhancer-Promoter Interactome Underlies Diabetic Endothelial Dysfunction

BackgroundDiabetic vascular complications are driven by endothelial dysfunction, yet the role of 3D genome organization in this process is unknown. We sought to define the alterations in chromatin architecture in diabetic endothelium and identify the key regulators involved. MethodsWe generated a high-resolution 3D epigenomic atlas of diabetic endothelial cells from mouse models and human subjects using H3K27ac HiChIP, complemented by ChIP-seq, ATAC-seq, and RNA-seq. A human cohort was used to assess protein expression in diabetic versus non-diabetic endothelial cells. To identify JUNB-interacting proteins, we performed rapid immunoprecipitation mass spectrometry of endogenous proteins (RIME), with protein-protein interaction validated by co-immunoprecipitation. Functional validation was performed using in vitro, ex vivo, and in vivo approaches, including endothelial-specific knockdown in a diabetic hindlimb ischemia model. ResultsMulti-omics profiling revealed extensive enhancer reprogramming in diabetic endothelium, with AP-1 binding motifs being consistently and selectively enriched in downregulated enhancers across three distinct diabetic models. Analysis of a human cohort confirmed significantly reduced JUNB protein levels in diabetic endothelial cells. We identified widespread disruption of JUNB-anchored enhancer-promoter interactions, which underlies transcriptional repression of key endothelial genes. RIME and co-immunoprecipitation established the E3 ubiquitin ligase RBBP6 as a direct JUNB interactor that promotes its polyubiquitination and proteasomal degradation in response to hyperglycemia. Human cohort analysis further showed reciprocal elevation of RBBP6 in diabetic endothelial cells. Either JUNB overexpression or RBBP6 knockdown restored enhancer-promoter connectivity, reactivated vasoprotective transcriptional programs, and rescued endothelial function. Critically, endothelial-specific knockdown of Rbbp6 in diabetic mice restored endothelium-dependent vasorelaxation and improved perfusion recovery after hindlimb ischemia, independent of systemic glucose levels. ConclusionsOur study unveils a novel mechanism whereby hyperglycemia induces enhancer reprogramming and disrupts endothelial 3D genome architecture through RBBP6-mediated degradation of JUNB. The RBBP6-JUNB axis is established as a crucial link between metabolic stress and epigenomic reprogramming in vascular disease, presenting a promising therapeutic target for diabetic vasculopathy.

genomics↗

bioRxiv2026

Retrotransposons Promote Clonal Hematopoiesis Through Aging-Related Stromal Inflammation

Clonal hematopoiesis (CH) is an age-related phenomenon driven by the expansion of mutant hematopoietic stem cell (HSC) clones, most commonly harboring mutations in DNMT3A. While inflammation is known to promote CH, the upstream triggers of this inflammatory state remain unclear. We show that aging selectively upregulates retrotransposons in the non-hematopoietic cell compartment of the murine bone marrow, particularly in mesenchymal stromal cells. Using in vivo competitive transplant models, we demonstrate that retrotransposon-induced inflammation cell-extrinsically promotes Dnmt3a-mutant HSC expansion. This competitive advantage arises from mutant HSC resistance to inflammation-driven cell cycle perturbations. Mechanistically, we show that retrotransposon activation induces type I interferon signaling via viral mimicry, such that stromal knockdown of either Irf3 or Sting abrogates the competitive advantage of Dnmt3a-mutant HSCs. Our findings establish stromal-selective retrotransposon reactivation as a previously unrecognized, non-cell-autonomous source of inflammation that contributes to age-associated CH. Key PointsO_LIRetrotransposons remodel the aging niche through viral mimicry pathways to promote Dnmt3a-mutant clonal hematopoiesis. C_LIO_LIRetrotransposon-mediated inflammation impairs wild-type HSC fitness, while Dnmt3a-mutant HSCs resist inflammatory changes. C_LI

cancer biology↗

bioRxiv2026

Multi-agent reasoning enables predictive design of living materials

Artificial intelligence is increasingly used to accelerate scientific discovery, but most successful frameworks operate within well-defined molecular, protein or materials spaces. Living materials present a more formidable computational problem because functions emerge from context dependent coupling among cells, matrices, fabrication processes and evaluation conditions. Here we introduce LiveMat, a multi-agent reasoning framework that transforms unstructured literature into a computable design space for living materials. LiveMat standardizes 34,215 living material records, integrating 16,769 microorganism and 17,446 polymer entries into a knowledge graph linking living components, abiotic matrices, functional outputs, evaluation contexts and performance metrics. Benchmarking across five large language models shows that living material reasoning is limited mainly by cross-domain feature integration rather than coarse classification. LiveMat overcomes this limitation through constraint decomposition, provenance-aware extraction, consistency checking and expert-anchored ranking. In a prospective wound-healing task, it prioritizes a four-component design with state-of-the-art in vivo performance, establishing a scalable infrastructure for interpretable, evidence-grounded living material discovery.

biochemistry↗

bioRxiv2026

MNS induces antiviral protection and suppresses inflammation

BackgroundIdentifying safe and broad-spectrum antiviral and anti-inflammatory agents remains an urgent need in infectious and inflammatory diseases. Here, we demonstrated that MNS (NSC170724), a small-molecule nitrovinyl benzodioxole, enhanced antiviral defense while limiting excessive inflammation. MethodsThe antiviral activity of MNS was evaluated in multiple cell lines and mouse infection models across DNA and RNA viruses. Virus-induced and LPS-induced inflammatory responses were assessed using RT-qPCR, ELISA and western blotting. Bulk RNA-seq and ATAC-seq were performed to define transcriptional and epigenetic mechanisms. ResultsMNS significantly suppressed viral infection in vitro and improved survival in four lethal viral infection models, accompanied by reduced viral loads and attenuated tissue injury. MNS also diminished virus-triggered and LPS-triggered inflammatory cytokine production in macrophages and multiple mouse organs, and protected mice from LPS-induced endotoxic lethality. Multi-omics profiling showed that MNS broadly repressed LPS-induced inflammatory transcriptional programs and reversed chromatin accessibility gains across promoters and transcription start sites. Joint analysis of RNA-seq and ATAC-seq data demonstrated consistent downregulation of pivotal inflammatory pathways, such as NF-{kappa}B, Toll-like receptor, and TNF signaling. ConclusionsWith potent activity against viral replication and inflammation in cellular and animal models, MNS emerges as a promising candidate for the treatment of viral infections and hyperinflammatory conditions.

systems biology↗

bioRxiv2026

A ML-framework for the discovery of next-generation IBD targets using a harmonized single-cell atlas of patient tissue

Target discovery for IBD has traditionally relied on genetic associations, which lack the cellular resolution needed to identify novel, actionable, cell type-specific disease pathways. Here, we describe an integrated analytical and experimental framework that leverages harmonized single-cell data to systematically discover novel therapeutic strategies for IBD. We used AMICA DBTM, Immunais harmonized database of single-cell RNA datasets to construct a harmonized 1 million single-cell atlas of the human intestine. We applied a machine learning framework (Immune Patient Representation, IPR) to identify disease-associated transcriptional programs and cell type-specific gene targets. Candidate targets were prioritized using atlas-derived metrics, refined using custom criteria emphasizing translational actionability, and validated across independent clinical cohorts. Select candidates were evaluated in human primary-cell models reflecting the targets cell-type context. The IPR framework identified 85 disease-associated transcriptional programs and ranked 400 cell type-specific target genes across immune and stromal lineages. Disease-associated programs were interpreted using a structured AI-assisted reasoning framework for structured biological reasoning, linking them to IBD-relevant pathways and guiding the identification of novel, promising gene targets. Functional validation of two cell-type-specific candidates, PTGIR in myeloid cells and IL6ST in fibroblasts, confirmed the reduction of inflammatory and fibrotic pathways linked to IBD pathology. Multi-omic profiling and projection of in vitro phenotypes to patient datasets demonstrated the reversal of disease-associated programs via mechanisms distinct from those of existing biologics. Our single-cell anchored, machine-learning framework integrates in silico discovery with experimental validation, revealing new cell type-specific therapeutic opportunities and supporting a scalable approach for precision target discovery in IBD and other immune-mediated diseases.

bioinformatics↗

bioRxiv2026

Novel repressors of cambium activity in Arabidopsis

Wood is the greatest reservoir of terrestrial biomass and an essential carbon sink. Formed of xylem, it is derived from the cambium, a meristematic zone within plant stems from which phloem also forms. In Arabidopsis, cell division within the cambium is promoted by three major factors: auxin, cytokinin, and the TDIF-PXY ligand-receptor pair. Meristems and other stem cell populations are typically regulated by a balance between cell division-promoting factors and those that repress cell division to control meristem size, however few factors with cambium-repressing activity are known. Here we combined transcriptomics and transcriptional network analysis, which led to identification of related homeodomain zinc-finger transcription factors, ATHB23, ATHB30, and ATHB34, that repress cambium activity. These factors inhibit cambium activity by directly binding of promoters from a subset of auxin, cytokinin and TDIF-PXY transcriptional target genes, resulting in attenuation of their transcription. Our findings thus reveal a new mechanism underpinning balanced cambium activity.

plant biology↗

bioRxiv2026

Melatonin alleviates acid-induced stress in peanut (Arachis hypogaea L.) through physiological, antioxidative regulation and molecular defense mechanisms

Acidic stress severely restricts crop growth by disrupting nutrient uptake, redox homeostasis, and membrane stability, yet mitigation strategies remain limited. Here, we investigated the role of melatonin (MT) in regulating growth, photosynthesis, oxidative stress, antioxidant defense and proton transport in peanut seedlings under controlled hydroponics acidic (pH 4.0) and near-optimal (pH 6.5) conditions, and validated these findings in naturally acidic field soil (pH 4.3-4.5). Acid stress markedly reduced biomass accumulation, chlorophyll content, and redox balance, while enhancing ROS (H2O2) and lipid peroxidation (MDA). Exogenous MT application, particularly at 50-100 {micro}M, significantly improved shoot and root biomass, restored chlorophyll pigments and reduced H2O2 and MDA accumulation, with more pronounced effects under pH 4.0 than pH 6.5. MT strongly activated antioxidant enzymes (SOD, CAT, APX), while POD activity declined, reflecting melatonins dual role as both a direct ROS scavenger and a regulator of enzymatic redox networks. Notably, MT induced strong, dose-dependent upregulation of HL-ATPase genes (AH1 and AH2) in both leaves and roots under acidic conditions, suggesting enhanced proton extrusion, intracellular pH homeostasis, and stress adaptation. The soil validation experiment confirmed the agronomic relevance of these findings, where MT dose-dependent concentrations improved germination, vegetative growth, chlorophyll fluorescence (Fv/Fm), and yield-related traits under natural acidic conditions. Although MT also conferred benefits at pH 6.5, responses were generally moderate compared with acid stress. Collectively, these results demonstrate that MT enhances peanut tolerance to acid stress across both controlled and natural field-relevant environments, highlighting its potential application for sustainable crop production on low-pH soils.

plant biology↗

bioRxiv2025

Targeting Putrescine Homeostasis with Lactiplantibacillus plantarum PS128 to Mitigate Autism-like Behaviors in Fmr1 KO Mice

BackgroundAutism spectrum disorder (ASD) is biologically heterogeneous and has limited mechanism-informed interventions targeting core behavioral symptoms. Emerging evidence suggests that gut microbial metabolism shapes neurobehavioral outcomes, yet the specific metabolic pathways linking gut ecology to brain function remain incompletely understood. One candidate pathway is polyamine metabolism, which links microbial amino acid metabolism to host regulation and represents a plausible but underexplored contributor to ASD-related phenotypes. Notably, the psychobiotic Lactiplantibacillus plantarum PS128 has been reported to improve social behaviors in individuals with ASD, although the molecular basis for these effects is unclear. In this study, we used Fragile X mental retardation 1 knockout (Fmr1 KO) mice, a well-established ASD model, to investigate microbiota-dependent metabolic mechanisms underlying autism-like behaviors. ResultsWe found that autism-like behaviors in Fmr1 KO mice were associated with gut dysbiosis, impaired intestinal barrier integrity, disruption of the arginine-ornithine-polyamine pathway, and elevated putrescine in the prefrontal cortex (PFC). Supplementation with PS128 remodeled the gut microbiota, reduced inflammation- and disease-associated taxa, improved intestinal structure and permeability, and restored polyamine homeostasis. Targeted metabolomics revealed an increased PFC-to-serum putrescine ratio in KO mice, which was normalized following PS128 intervention. This correction was accompanied by reduced expression of polyamine transporters in the PFC, including ATP13A family members. Causal experiments supported a functional role for putrescine, as peripheral elevation of putrescine induced autism-like behaviors in wild-type mice, whereas pharmacological inhibition of putrescine synthesis ameliorated behavioral deficits in Fmr1 KO mice. ConclusionsThese findings identify putrescine metabolic dysregulation as a key contributor to autism-like phenotypes and support the existence of an ASD subtype defined by disruption of the arginine-ornithine-polyamine axis. Integrated multi-omics and causal perturbation analyses support a model in which microbiota-targeted intervention rebalances systemic polyamine regulation along the gut-brain axis, thereby improving ASD-relevant behaviors. Our work provides mechanistic evidence linking microbial metabolism to neurochemical homeostasis and highlights the translational potential of metabolic stratification in ASD. Graphical AbstractWe identify dysregulated microbiota-driven polyamine homeostasis as a causal contributor to autism-like behaviors in Fmr1 knockout mice. A psychobiotic intervention restores gut-brain metabolic balance and rescues social and repetitive behavioral deficits through coordinated microbial and host transport regulation. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=200 SRC="FIGDIR/small/694366v2_ufig1.gif" ALT="Figure 1"> View larger version (53K): org.highwire.dtl.DTLVardef@e1e780org.highwire.dtl.DTLVardef@125a3baorg.highwire.dtl.DTLVardef@201228org.highwire.dtl.DTLVardef@1880650_HPS_FORMAT_FIGEXP M_FIG C_FIG

neuroscience↗

bioRxiv2025

Lessons learned from a Kaggle challenge for particle picking in cryo-electron tomography

The difficulty of particle picking in cryo-electron tomography remains a barrier to routine in situ structure determination. Machine learning is well-suited to overcome this bottleneck with efficient algorithms that generalize across molecular species. To spur new algorithm development, we held a three-month Kaggle challenge that tasked contestants with annotating five molecular species across hundreds of experimental tomograms. This competition successfully engaged >1000 participants from diverse fields and delivered particle pickers that outperformed existing state-of-the-art. Systematic comparisons of the contestants submissions revealed the tolerance of subtomogram averaging to moderate but not severe over-picking and underscored the need for more robust measures of annotation quality. The winning models also highlighted the importance of data augmentation to overcome limited training data. All competition tomograms along with the ground truth and winning teams annotations have been released on the CryoET Data Portal as a resource to benchmark current and future tools for particle picking.

cell biology↗