The artificial intelligence (AI)-driven generation of genetic sequences holds transformative potential for addressing global challenges in agriculture, medicine, and bioenergy. Traditional approaches including hybridization, mutagenesis, and CRISPR-based editing enable targeted modification of endogenous DNA, yet remain constrained by natural sequence diversity. We here introduce PlantGFM, an application of the Hyena operator within a plant-oriented genomic foundation model, which was pre-trained on 10.84 billion nucleotides from 12 plant species and supports long-context (64 kb) prediction and sequence generation within a unified architecture. After fine-tuning on 10 annotated plant genomes, PlantGFM matched or exceeded the performance of specialized gene prediction tools. Beyond reproducing natural genes, it enables de novo design of novel candidates through the emergence capability of AI. Seven candidates selected through an AI-Human Knowledge fusion screening pipeline all showed transcriptional activity in Nicotiana benthamiana, two with stable protein expression-representing the first demonstration of DNA-RNA-protein expression of Large Language Model-generated sequences in plants. As a proof of concept, PlantGFM also exhibits emergent abilities in generating plant NLR genes. Our findings establish the feasibility of LLM technology for de novo plant gene design, providing a foundation for plant synthetic biology and AI-assisted breeding.
Shoot apical meristem (SAM) homeostasis integrates environmental and genetic cues to regulate growth dynamics that drive biomass accumulation and crop yield; however, no robust, non-destructive, quantitative proxy has been established for modeling or monitoring SAM-homeostasis-associated dynamics. Here, we developed a novel robot-based 3D imaging system and a custom pot-chamber gas exchange system to non-destructively measure plant occupation volume (POV) and whole-plant photosynthetic rate in wild-type Arabidopsis plants and nine mutants with disrupted SAM homeostasis. We demonstrate that POV robustly captures 3D plant architecture, whereas whole-plant photosynthetic rate serves as a superior proxy for optimal growth dynamics and final biomass associated with SAM homeostasis, outperforming conventional traits such as leaf number, leaf size, total leaf area, and rosette diameter. The strong positive correlations among POV, whole plant photosynthesis, and biomass accumulation establish a powerful new framework for quantitative studies of SAM homeostasis and data-driven evaluation of plant architecture.
暂无摘要（点击查看详情）
Single-molecule fluorescence in situ hybridization (smFISH) has emerged as a powerful tool to study gene expression dynamics with unparalleled precision and spatial resolution in a variety of biological systems. Recent advancements have expanded its application to encompass plant studies, yet a demand persists for a simple and robust smFISH method adapted to plant tissue sections. Here, we present an optimized smFISH protocol (cryo-smFISH) for visualizing and quantifying single mRNA molecules in plant tissue cryosections. This method exhibits remarkable sensitivity, capable of detecting low-expression transcripts, including long non-coding RNAs. Integrating a deep learning-based algorithm in our image analysis pipeline, our method enables us to assign RNA abundance precisely in nuclear and cytoplasmic compartments. The method also enables a robust combination with immunofluorescence, as cryosectioning enhances antibody penetration. This allows for the sequential visualization and quantification of both RNAs and endogenous proteins within the same cells. Finally, this study demonstrates the use of smFISH to validate single-cell RNA sequencing (scRNA-seq) expression patterns in plant tissues. By extending the smFISH method to plant cryosections, an even broader community of plant scientists will be able to exploit the multiple potentials of quantitative transcript analysis at cellular and subcellular resolutions.
Circular RNAs (circRNAs) are noncoding RNAs formed by back-splicing, characterized by covalently closed-loop structures with enhanced stability. Although growing evidence highlights their regulatory roles in plants, the precise biological functions of circRNAs remain largely unclear. In this study, we identified Arabidopsis thaliana circP5CS1, a circRNA derived from the delta-1-pyrroline-5-carboxylate synthase (P5CS1) gene through back-splicing of exons 11 to 13. The junction between exons 11-13 generates a de novo miR167 binding site, validated by sequence alignment and RNA pull-down assays, which indicate that circP5CS1 preferentially binds miR167a/b. circP5CS1 acts as a miR167 sponge to upregulate auxin response factors 6 (ARF6) and ARF8, fine-tuning salicylic acid (SA)- and jasmonic acid (JA)-mediated immunity during Pseudomonas syringae pv. tomato (Pst) DC3000 infection. In addition, circP5CS1 suppresses its host gene P5CS1, a rate-limiting enzyme for proline biosynthesis, through both splicing competition and post-transcriptional mechanisms, thereby linking circRNA biogenesis with gene expression control. Our results suggest that circP5CS1 directly modulates plant immunity via both hormone and proline pathways. circP5CS1 expression was induced by both Pst DC3000 and AvrRpt2 infections. Consistent with this, resistance to these pathogens was compromised in circP5CS1-OE, p5cs1, and miR167-STTM plants, whereas circP5CS1-RNAi, P5CS1-1-OE, and miR167-OE showed enhanced resistance. Proline levels increased in P5CS1-1-OE and circP5CS1-RNAi plants but decreased in circP5CS1-OE and p5cs1 mutants, supporting a link between circP5CS1 activity and proline metabolism. Our findings suggest that circP5CS1 may be an immuno-hub integrating hormone signaling and proline homeostasis through miRNA sponging and host gene suppression, revealing its dual regulatory role in coordinating RNA-mediated regulation with plant immune response.
The unfolded protein response (UPR) is a highly coordinated signaling network that mitigates endoplasmic reticulum (ER) stress, a condition induced by diverse environmental challenges in plants. Over the past two decades, substantial progress has been made in elucidating the molecular and genetic mechanisms of ER stress sensing and signal transduction in plants, largely through studies in the model Arabidopsis thaliana. These advances have established the UPR as a central regulator of proteostasis and underscored its broader relevance to plant growth, development, and crop productivity under stress conditions. Despite this progress, critical knowledge gaps remain, particularly concerning the downstream biological processes required for growth recovery once ER stress is subsided and how these processes are coordinated by UPR regulators. Recent systems-level and integrative studies have begun to reveal critical roles of UPR signaling in pathways governing growth re-establishment and homeostasis of nutrient allocation and energy metabolism. In this review, we highlight recent findings on the functional roles of the plant UPR in recovery from ER stress, with a focus on mechanisms mediated by the UPR regulators and downstream biological pathways that enable the transition from stress mitigation to growth restoration. Although this research area is still emerging, accumulating evidence supports a model in which the UPR functions as a dynamic regulatory network that actively coordinates post-stress physiological recovery and plant fitness.
Structural elucidation of unknown metabolites remains a fundamental bottleneck in plant metabolomics, where the vast chemical diversity of plant secondary metabolites far exceeds the coverage of existing spectral libraries. Here, we present DeepMASS v2, a substantially enhanced platform for LC-MS/MS-based compound annotation designed to address this challenge at scale. DeepMASS v2 leverages a semantic representation model trained on millions of spectra from GNPS, NIST, and in-house resources. By integrating Spec2Vec-based embeddings with Hierarchical Navigable Small World (HNSW) graph retrieval and a unified chemical space defined by molecular fingerprints, DeepMASS v2 identifies structurally related neighbors for unknown spectra and ranks candidate structures based on spatial proximity to these predicted chemical contexts. Benchmarking using CASMI datasets and a curated natural product collection demonstrates that DeepMASS v2 outperforms state-of-the-art in silico annotation tools including SIRIUS, CFM-ID, MetFrag, and MS-Finder. Importantly, DeepMASS v2 maintains strong performance for metabolites absent from spectral libraries, highlighting its capability for genuine unknown discovery. Application of DeepMASS v2 to large-scale plant datasets further reveals its power to expand the accessible metabolome space. Delivered as an intuitive web platform, DeepMASS v2 provides the community with a scalable, interpretable, and high-throughput solution for structural annotation, enabling more comprehensive characterization of plant chemical diversity and accelerating natural product discovery in molecular plant science. The web server of DeepMASS v2 can be accessed through http://deepmass.cn.
Plants associate with diverse microbial communities that influence growth and health. Although the plant immune regulatory network balances defense activation and microbial accommodation during pathogen attack, how it coordinates beneficial plant-microbe interactions across complex microbial contexts remains unclear. Here, we performed a systematic screen of 39 immune-pathway mutants using individual plant growth-promoting bacteria (PGPBs), binary combinations, synthetic and natural communities. We identified the bik1-1 mutant as exhibiting a broad defect in growth promotion across multiple beneficial microbial systems. Extensive genetic analyses using independent BIK1 CRISPR and T-DNA insertion alleles, as well as overexpression lines, demonstrated that the growth promotion defect observed in the bik1-1 line is not caused by loss of BIK1 function but instead correlates with a chromosomal fragment duplication. Although ISR and beneficial bacteria-mediated pathogen protection remain intact in bik1-1, immune activation is elevated during beneficial interactions. Microbiome profiling revealed reduced diversity and altered community structure, and microbiome transfer experiments indicate that host immune status influences the selection of microbial taxa associated with growth promotion. These findings indicate that immune signaling balance is a key determinant of plant-microbiome compatibility across diverse microbial contexts.
Salicylic acid (SA) is essential for plant immunity, but excessive SA accumulation accelerates leaf senescence, necessitating tight control of its biosynthesis. Although AVRPPHB SUSCEPTIBLE3 (PBS3) is a key enzyme in SA biosynthesis, how PBS3 abundance is regulated to coordinate immunity and longevity remains unclear. Using genetic, biochemical, and physiological analyses, we show that PBS3 functions as a quantitative regulator of the immunity-longevity balance. Loss of PBS3 compromises disease resistance but delays senescence, whereas graded increases in PBS3 abundance progressively enhance pathogen-induced SA accumulation, systemic acquired resistance (SAR), and senescence severity. We further identify the E3 ubiquitin ligase PLANT U-BOX PROTEIN 13 (PUB13) as a direct regulator of PBS3. PUB13 physically associates with PBS3 and promotes its polyubiquitination and degradation through the 26S proteasome pathway. Disruption of PUB13 stabilizes PBS3, resulting in elevated SA accumulation, enhanced SAR, and accelerated leaf senescence. Time-course analyses revealed that pathogen-induced PBS3 accumulation and SA biosynthesis are transient in wild-type plants but remain elevated in pub13 mutants, indicating that PUB13 promotes the attenuation of immune-associated SA production after defense activation. Together, our findings establish the PUB13-PBS3 module as a post-translational mechanism that fine-tunes SA biosynthesis, enabling effective immunity while preventing prolonged SA accumulation and its detrimental effects on plant longevity.
Plant diseases cause 20-40% annual crop losses worldwide, yet conventional detection methods remain slow, subjective, and inaccessible to smallholder farmers. This work presents GreenAid, an end-to-end plant disease detection and management system that bridges the gap between laboratory-level deep learning performance and practical agricultural deployment. The system integrates a confidence-weighted ensemble of three CNN architectures (VGG16, ResNet50, InceptionV3), fused through per-class F1-score reliability weights, with a cross-platform mobile application supporting offline inference via TensorFlow Lite, a web-based analytics dashboard, and an NLP-powered chatbot. On the PlantVillage benchmark (87,000 images, 38 classes, 14 species), the ensemble achieves 98.74% accuracy and 98.48% F1-score. Systematic comparison of six fusion strategies confirms that per-class F1 weighting outperforms alternatives including majority voting, simple averaging, and stacking. The INT8-quantised deployment model (78 MB, 127 ms on a mid-range smartphone) retains 98.43% accuracy with per-class analysis confirming disproportionate impact on the five most challenging categories. All pairwise model comparisons are validated by McNemar's test ([Formula: see text]). The primary contribution is the complete, reproducible integration of competitive classification, edge deployment, and an end-to-end agricultural delivery pipeline (mobile application, web dashboard, and NLP chatbot) rather than the ensemble mechanism itself.
Multiplex genome editing is a powerful approach for dissecting gene networks and engineering complex traits in crops because it enables the simultaneous modification of multiple genomic loci. However, achieving high editing efficiency across multiple targets remains a significant challenge. To address this, we developed an optimized CRISPR system for rice that combines a monomeric TREX2-SpCas9 fusion with a novel array of tRNA-based gRNA processing elements. The TREX2-SpCas9 fusion significantly enhanced editing performance, resulting in higher editing efficiency, larger deletions, and increased mutation frequencies compared with wild-type SpCas9 and other exonuclease fusions. By systematically evaluating 38 endogenous rice tRNA genes, we identified 13 high-performing candidates, including tRNALeu-1 and tRNAPro-1, that outperformed the widely used tRNAGly and tRNAMet elements, enabling highly efficient processing of multiplexed gRNA arrays. Incorporating these top-performing tRNAs into our system enabled simultaneous editing of up to 29 OsCPK genes in a single rice plant. Furthermore, we demonstrated the cross-species applicability of this platform in the dicot Nicotiana benthamiana using transient expression, where rice-derived tRNA elements facilitated high-efficiency editing. This optimized multiplex gene-editing system provides a robust, scalable platform for accelerating plant functional genomics and engineering complex agronomic traits.
Plant cell walls are dynamic signaling platforms that integrate mechanical, biochemical, and immune cues during development and in response to environmental stress. Among wall polysaccharides, pectin has emerged as a central regulator of cell wall integrity and stress-responsive signaling. Here, we highlight pectin methylesterases and their regulators as a coordinated modulatory system controlling pectin remodeling in space and time during biotic and abiotic stresses. Recent discoveries on isoform-specific pectin methylesterase functions, proteolytic activation, and trafficking routes are discussed. During biotic interactions, pectin methylesterase activity promotes pectin remodeling for cell wall reinforcements. It also contributes to cell wall integrity sensing by supporting the generation of methanol and, indirectly, oligogalacturonides, and by enabling the formation of homogalacturonan structures that facilitate mechanosensor and receptor binding. Under abiotic stress, pectin methylesterase-driven de-methylesterification influences cell wall stiffness, mechanics, porosity, and ion homeostasis. This review highlights pectin methylesterases as central regulators of pectin dynamics that underlie cell wall remodeling, signaling, and stress responses, identifies key knowledge gaps and outlines future directions to enhance plant stress resilience.
Singlet oxygen (1O2) is a highly reactive chloroplast-derived reactive oxygen species (ROS) that functions not only as a damaging oxidant but also as a signaling molecule controlling gene expression, stress responses, and development. The EXECUTER (EX) family, comprising EX1 and its paralog EX2, is central to 1O2-triggered retrograde signaling in Arabidopsis thaliana. Recent mechanistic advances have revealed that EX1 undergoes post-translational oxidation, is proteolytically turned over, and that its small peptides interact with TIC/TOC translocons to activate signaling cascades. EX1 also exhibits unexpected functions in darkness, where it modulates auxin-responsive gene networks. Meanwhile, comparative genomic analyses further show that EX-like proteins containing the conserved SOS (Singlet Oxygen Sensor) domain are present in green algae. In contrast, the complete EX architecture, including both SOS and UvrBC-like domains, emerged in charophytes and subsequently diversified across major plant lineages. In this review, we present classical and emerging insights into EX-mediated signaling, highlight evolutionary patterns of EX proteins across the plant kingdom, and discuss how understanding EX1/EX2 dynamics may inform strategies to engineer ROS perception, stress tolerance, and crop growth optimization. We propose a unified model of EX-mediated 1O2 sensing and outline key unresolved questions that will guide future research.
Abscisic acid (ABA) is a central regulator of plant adaptation to abiotic stress, balancing stress responses with growth and development through the dynamic modulation of ABA signaling. Rapid activation of ABA signaling is essential for enhancing stress tolerance, whereas prolonged stress requires timely attenuation of the pathway to restore growth and prevent excessive defense responses. Recent studies have uncovered diverse mechanisms underlying ABA desensitization, including regulation by SnRK2 kinases, phytohormone crosstalk, nutrient signaling, protein trafficking, post-translational modifications, and feedback regulatory networks. Together, these interconnected mechanisms enable plants to fine-tune ABA signaling in response to developmental and environmental cues. In this review, we summarize recent advances in the molecular mechanisms that attenuate ABA signaling and rebalance growth and stress adaptation during prolonged stress. We also highlight questions and discuss strategies for engineering ABA signaling dynamics to improve crop resilience, productivity, and climate adaptation in increasingly variable environments.
Enhancing cotton yield remains a paramount breeding objective. Given limited arable land, increasing planting density is an effective strategy to boost cotton yield. However, the genetic basis of plant architecture suitable for high-density planting, particularly the molecular mechanism controlling fruit branch angle (FBA), remains largely unknown. This study identified qFBA-A11, a major QTL regulating FBA, using a Gossypium hirsutum × G. mustelinum introgression lines population. Map-based cloning revealed that GhFBA1_At was the major gene positively regulating FBA. GhFBA1_At encodes a protein with no known functional domains. Further investigations identified that GATA5, a light-responsive transcription factor, is a positive regulator of GhFBA1_At expression. The GhFBA1_At protein inhibited KNAT7 accumulation through direct protein-protein interaction and downregulated the expression of cell wall biosynthesis-related genes. This process promoted cell expansion while reducing cell wall thickness, ultimately weakening the mechanical strength of the cell wall and leading to a loose plant architecture. A structural variation (SV) at the GhFBA1_At locus in G. mustelinum caused complete gene loss, resulting in a compact plant architecture. Phylogenetic analysis showed that this SV was unique to G. mustelinum. CRISPR-Cas9 editing of GhFBA1_At generated a compact plant architecture and enhanced cotton yield under high-density planting. Our findings established GhFBA1_At as a crucial regulator of FBA, elucidated its molecular mechanism, and provided valuable germplasm resources for ideal plant architecture breeding in cotton.
暂无摘要（点击查看详情）
暂无摘要（点击查看详情）
Industrial Control Systems (ICS) face escalating cyber threats as adversaries increasingly exploit artificial intelligence (AI) to evade conventional defenses. This paper introduces a Digital Twin-enhanced security framework in which a real-time, physics-consistent virtual replica of the controlled industrial process is synchronized with sensor and actuator telemetry from the physical plant and used to validate, suppress, or confirm anomaly scores produced by a deep-learning ensemble. The physical twin is the closed-loop ICS plant (water treatment, water distribution, or chemical process); the Digital Twin is a state-space process model coupled to an Extended Kalman Filter that predicts the next sensor measurement and emits a residual whenever the observation deviates from the physics-consistent prediction. The detection layer combines this Digital-Twin residual signal with a Long Short-Term Memory (LSTM) autoencoder, an attention-based transformer, and an Isolation Forest, fused through a calibrated weighted score that is gated by the residual, so that purely data-driven anomalies that do not violate physics are downweighted and stealthy attacks that violate physics are escalated. Evaluated on three benchmark datasets (Secure Water Treatment testbed [SWaT], Water Distribution [WADI], and Tennessee Eastman) comprising 56 attack scenarios, the framework achieves 97.6% precision, 96.2% recall, an F1-score of 96.9%, and sub-50 ms inference latency. This corresponds to a 3.2 percentage-point F1-score improvement over the strongest baseline (transformer at 93.7%) and a roughly 50% reduction in residual error. Interpretability is supported through attention visualization and Digital-Twin residual analysis, enabling operators to validate detection outcomes. With native Message Queuing Telemetry Transport (MQTT) and Open Platform Communications Unified Architecture (OPC UA) integration, Byzantine fault-tolerant consensus for distributed deployments, and formal verification of safety properties, the framework supports deployment-oriented protection for critical infrastructure aligned with International Electrotechnical Commission (IEC) 62443-4-2 requirements.
Salt stress causes concurrent Na+ and Cl- toxicity that severely inhibits plant growth and reduces yield. While Na+ homeostasis has been extensively studied, the molecular mechanisms underlying Cl- homeostasis under salt stress remain poorly characterized in plants. To prevent excessive Cl- accumulation in shoots, root-to-shoot Cl- translocation is tightly regulated. Here, we show that the zinc finger protein ZAT7 directly binds the promoter of the Cl- transporter gene NRT1/PTR FAMILY 2.4 (NPF2.4) and represses its expression, thereby inhibiting NPF2.4-mediated Cl- upward transport. Through yeast two-hybrid screening and functional validation, we identified three upstream signaling regulators of ZAT7: DUF295 ORGANELLAR A 9 (DOA9), deubiquitinase OVARIAN TUMOR 11 (OTU11), and cystatin CYSTATIN 13 (CYS13). Under normal growth conditions, OTU11 stabilizes DOA9 to promote ZAT7 degradation, maintaining NPF2.4 expression and basal Cl- supply to shoots. Upon Cl- stress, rapidly induced CYS13 inactivates the OTU11-DOA9 module, leading to ZAT7 accumulation, NPF2.4 suppression, and reduced excessive Cl- translocation to shoots. This study reveals that plants fine-tune the abundance of the core chloride-resistance regulator ZAT7 via CYS13 under salt stress to precisely control root-to-shoot Cl- delivery.
Limonoids are a structurally diverse class of highly oxidized triterpenoids produced by Meliaceae and Rutaceae species. They exhibit remarkable biological activities but are difficult to produce due to their complex biosynthesis. Plants evolve ecologically adaptive specialized metabolites through the amplification and functional differentiation of gene families, yet our understanding of the molecular mechanisms remains very limited. A chromosome-level genome assembly of Melia toosendan (219.5 Mb) showed no recent whole-genome duplication, and we annotated 14 key enzymes for triterpenoid biosynthesis, the oxidosqualene cyclases (OSCs). Transient expression of these OSCs in Nicotiana benthamiana produced 11 distinct triterpene skeletons, including eupha-7,24-dien-3β-ol, the product of MtOSC10. Phylogenetic analysis and dN/dS ratio assessments suggest that this OSC lineage may have undergone neofunctionalization events, correlating with signals of strong positive selection. Ancestral sequence reconstruction traced the divergence of an ancestral β-amyrin synthase into two evolutionary lineages: one lineage retains the conserved activity of tirucalla-7,24-dien-3β-ol synthase, which produces the canonical limonoid precursor, and the other diversifies into novel eupha-7,24-dien-3β-ol synthases that produce a hypothetical alternative precursor. Strikingly, four key amino acid residue substitutions (W258L, T413S, M730Y, L735H) in tirucalla-7,24-dien-3β-ol synthase were sufficient to switch MtOSC1 product specificity from tirucalla-7,24-dien-3β-ol to eupha-7,24-dien-3β-ol. Our findings uncover a potential alternative pathway for limonoid biosynthesis and reveal the molecular basis of triterpene skeleton diversification in Meliaceae. More broadly, they illustrate how neofunctionalization of OSCs under positive selection drives metabolic innovation across plant lineages. The finding also provides a foundation for synthetic biology approaches to engineer plant-derived insecticides.