Frontiers in Plant Science最新文献

Rubus chingii Hu. is a versatile plant valued for its medicinal and edible properties. Its fruits are rich in flavonoids, with unripe fruits exhibiting higher flavonoid content. This study identified the WRKY gene family in this plant and explored its regulatory mechanisms in flavonoid synthesis. In total, 52 members of the RcWRKY family were identified; they were unevenly distributed on six chromosomes and all located in the nucleus. Based on the phylogenetic tree, these 52 members were divided into seven subgroups, and the motifs and gene structures of different subgroups were highly consistent. Their promoters were rich in elements such as light and jasmonic acid methyl response elements. Additionally, 20 collinear gene pairs were identified in the genome, most of which underwent purifying selection. Whole-genome duplication was the main cause of expansion of the RcWRKY family, and most members showed obvious collinearity with dicotyledonous plants. Transcriptome analysis revealed that 47 RcWRKY members were differentially expressed during fruit ripening, and 11 were highly expressed at the mature green (MG) stage with high flavonoid content. Additionally, the 30 identified flavonoid synthesis-related genes were highly expressed in the MG period, with six RcWRKY members being significantly positively correlated with most flavonoid synthesis genes. Reverse transcription-quantitative polymerase chain reaction and subcellular localization confirmed that RcWRKY34 and RcWRKY37 were highly expressed in the MG period and located in the nucleus. Dual-luciferase assay showed that both RcWRKY34 and RcWRKY37 positively regulated the flavonoid synthesis gene LG07.48. Overall, this study lays a foundation for enhancing the medicinal value of palm-leaved R. chingii.

Weeds remain a major constraint to row-crop productivity, yet current deep learning approaches for UAV imagery often require extensive annotation, generalize poorly across fields, and provide limited interpretability. We investigate whether modern vision-language models (VLMs) can address these gaps in a zero-shot setting. Using drone images from soybean fields with ground-truth weed boxes, we evaluate six commercial VLMs, ChatGPT-4.1, ChatGPT-4o, Gemini Flash 2.5, Gemini Flash Lite 2.5, LLaMA-4 Scout, and LLaMA-4 Maverick under a unified prompt that elicits (i) weed presence, (ii) spatial localization, (iii) reasoning, (iv) crop growth stage, and (v) crop type. We further introduce Error-Probing Prompting (EPP), a counterfactual follow-up that forces re-analysis under the assumption that weeds are present, and we quantify self-correction with expert-rated interpretability scores (Grounding, Specificity, Plausibility, Non-Hallucination, Actionability). Across models, Gemini Flash 2.5 delivers the most consistent zero-shot performance and highest interpretability, ChatGPT-4.1 provides the strongest reasoning but lower raw detection, ChatGPT-4o offers a balanced profile, and LLaMA-4 variants lag in localization and specificity. Gemini Flash Lite 2.5 is efficient but fails EPP stress tests, revealing brittle reasoning. Visual grounding analysis and a text-to-region overlap metric show that interpretability tracks spatial correctness. Results highlight that explainability and feedback driven adaptability not scale alone best predict reliability for field deployment, and position VLMs as promising, low-annotation tools for precision weed management.

Entomopathogenic nematodes (EPNs) of the genera Heterorhabditis and Steinernema are increasingly recognized as potent biological control agents due to their ability to infect and kill diverse insect pest taxa through a symbiotic partnership with insect-pathogenic bacteria. Over the last decades, substantial progress has been made in improving EPN field performance through advances in formulation and application methods, use of biodegradable polymers and nanocarriers, and elucidation of stress tolerance mechanisms. However, despite their proven efficacy, large-scale commercialization of EPNs remains limited by high production costs, formulation instability, and environmental constraints. While numerous reviews have separately addressed EPN biology, mass production, or field application independently, a critical and integrative synthesis linking molecular mechanisms, and formulation strategies remains lacking. This review synthesizes current understanding of EPN biology with emphasis on molecular mechanisms governing host localization, invasion, and immune suppression, as well as their biotic ecological interactions within soil environments. We also discuss advances in stress tolerance mechanisms, innovations in formulation, and outline future research priorities needed to develop ecologically resilient EPN-based biocontrol products. As agriculture shifts toward more regenerative and environmentally sustainable systems, a comprehensive understanding of EPN biology, full ecological potential of EPN-bacteria partnerships holds promise not only for effective pest suppression but also for advancing fundamental understanding of host-microbe interactions and ecosystem resilience.

Soybean (Glycine max [L.] Merrill) is one of the world's most important agricultural crops, playing a strategic role in global protein and lipid production. However, its productivity is severely constrained by defoliating lepidopterans such as Anticarsia gemmatalis, Chrysodeixis includens, Helicoverpa armigera, and species of the genus Spodoptera, which cause substantial yield losses due to their intense herbivory and remarkable adaptive capacity. Conventional management strategies relying on chemical insecticides provide only partial control and are associated with negative environmental and ecological impacts. Although transgenic Bt soybeans have demonstrated efficacy against certain pest species, they exhibit limited toxicity toward Spodoptera spp. In this context, the exploitation of soybean genotypes with natural resistance represents a promising alternative within the framework of Integrated Pest Management. This review summarizes the principal chemical defense mechanisms underlying soybean resistance to lepidopterans, emphasizing the role of secondary metabolites, such as flavonoids, phenolics, tannins, and volatile organic compounds, that function as toxic, antinutritional, or repellent agents. Several genotypes, including IAC 100, PI 227687, and PI 227682, have displayed resistance to multiple caterpillar species, establishing themselves as valuable genetic resources for breeding programs. Furthermore, recent studies indicate that environmental conditions, plant developmental stage, and multitrophic interactions strongly modulate the expression of these defense traits. A comprehensive understanding of the chemical interactions within the soybean-lepidopteran system is therefore crucial for the development of more tolerant and sustainable cultivars, reducing dependency on insecticides and slowing the evolution of insect resistance. Future perspectives emphasize the integration of omics technologies, bioinformatics, and biotechnology to elucidate key metabolic pathways and accelerate the generation of resistant soybean varieties, ultimately promoting higher productivity and agricultural sustainability.

The chloroplast genome serves as a valuable tool for plant phylogenetic studies due to its conserved structure and slow evolutionary rate. Despite the economic importance of the genus Citrus, comprehensive comparative analyses of chloroplast genomes across multiple species remain limited. As one of the ancestors of Citrus, Citrus medica L. serves as a key representative in studies of the origin and evolution of the genus. In this study, we conducted a comprehensive comparative analysis of chloroplast genomes from 35 Citrus species and their close relatives, with C. medica as a focal species, to investigate structural variation, codon usage patterns, and phylogenetic relationships. All genomes exhibited a typical quadripartite structure, ranging from 159 to 161 kb with GC contents of 38.41-38.49%. While genome synteny was highly conserved, expansions and contractions at IR boundaries provided species-specific variation. SSR analysis revealed abundant mononucleotide repeats with a strong AT bias, predominantly distributed in non-coding regions. Codon usage analysis indicated a preference for A/U-ending codons, and ENC-GC3s, neutrality plot, and PR2 analyses suggested that natural selection was the main force shaping codon usage bias. Most protein-coding genes were under strong purifying selection, whereas matK and rps16 exhibited elevated Ka/Ks ratios, suggesting relaxed selective constraints or potential signals of positive selection. Phylogenomic analysis strongly supported the monophyly of Citrus and resolved intrageneric relationships, grouping species into distinct ancestral and cultivated clades. Overall, this study provides essential chloroplast genomic resources for molecular breeding, species identification, and understanding Citrus adaptation.

18-Hydroxyoleic acid and its dioic acid derivative, oleic-1,18-dioic acid, are the two most prominent aliphatic monomers in soybean root suberin. While hydroxylated fatty acids are known to be formed by cytochrome P450 monooxygenases (P450), mainly from the CYP86A and CYP86B subfamilies, the biosynthetic origin of their corresponding dioic acids in soybean remains unclear. Two root-expressed soybean P450 genes, GmCYP86A37 and GmCYP86B9 were cloned and expressed as recombinant enzymes in yeast. A third root-expressed soybean P450 gene (GmCYP86A38) was also cloned, but no recombinant protein was produced. In vitro assays demonstrated that GmCYP86A37 and GmCYP86B9 exhibited preference for the ω-hydroxylation of oleic acid (C_18:1) and lignoceric (C_24:0) acids, respectively. Surprisingly, in vitro production of oleic-1,18-dioic acid was also detected when GmCYP86A37 was supplied with oleic acid substrate. Furthermore, CRISPR/Cas9-mediated double knockout of Gmcyp86a37/38 resulted in substantial reduction of ω-hydroxylated fatty acids and dioic acids. These findings underscore the role of the CYP86A subfamily in soybean aliphatic suberin biosynthesis and provide direct evidence for GmCYP86A37 in the formation of oleic-1,18-dioic acid.

Pine wilt disease, often referred to as the "cancer of pine trees," is characterized by its rapid spread and extremely high mortality rate, posing a severe threat to forest ecosystems. Currently, most automatic identification methods for pine wilt disease based on UAV remote sensing imagery rely on a single architecture of Convolutional Neural Networks (CNNs) or Transformer, which suffer from limitations such as restricted receptive fields, insufficient global context modeling, and loss of local details. Existing fusion strategies typically adopt simple stacking or parallel designs without an effective hierarchical feature interaction mechanism, resulting in inadequate integration of semantic and detailed information, as well as high computational overhead, which hinders their deployment in edge computing environments. To address these issues, this study proposes PGCNet, a semantic segmentation model that efficiently fuses CNN and Transformer representations. The model employs CSWin Transformer as the backbone network to capture comprehensive global contextual information. A Progressive Guidance Fusion Module (PGFM) is designed to achieve effective cross-layer fusion of semantic and detailed features through a spatial-channel collaborative attention mechanism. Furthermore, a lightweight Context-Aware Residual Atrous Spatial Pyramid Pooling module (CAR-ASPP) is introduced to enhance multi-scale feature representation while significantly reducing the number of parameters and computational complexity. Experimental results demonstrate that PGCNet outperforms mainstream semantic segmentation models across multiple evaluation metrics, showing especially strong performance in scenarios with complex background interference and small-scale disease target identification. The proposed model achieves high accuracy with excellent computational efficiency, offering a practical solution for real-time monitoring and edge deployment of forestry disease detection, and exhibiting strong potential for extension to agricultural remote sensing disease identification tasks.

Wild rice species serve as a valuable genetic reservoir for rice improvement, and the DNA segments containing useful genes/alleles can be transferred to elite rice varieties through crosses. However, DNA markers, which are essential tools for genetic analysis and molecular breeding, are not yet well-established for harnessing wild rice species. To enable an efficient utilization of the CC-genome wild rice species in the genus Oryza, we developed a genome-wide polymorphic InDel marker (≥20 bp) set comprising 182 markers through positional sequence alignments between O. officinalis (CC-genome) and rice cultivars (AA-genome). These markers were evenly distributed across the 12 chromosomes, with ~2Mb marker intervals. For validation of marker polymorphism, all the markers were tested by using PCR-agarose gel analysis with 12 accessions of CC-genome species (four accessions each from O. eichingeri, O. officinalis, and O. rhizomatis), two accessions of BBCC-genome species (O. minuta), and two cultivars, Nipponbare (japonica) and IR24 (indica). Out of 182 markers, 172 markers (94.5%) successfully amplified and exhibited polymorphism between rice and the CC-genome accessions, corresponding to an average marker interval of ~2.17Mb across the rice genome. Moreover, the marker set also showed high polymorphism (84.1-92.9%) when applied to BBCC-genome species. Based on the marker validation data, five markers were selected for species identification within the three CC-genome species. In addition, the polymorphic markers successfully detected wild rice introgressions from the wide hybridization progenies. The newly developed marker set will function as valuable genomic tools for harnessing CC-genome and CC-genome containing germplasm for rice improvement.

Calcium-dependent protein kinase (CPK) gene family, which can be activated directly by Ca²⁺, plays an important role in Ca²⁺ signal transduction and stress response and is widely present in green plants. So far, the role of CPK gene family evolution for species in karst area is far from understanding. Hemiboea subcapitata (Gesneriaceae) was used to explore this issue, for its mainly distributing in the karst area of south China. Our results indicated that 32 highly conserved CPK genes identified were distributed across 14 chromosomes in H. subcapitata. Additionally, 10 gene pairs were generated by fragment replication. Subcellular localization analysis revealed that HsCPKs were mainly localized in chloroplasts and cytoplasm. This gene family experienced intron loss events, but its motif structure was highly similar. Phylogenetic analysis showed that the HsCPKs were divided into four subfamilies. Subfamilies I and II were under neutral selection, while subfamilies III and IV were under strongly positive selection. The HsCPKs showed different expressions in three vegetative organs of H. subcapitata. Meanwhile, the expression levels under calcium stress revealed an overall increasing trend for all HsCPKs examined. Cis-acting elements analysis revealed that HsCPKs contained hormone-responsive elements related to stress. The expansion and evolution of CPK gene family in H. subcapitata may be related to its adaptation to calcium-rich and stressed habitats. This study provides a valuable understanding for the roles of the CPK gene family within karst species.