Visualization and Prediction of in vivo Phosphate Dynamics via Auto-Glowing Plant Sensors

Abstract

SUMMARY Monitoring endogenous nutrient levels is crucial for maximizing crop yields and optimizing fertilizer use. Here, focusing on phosphorus, an essential nutrient for plant growth, we developed a low-cost and non-invasive biosensor to visualize and predict early stress signaling in plants. By combining plant phosphate (Pi)-deficiency-induced promoter systems with fungal self-sustained bioluminescence systems genetically engineered into tobacco plants, we created sensor plants that emitted more light when experiencing Pi deficiency. This light emission correlated with the expressions of known phosphate-responsive genes and the total phosphorus content in plants, and decreased during Pi recovery conditions, demonstrating the responsiveness and robustness of the sensor plants in reflecting endogenous phosphorus deficiency. The sensor plants responded primarily to Pi deficiency rather than nitrogen or potassium deficiencies and were sensitive to different ranges of external Pi concentrations. Additionally, when grafted onto tomato and chili pepper plants, the sensor plants responded to external phosphorus deficiency, showing promise for monitoring stress signals in different crop species. Using deep-learning-based image analysis techniques, auto-luminescent signals of sensor plants could be detected and used to predict phosphorus deficiency. This study outlines a strategy of creating a self-luminous biosensor to visualize phosphate dynamics in planta and predict nutrient deficiency for sustainable agriculture.

Plant phenotyping relevance

植物内リン欠乏状態を自己発光センサーと画像解析で可視化・予測する手法を開発し、応答性・頑健性を検証しているため、植物フェノタイピング手法が中心です。

Code and data availability

The supplied blocks describe the paper's auto-luminescent tobacco sensor plants, imaging datasets, and deep-learning pipeline, but contain no public deposit, availability statement, or authors' URL for the phenotype/image datasets, analysis code, or trained models. Solgenomics Network and Gene Ontology are cited only作为

No evidence-backed public reproduction asset is currently recorded.