By re-engineering the internal genetic pathways of grasses, the team successfully created “sentinel plants” that produced a distinct purple pigment when they encountered specific chemical signals. These changes were detected via remote sensing technology, potentially ushering in a new era of automated field monitoring for the global agricultural sector.
The study marks a significant milestone because it focuses on monocots. While model plants like Arabidopsis have long been the focus of synthetic biology, grasses like maize, rice and wheat have been notoriously more difficult to engineer.
The research team successfully adapted a ligand-inducible sensor to activate the production of anthocyanin — a natural pigment — in Setaria viridis, a model grass species.
Developing the biological toolkit for monocots
The agricultural industry relies heavily on monocot grasses for the majority of the world’s grain production. However, developing synthetic genetic tools for these plants has traditionally lagged behind other species.
This is largely due to differences in promoter structures and nucleotide compositions that make direct adaptation of existing tools challenging. The researchers sought to bridge this gap by creating a system that could predictably control gene expression in response to external stimuli.
To achieve this, the team identified two specific transcription factors in Setaria, known as SvR1 and SvC1. When these factors are expressed together, they trigger the plant’s natural anthocyanin pathway, turning the green tissue purple. The researchers engineered a single genetic transcript that could produce both proteins simultaneously, ensuring the system was efficient and functional.
“This work demonstrates the use of inducible expression systems in monocots to manipulate endogenous pigmentation production for remote detection,” the researchers stated. “Applying inducible anthocyanin production coupled with sensitive detection algorithms could enable crop plants to report on the status of field contamination or detect undesirable chemicals impacting agriculture.”
Overcoming the barriers of chemical uptake
One of the primary hurdles in creating responsive plant sensors is ensuring the ‘trigger’ chemical can actually enter the plant tissue. The researchers tested several ligands, including the steroid hormone dexamethasone (Dex), to see which would most effectively turn on the purple signal. They found that standard chemicals often struggled to penetrate the waxy outer layer and rigid cell walls of the grass leaves.
To solve this, the team explored different application methods and chemical variants. They discovered that triamcinolone acetonide (TA) was a far more potent inducer than standard dexamethasone.
When applied via ultrasonic nebulisation — a fine chemical fog — the TA ligand successfully bypassed physical barriers to reach the internal genetic circuit. This resulted in persistent pigment accumulation that lasted for over a month, even spreading to newly developed tissues like the panicles and flag leaves.
Remote detection through hyperspectral imaging
For these sentinel plants to be useful in a commercial setting, farmers and land managers must be able to detect the colour changes without walking through every acre of a field. The research team integrated their biological work with advanced imaging technology to ensure the signals could be picked up from a distance. They used hyperspectral imaging, which captures a much broader spectrum of light than the human eye or standard cameras can see.
Initially, the team used the Anthocyanin Reflective Index (ARI), a standard vegetation index, to measure pigment levels. While the ARI could distinguish between a fully purple plant and a green one, it struggled to identify the subtle, non-uniform patches of purple caused by chemical induction in the field. This prompted the researchers to employ a more sophisticated machine learning approach.
The team utilised a Multiple Instance Adaptive Cosine Estimator (MI-ACE), a trained algorithm designed for sub-pixel target detection. Unlike simple spectral indices that only look at a few wavelengths, MI-ACE leverages the entire hyperspectral signature to find specific patterns. This allowed the system to detect the engineered purple signal with much higher accuracy and a lower rate of false alarms.
Commercial implications for precision agriculture
The ability to turn crops into environmental reporters offers significant opportunities for the agri-tech sector. Sentinel plants could be used to detect the presence of specific herbicides, industrial pollutants, or even early signs of pathogen stress before visible damage occurs. By linking these biological signals to drone or satellite imagery, companies can implement high-precision management strategies that save time and reduce chemical usage.
The researchers believe this ‘phytosensor’ technology can be expanded to detect a wide array of molecules. By using recently developed receptor circuits based on natural plant proteins, the sensitivity of these sensors could be tuned to specific commercial needs. This would allow for the development of bespoke crop varieties that monitor everything, from soil health to the presence of specific explosives or pollutants in high-security zones.
Future directions and field deployment
While the current study focused on indoor near-remote imaging, the transition to the field is the next logical step. The MI-ACE approach is highly adaptable and can be integrated into unmanned aerial vehicle (UAV) platforms or aircraft-mounted sensors. This would enable the monitoring of entire landscapes in real-time.
However, the team acknowledged that the environment played a role in how these sensors performed. Factors like natural plant stress can also trigger anthocyanin production, which might create ‘noise’ in the data. The researchers suggested that the specific spatial patterns of the engineered induction — such as the way the pigment follows the leaf vasculature — can be used to train AI models to distinguish between a synthetic signal and a natural stress response.
The commercialisation of such systems is already beginning to emerge in the biotech space. By combining the latest in synthetic biology with sophisticated computer vision, the agricultural industry is moving toward a future where the plants themselves serve as the primary data collection tool for the field.
The researchers concluded: “Such circuits could be combined with induced pigmentation to develop plant-based sensors that monitor chemical exposure, thereby enhancing both plant and human health in agricultural settings.”
Source: Plant Biotechnology Journal
“Remote Sensing of Endogenous Pigmentation by Inducible Synthetic Circuits in Grasses”
https://doi.org/10.1111/pbi.70480
Authors: Lee Dong-Yeon, et al.
