A research team has engineered grasses to turn purple when they detect specific chemicals, creating living biosensors that could help farmers spot problems early and protect global grain supplies.
In a glimpse of agriculture’s future, scientists have turned humble grasses into living warning systems that change color when they detect tiny amounts of chemicals in the environment.
A collaborative team from the Donald Danforth Plant Science Center, the University of Florida and the University of Iowa has engineered grass plants to produce a vivid purple pigment when exposed to specific chemical cues. Paired with advanced imaging tools, these plants can act as biosensors, signaling low levels of chemical exposure, pollution or other stressors long before damage is obvious.
The research, published in the Plant Biotechnology Journal, focuses on Setaria viridis, a small grass species closely related to major grain crops like corn and sorghum. In side-by-side images, wild-type plants remain green, while engineered plants turn a striking purple, thanks to a natural pigment called anthocyanin.
The idea is simple but powerful: use color as a built-in alarm system.
The research team began with a question that could reshape how farmers manage their fields: “What if plants could alert farmers to adverse conditions or unwanted chemicals?” Today, growers often rely on lab tests, sensors or visual inspection to detect problems such as pesticide drift, soil contamination or nutrient imbalances. Those methods can be slow, expensive or too late to prevent yield losses.
By contrast, plants that visibly respond to their environment could serve as early-warning “sentinels” across entire fields.
To build that capability, principal investigators Dmitri Nusinow and Malia Gehan from the Danforth Plant Science Center led an effort to adapt a synthetic genetic circuit that taps into the plant’s own anthocyanin pathway. Anthocyanins are natural pigments that give many fruits and flowers their red, purple or blue hues. In this system, the pathway is switched on only when the plant encounters a chosen chemical signal.
The team identified two key transcription factors — proteins that control gene activity — that can be expressed together from a single genetic transcript to reliably trigger anthocyanin production. That design makes the system more compact and easier to move into different grass species, including staple crops.
They then showed that the engineered circuit works in both isolated plant cells, called protoplasts, and in whole plants. In some cases, the pigment production is always on; in others, it is ligand-inducible, meaning it only activates in the presence of a specific chemical.
Color change is only half the story. To turn these plants into practical biosensors, the researchers also developed hyperspectral imaging and analytical methods that can detect subtle shifts in pigmentation from a distance, without harming the plants. Hyperspectral imaging captures information across many wavelengths of light, far beyond what the human eye can see, allowing computers to pick up early or faint color changes.
Together, the genetic tools and imaging techniques create a system for precise, remote sensing of chemical exposure in grasses. In a real-world setting, that could mean drones, tractors or satellites scanning fields to spot patches of purple that signal contamination, chemical drift or emerging stress.
“Grain crops are at the heart of global food security,” Nusinow said in a news release. “Having plants act as sentinels in the field could increase food security and improve the sustainability of agriculture.”
Plant-based biosensors are an emerging area of synthetic biology, where researchers design new biological functions by rewiring genes and pathways. Until now, most plant biosensor tools have been developed in dicot model species such as Arabidopsis thaliana. Grasses, or monocots, have lagged behind despite being the backbone of global grain production.
By demonstrating a robust, inducible pigment system in a C4 model grass, the team has helped close that gap and opened the door to similar systems in corn and other cereals. In the future, different circuits could be tuned to respond to different chemicals, allowing plants to “report” on a range of environmental conditions that affect crop performance and human health.
The potential applications are wide-ranging: detecting industrial pollutants that seep into farmland, spotting herbicide drift before it harms neighboring crops, or monitoring for stressors linked to climate change, such as heat or drought, by coupling pigment production to stress-response pathways.
The researchers also made a deliberate choice to share their tools widely.
“We wanted to build a system that other researchers could easily use. Making our constructs and imaging approaches publicly available will accelerate innovation across the community,” added Gehan.
To that end, the molecular components needed to build these biosensors in grasses, along with the methods for sensitive pigment detection, have been deposited in public repositories. That open-science approach is designed to help other labs adapt and expand the system, speeding progress in plant synthetic biology.
The project brought together expertise in plant genetics, engineering and remote sensing. Co-authors included Alina Zare, a professor of electrical and computer engineering and director of the Artificial Intelligence and Informatics Research Institute at the University of Florida, and Susan Meerdink, an assistant professor in the School for Earth, Environment, and Sustainability at the University of Iowa.
While the current study is a proof of concept in a model grass, the researchers see it as a steppingstone toward real-world applications in major crops. Future work will likely focus on tailoring the circuits to respond to specific agrichemicals or pollutants, improving sensitivity and integrating the system with field-ready imaging platforms.
If successful, tomorrow’s grain fields may not just feed the world — they may also help protect it, turning purple to warn us when something is wrong.