|
Getting your Trinity Audio player ready...
|
For as long as farmers have grown crops, spotting trouble has meant walking the rows and looking. A new generation of engineered plants is trying to change that by making the crop itself raise the alarm. Using genetic engineering, companies are building plants that produce a light signal, invisible to the human eye, when they come under stress, so that drones, satellites or ground sensors can read it and tell a farmer what is wrong, often weeks before leaves show damage. The idea, which is being described in coverage under the banner of gene editing, is moving from laboratory papers and patents into commercial fields, though on a small scale so far.
The best-known example is InnerPlant, a seed technology company based in Davis, California, founded in 2018. It engineers plants to produce optical fluorescence signals when specific stress pathways are activated, including pathogen infection, insect pressure, water stress or nutrient deficiency, according to a company profile. Its method, as described in regulatory filings summarised by Whiteford Research, is to identify stress-responsive genetic pathways through genomic analysis and then use the plant’s own regulatory elements to drive a fluorescent protein that can be detected remotely. In plain terms, the plant’s own immune response flips a switch, and the switch turns on a glow.
The timing is the selling point. Sean Yokomizo, the company’s vice president of communications, told Michigan Ag Today that because the signal is tied to the plant’s immune response, it begins within 24 hours of infection or stress, which is four to six weeks before it would be detectable in the field. Another company executive, Shely Aronov, told SynBioBeta that signals come up within two days of initial stress, whereas it takes three weeks until external symptoms appear. These are the company’s claims, and the reports reviewed do not include independent field trials comparing the system with traditional scouting. Even so, an early warning of that kind would give a farmer time to act before a fungal infection spreads and costs yield.
Importantly, the plants do not visibly glow. Farm Progress noted that a crop with the technology is not going to glow in a way a person can see; the fluorescence falls in a part of the spectrum that can only be picked up by spectrometers or multispectral imagery. That is why the system depends on cameras mounted on drones, aircraft and satellites, or on ground detectors. InnerPlant has demonstrated hyperspectral satellite detection using Germany’s EnMAP satellite and an instrument aboard the International Space Station, Payload reported. A satellite pass can, in the company’s words, deliver a signal saying that a field was infected by a fungus within the last 48 to 72 hours, according to the Soybean Research and Information Network.
Using satellites at scale has one catch. To use satellite imagery as a detection tool, entire large fields would need to be planted with the engineered soybeans, the same network noted, which is a big commitment. The commercial product sidesteps that with sentinel plots. InnerPlant’s first product is CropVoice, a service powered by InnerSoy, a soybean engineered to signal fungal infection, launched for the 2025 growing season in selected parts of Illinois, according to Whiteford Research. The approach is to plant small plots of the engineered soybeans, monitor them, and use the signals to advise nearby farmers growing ordinary crops. If a detector sees an active mould outbreak in one plot, the company sends a notice to farmers growing nearby, Payload reported, since diseases tend to follow geography. Alerts reach farmers by text message or through John Deere’s Operations Center, with recommendations on spraying, and the service is sold through agricultural retailers.
The adoption numbers are modest but growing. Yokomizo told Payload that the service grew from around 50 farmers in its first season to between 160 and 180 customers in the next, though the report’s dates are garbled. The company says it provides actionable data, not raw images, to keep the process simple for growers. Its stated aim is to cut broad, calendar-based chemical spraying by targeting treatment where and when it is needed, an environmental and economic argument that is common in precision agriculture. The Agtech Industry Examiner repeats the company’s case that targeted treatment can cut overuse of pesticides and save money, but that remains a claim to be tested on farms over several seasons.
A word on terminology is needed, because “gene editing” is being used loosely. SynBioBeta reported that InnerPlant’s soybeans are transgenic, meaning they contain fluorescent genes that do not occur naturally in plants. That is a different technique from the targeted edits made with tools such as CRISPR, which change a plant’s existing DNA without necessarily adding foreign genes, as a review of CRISPR for abiotic stress management explains. Regulators treat the two differently in many countries. In the United States, InnerPlant’s fungal-sensing soybeans were approved by the Department of Agriculture in just nine months under the SECURE rule, because fluorescent traits had already been approved, according to the same report, and the company completed the Food and Drug Administration’s new protein consultation process for its fluorescent protein. Rules in other countries vary, and the reports reviewed do not describe approvals outside the US.
The science behind sensing plants is older than the company. A research paper on potato described phytosensors as genetically engineered plant sensors that fuse synthetic promoters to reporter genes to report specific stressors, and noted a practical problem: if the signal from a weak promoter falls below detectable limits, the sensor may not work as intended. Scientists have tried boosting signals with a synthetic transcription factor system, and patents describe networks of sentinel plants that communicate stress visually, thermally or chemically to a farmer, including an example in which a red fluorescent protein reports water stress. These documents show the idea has been explored for years, and that making the signal strong, specific and reliable under real field conditions is the hard part.
There are limits to what has been shown. The system so far is mainly aimed at fungal stress in soybeans, with insect detection described as a future plan and expansion to corn and cotton envisaged. Satellite detection depends on clouds, timing and spectral resolution. And a signal only helps if farmers trust it and have an effective response, such as a fungicide or a changed practice, which is not always available or affordable. For smallholder farmers in places such as Nigeria, the technology as described is built around US-style commercial farming, satellite services and retail networks, so its direct relevance would depend on regulation, infrastructure and cost, none of which the reports address.
The bigger picture is a shift in how farms are monitored. Instead of adding more sensors to a field, the plants themselves become the sensors, and the data flows to a software platform. If it works at scale, it could reduce guesswork and chemical use, and it could change the way seed companies sell traits, from yield to information. If it fails to deliver, it will join a long list of agricultural technologies that looked promising in the lab. The company describes its work on its InnerPlant site, and Farm Progress has an early account of the approach in its report on crops as field sensors.
Readers who follow global technology, agriculture and business stories can find more coverage at BusinessTech Nigeria. For now, the glowing soybean is a small but vivid example of engineers trying to give plants a voice, and the test will be whether farmers find that voice worth paying for.
