There is a spray you can now buy in the United States that kills the Colorado potato beetle and almost nothing else. It carries no nerve agent, no Bt protein, no chitin inhibitor. Its active ingredient is a short loop of double-stranded RNA whose sequence is a near-perfect mirror of one gene inside the beetle. The beetle eats a leaf, swallows the RNA, and its own cellular machinery reads the molecule as an instruction to shred that gene’s output. The gene keeps the beetle’s cells from committing suicide on schedule. Switch it off and the beetle dissolves itself from the inside.

The product is called Calantha. Its active ingredient, ledprona, became the first sprayable RNA pesticide ever registered when the Environmental Protection Agency cleared it in January 2024. And the mechanism it runs on is the same one that won Andrew Fire and Craig Mello the 2006 Nobel Prize in Medicine and became the basis of a class of human rare-disease drugs that sell for six figures a year.

That is the collision worth watching. A trick discovered in a soil worm, turned into Nobel-grade medicine, is now being sprayed on row crops. And when you pull the US patent record, the surprise is not that it happened. It is who spent fifteen years quietly making sure they owned it when it did.

The leaderboard nobody published

Search the 9.3 million US patent grants for the ones that actually describe double-stranded RNA used to silence a gene in an insect or crop pest, and you get 95 grants. Read the assignees and the field’s whole history falls out of the list. The top three holders are the same companies that spent the 1990s building the genetically modified seed business:

The agrochemical majors hold more than half the field. Tucked near the bottom is a single grant from Beeologics, a startup Monsanto bought in 2011 to chase RNA treatments for collapsing honeybee hives. The seed giants did not stumble into this. They were assembling the position while the rest of agriculture was still arguing about Roundup.

The science that fed those filings climbed on its own quiet curve. In the scientific literature indexed by OpenAlex, papers on RNA interference for insect pest control go from a single paper in 2006, the year of the Nobel, to 115 in 2025. The chart is a near-perfect ramp, and almost none of it made the trade press until a product reached a shelf.

The mechanism is the same every time, which is the tell

The coherence test for any patent story is whether the documents share an engineering mechanism or just a phrase. Here they share a mechanism so exact you can read it off the claims.

GreenLight Biosciences’ US Patent 11,185,079, granted in 2021, claims a double-stranded RNA “at least 100 nucleotides in length and 90% to 100% complementary” to the gene encoding the Colorado potato beetle’s Inhibitor of Apoptosis protein. Match that sequence, get the beetle to eat it, and the beetle’s own RNA-interference pathway destroys the matching messenger RNA. The apoptosis brake comes off and the cells die. That is ledprona.

Bayer’s in-plant version targets a different gene in a different beetle by the identical logic. SmartStax PRO corn, commercial since 2022, carries an engineered stretch of DNA that makes the plant produce dsRNA against DvSnf7, a housekeeping gene the western corn rootworm cannot live without. According to Bayer Crop Science, field trials across 42 sites in nine states cut rootworm root injury and adult emergence by 80 to 95 percent, including in fields where the rootworm had already evolved past the Bt proteins farmers had leaned on for a decade. It was the first corn trait to kill an insect with RNA instead of a toxin.

The newest filings push the trick past insects entirely. GreenLight’s US Patent 12,077,763, granted in 2024, describes RNA “triggers” sprayed onto grapevines to silence a gene in Erysiphe necator, the fungus behind powdery mildew. Same molecule, same silencing pathway, a fungal target, no genetic modification of the plant at all. Delete the phrase “RNA interference” from all three patents and they still describe one invention pointed at three problems. That is a technology, not a buzzword.

The payoff a farmer cares about is specificity. A sequence 100 bases long that matches the potato beetle’s apoptosis gene does not match a bee’s, a ladybug’s, or a human’s. The EPA’s own review found ledprona active against its target and not against the beneficial insects tested. For an industry under regulatory and public pressure over broad-spectrum insecticides that kill pollinators along with pests, a chemistry you can aim at a single species is the most valuable property on the label.

Why the door only opened in 2024

If the biology has been understood since Fire and Mello’s 1998 paper, and the patents date to the mid-2000s, why is the first sprayable product only two years old?

Because spraying was an economics problem, not a biology problem. Building a gene into a seed costs money once. Spraying a field means manufacturing kilograms of double-stranded RNA cheaply enough to compete with a chemical that costs pennies a gram, and for two decades RNA was a reagent you bought by the milligram for a lab. The molecule was ready. The factory was not.

That is the adjacent-possible turn, and it is why a company nobody in agriculture had heard of made the first product. GreenLight Biosciences did not start as a pest-control firm. Read its patent trail in order and you watch the pivot happen: cell-free production of sugars, of nucleoside triphosphates, of ribonucleic acid, a string of filings from 2014 onward on making RNA in a vat without living cells. The company was building a cheap RNA factory. Only once it had one did the grants turn to “Control of Coleopteran insects,” then powdery mildew, then, in 2025, back to “nucleic acid therapeutics for genetic disorders.” The manufacturing breakthrough was the key. Pests and medicine were two doors it opened.

The market nearly killed the company first. GreenLight went public through a blank-check merger in 2022 at a valuation around $1.5 billion, pitching RNA for both vaccines and crops, as Fierce Biotech reported. Eighteen months later the vaccine dream had collapsed and a consortium led by the farmland investor Fall Line Capital took it private for about $45.5 million, a wipeout of roughly 97 percent. The EPA cleared Calantha a few months later. The first sprayable RNA pesticide in history shipped from a company the public markets had just left for dead.

What the incumbents do next

The majors are not sitting still, and the honeybee thread shows where this goes. In 2019 Bayer submitted a dsRNA product called BioDirect to the EPA to kill Varroa mites, the parasite chewing through commercial bee colonies, then in 2021 licensed the underlying patent to GreenLight, according to a 2026 review in the journal Horticulturae. GreenLight’s bee product, vadescana, silences a gene called calmodulin in the mite while leaving the bee unharmed. Syngenta, the second-largest patent holder on this list, has validated its own dsRNA spray against the potato beetle in the field and put commercialization on a seven-to-ten-year horizon. Science called the approach “the perfect pesticide” for its accuracy.

For an R&D director in crop protection, the strategic read is stark. The foundational US patents on silencing pest genes with RNA were granted years ago, and they belong to three companies. The piece everyone underpriced was manufacturing, and the firm that solved it traded for the price of a mid-sized office building. The next decade of this market will be decided less by who can find the right gene to switch off, which is now close to a search problem, and more by who can make double-stranded RNA at the price of a commodity chemical. The biology has been waiting in the patent record since the Bush administration. The factory is what finally arrived.

Method note

Counts come from roughly 9.3 million US utility patent grants sourced from USPTO bulk grant data, filtered by full-text search for granted patents describing double-stranded RNA or RNA interference used to control insect or crop pests, through May 2026. Assignee figures combine variant spellings and corporate successors (Monsanto under Bayer, Devgen under Syngenta, Dow AgroSciences under Corteva); a loose keyword filter on a fast-moving field both misses relevant grants and admits marginal ones, so treat the 95-grant total as a grounded floor, not a census. Literature counts are from roughly 357 million works indexed by OpenAlex. Regulatory, commercial, and licensing details are drawn from the EPA, Bayer Crop Science, Science, Fierce Biotech, and a 2026 review in Horticulturae, each named in the text. Patent claim language is quoted from the granted US patents cited by number.