The search for new crop protection chemicals was once described as trying to find a needle in a haystack. But that old cliché doesn’t work anymore. These days the process is harder than that. It’s more like a scavenger hunt followed by a footrace on a track filled with trap doors and ending in a pole vault.
The big agriculture players Bayer, Syngenta, Corteva Agriscience, and BASF built their businesses by investing billions of R&D dollars in high-throughput in vitro screening, vast compound libraries, and other large-scale methods to discover new products. But compared to productivity during their heyday, these firms are no longer getting a good return on their research investments.
That’s because what the industry needs, and regulators now require, are well understood candidate molecules that are safe for nontarget species, human health, and the environment and that work by new modes of action to prevent the rise of resistant pests. Despite the big spending on R&D, it has become harder and harder to find new chemicals that meet these challenges. Successful new molecules are so rare, in fact, that the agriculture industry is looking beyond chemistry for crop innovation. A lot of attention and funding is going to firms touting all manner of hyperspectral satellite images, nitrogen-fixing microbes, digital sensors, robotic field weeders, and vertical farming gear.
Yet in the mix of hundreds of ag start-ups, there is a small cohort of risk takers aiming directly at the most difficult science endeavor in agriculture: discovering new crop protection molecules.
MoA Technology, Boragen, and Enko Chem are three firms taking a back-to-the-drawing-board approach to pesticide discovery. They say they’re not hindered by existing investments in research infrastructure. Instead, they are free to probe for novel biological targets inside weeds and other pests, experiment with new chemical building blocks, and take advantage of cutting-edge tools like DNA-encoded libraries.
But to bring new crop protection products to market, the start-ups will need to work with agriculture giants. Only entrenched firms can orchestrate and pay for the global field trials and expensive toxicological tests required to win regulatory approval. And it is too soon to tell if the big players will see working with the chemistry start-ups as a must, given the raft of flashier technologies out there for the taking.
Big agriculture firms do need new strategies to solve the problems of today’s farmers, says Decker Walker, a managing director at Boston Consulting Group. Company scientists have already plumbed the depths of known biological targets with their chemistry tools. Any new molecules will be under such regulatory scrutiny that developing them will require prodigious spending, he says. “Just like at big pharma companies, without blockbusters it is hard to justify spending on R&D.”
Meanwhile, pests are becoming resistant to existing chemistry, so growers are looking to the crop protection industry to provide new modes of action. “The big companies are in the business of innovation and are open to any type of partnership. If there’s a better mousetrap, they are interested in acquiring that technology,” Walker says.
MoA, Boragen, and Enko all prioritize finding new modes of action for killing a weed, insect, or disease organism as early as possible in the discovery process. That’s a big change from how the industry traditionally operated. For example, the weed-killing ability of glyphosate, the active ingredient in Roundup, was discovered in 1970, but researchers were still working out its precise mode of action 30 years later (Proc. Natl. Acad. Sci. U.S.A. 2001, DOI: 10.1073/pnas.061025898).
With their new approaches, the start-ups say they’ll advance a handful of contenders rather than hundreds or thousands of hits from in vitro screens. They say their strategies will pay off. A molecule that kills pests via a new mode of action is less likely to fail at later, and more expensive, stages of development.
Investors seem to agree. MoA Technology, a UK start-up, raised $8 million in its first round of funding from Oxford Sciences Innovation, a university-linked venture fund, and Parkwalk Advisors. Boragen has taken in over $15 million in investments and grants. And Enko Chem recently raised over $50 million from venture investors and the Bill & Melinda Gates Foundation.
With its very name, MoA promises to deliver new modes of action. It was founded in 2017 to use a new tool for screening herbicides: a tiny but fully formed plant. Liam Dolan, a botany professor at Oxford and the company’s founder, developed the model organism to research the genetics of plant evolution and didn’t imagine he’d later be using it to kill plants.
Genetically the plant, which Dolan doesn’t identify, is similar to Arabidopsis, which is used widely in agriculture research.
Dolan’s team makes as many as millions of the plants and is very familiar with their tidy genetic code. “We were using this neat system to answer fundamental genetic questions about all plants that grow on the land,” Dolan recalls. “And in actual fact we developed a technology perfect for discovering herbicides and also for discovering how herbicides work.”
The approach involves exposing tiny, whole plants to chemicals and assaying their response. The in vivo approach doesn’t give false positives, Dolan notes. “If it doesn’t kill a plant, we don’t even know about it.”
As a plant succumbs to a chemical, its symptoms can tell a lot about how the compound works. For example, there may be evidence it is inhibiting photosynthesis, Dolan says. Using special cameras, MoA extracts about 50 pieces of data about each plant. The data are mapped to a 3-D space, with a technology MoA calls Galaxy, that shows how symptoms are related. Some candidates will group together, as in a solar system, suggesting the same or related modes of action, while others stand apart.
“We are looking for the stars out on their own, because that suggests a new mode of action,” Dolan says.
He says Galaxy has identified some strong contenders, including natural products. The next step is to identify a compound’s specific biological target—say, a step in an enzyme cascade or a protein in the plant. Finally, with a known chemical and a known target, researchers can look for related chemicals that work even better than the first, or screen again for other chemicals that also have the novel mode of action.
At Boragen, meanwhile, researchers are building out a new universe of candidates based on their boron platform chemistry. Chief scientific officer Tony Liu cofounded the company in 2015 to translate boron’s ability to fight fungal infections in people to the agriculture realm. “Our angle is using unique attributes of boron to access new modes of action and new targets that traditional chemistry cannot access,” Liu says.
The first candidate in Boragen’s pipeline boasts activity on two different molecular sites in fungi. Likewise, a second compound can treat a common disease organism as well as block a root-damaging nematode.
Farmers need new modes of action as regulators take older fungicides off the market, says Hannah Webb, Boragen’s head of business development. Another constraint is that fungicides are often delivered on seed coatings. “Everyone is competing for space on that seed, so if you have two activities, that helps to fill a market need,” Webb says.
Across the industry, companies are devoting significant resources to new fungicides. But fungi replicate rapidly and can quickly develop resistance—a particular problem for soybean growers in Brazil, Webb says. By combining a partner company’s older chemistry with a new Boragen pesticide featuring a new mode of action, Webb says, crop protection companies can extend the useful life of their existing products.
“We will work with all the major ag companies. We are involved in collaborations in global field testing, and we see the sense of urgency to refresh their pipeline,” Webb says. “They’re pretty receptive to working with a 20-person shop.”
Boragen has learned that each company has its own culture and works at different speeds. “Some interact with us more and are more used to outside innovation,” Webb says. “We adapt how we explain ourselves and our boron chemistry—what is unique about it and what it brings to the table.”
The founders of Enko Chem are also familiar with the culture inside big agriculture firms because both spent much of their careers there. CEO Jacqueline Heard worked for the venture capital arm of Monsanto before the company was acquired by Bayer. Chief science officer Tom Meade worked on trait development at Mycogen and later was head of trait development at Dow AgroSciences, now Corteva.
“I knew how hard we worked to find valuable starting points in the crop protection discovery area,” Meade says. “And I knew the industry had a very mixed track record in target-based discovery—and even tried a lot of technologies like combinatorial chemistry.”
Enko’s answer is to look for chemicals that work against biological targets specific to a pest, such as the farmer nemesis fall armyworm. Of the potential biological targets scientists have discovered in fall armyworm, existing products work on only a few of them, Heard explains. That leaves a lot of room for exploration.
Just as at pharmaceutical firms, Meade says, Enko’s goal is to find chemicals that bind to a target. To do that, the firm is using DNA-encoded libraries developed for drug discovery.
DNA-encoded libraries are mixtures of molecules, each tagged with a string of DNA that acts like a bar code and contains information about how the molecule was made. The technique allows a library of billions of tagged compounds to be stored in a single small vessel. Traditional high-throughput screening methods are expensive and bulky, but with DNA-encoded libraries, Enko can run massive, parallel tests against its targets.
Rather than build the libraries itself, Enko is collaborating with X-Chem Pharmaceuticals, a library-builder that traditionally works in human health.
Heard says Enko’s founding in 2017 was timely. The agriculture industry has consolidated, and the now-larger firms have to stretch their R&D dollars across genetics, biologics, and digital tools as well as chemistry. With the dust settled on their mergers, companies are ready to bring in additional molecules to fill the holes in their pipelines.
“The appetite is broad in the sense of what we are looking for,” confirms Marco Busch, head of weed control research at Bayer. “For partnering, we have grand targets that we are putting out there if they have interesting new modes of action for insect, disease, or weed control.”
As evidence that Bayer is happy to work with start-ups pursuing cutting-edge science, Busch points to a joint venture the company launched last year with Arvinas, a human health biotech company with expertise in targeted protein degradation. The venture, called OerthBio, will take advantage of small molecules that act selectively to interfere with proteins in insects, weeds, and disease-causing fungi.
Busch says Bayer is reordering its research efforts to get early confirmation that a candidate chemical works by a new mode of action and is likely to pass rigorous safety testing. Using this approach, the firm is developing a new herbicide it says controls grass weeds resistant to glyphosate. Because researchers already know its mode of action, they have a big head start on work to engineer crops that tolerate the molecule, Busch says.
Bayer is interested in promising chemistry developed by start-ups, but it is also on the hunt for new research tools to use in its own labs. Those include virtual library screening, new types of assays, and models that can provide an early safety assessment.
For start-ups that wish to work with Bayer, Busch has some words of caution. “Is it a realistic dream? We often need to explain that the time to develop a molecule from initial lead to product can take 12 years and cost about $250 million,” he says, citing figures from the consulting firm Phillips McDougall. “If they expect royalties within 3 to 5 years, we have to dampen that hope.”
Still, Busch says Bayer frequently works with start-ups and universities to test their leads and provide an early read on any promise. That can lead to a deeper partnership.
“We know the market, what the product looks like, and what we have to deliver,” Busch says. “You need a lot of experiments that do not work to find the ones that do.”