A Texas A&M researcher spent 30 years making edible cottonseed. Will anyone ever grow it?

From Harvest Public Media:

Growing cotton can be a tough way to make a living in 2025.

Global demand has waned for several years, thanks in part to the rise of synthetic fibers. At the same time, high prices for supplies like fuel and fertilizer cut into farmers’ bottom line.

Researchers have long looked for a way to get more revenue out of the crop.

“Right now most of [cotton farmers’] income comes from selling the fiber,” said Keerti Rathore, a professor of soil and crop sciences at Texas A&M University who specializes in genetic modification. “But if you make the seed valuable in terms of being able to utilize that for food, and also feed for other animals, maybe they’ll get more value for their seed.”

Professor Keerti Rathore has spent the past 30 years working to produce cottonseeds with less toxin. If ground into flour, the protein-packed seeds could provide food for millions of people. Lacy Roberts / Texas A&M University

For every pound of cotton fiber produced, there are 1.5 pounds of seed. But the seed’s uses are limited by a toxin called gossypol, which makes it inedible for most animals, although cattle are an exception.

“The cotton industry has always been incredibly excited about the potential of getting rid of gossypol and what that will do for the value [of cottonseed],” said Tom Wedegaertner, a veteran of the cotton industry.

Wedegaertner, retired three years ago from Cotton Inc., a grower-supported research and marketing group which funded research by Texas A&M’s Rathore.

Gossypol appears as little black dots on the cotton plant. Rathore’s goal was to make those dots disappear – a target he worked toward for 30 years.

“Most other scientists would have moved on to something else,” Rathore said. “I guess it’s probably a little stubbornness on my part.”

A lot of protein

Cottonseeds are about 23% protein. Globally, farmers produce roughly 45 million tons of them annually.

“If you could utilize that protein directly for human nutrition, you could meet the basic protein requirements of 500 million people. That’s a lot of protein,” Rathore said.

He envisions the seeds will be used more for livestock than for humans. Wedegaertner, from Cotton Inc., said they show particular promise for poultry and aquaculture.

“The critters, especially the shrimp and the black sea bass, all those guys just seem to greatly prefer [cottonseed] to a pellet that had soybean protein,” Wedegaertner said.

But cotton plants need the gossypol toxin to help them resist pests and diseases. Rathore’s challenge is to lower gossypol levels in the seed so it can be eaten, without disrupting the toxin in other parts of the plant – an obstacle that previous efforts to make the plant useful couldn’t quite clear.

In the late 1950s a scientist named Scott McMichael published research on Hopi cotton – so named for the Native Americans who cultivated it. It’s also called glandless cotton, because it doesn’t have the glands that produce gossypol, making it edible.

Researchers thought they had the key to turn cottonseed into a major food product. They worked for decades to do so.

“We had edible nuts and flavored salad toppings and cookies and crackers. Oh gosh, we made all kinds of stuff out of it,” Wedegaertner said.

Food scientists at Texas A&M University in the 1980s created TAMUNUTS: roasted and salted edible cottonseeds. The glandless Hopi cotton inspired optimism up and down the plant’s supply chain.

“The ginners [were] also quite excited because that’s the more value we have in cottonseed, the less they have to charge for ginning,” said Wedegaertner.

But none of these products ever made it to the grocery store. Without the gossypol, pests chewed it up in the field. It didn’t produce good enough fiber, either, so Hopi cotton never took off.

Research into low-gossypol cotton mostly lay dormant until Keerti Rathore came to Texas A&M in 1995.

Manipulating the genome

Rathore had never seen cotton growing in the field before arriving in Texas. His new colleagues said that the crop would be a good candidate for the cutting-edge gene modification in which he was interested.

To lower gossypol levels in just the seed, Rathore landed on a technique called gene silencing, which mostly mutes the function of a specific gene.

The same year Rathore arrived in Texas, scientists at Purdue University happened to publish the genetic sequence for an enzyme used to create gossypol in the cottonseed. Rathore then used a technology called RNA interference, or RNAi, to silence that gene, and only that gene.

“There was a lot of excitement when he started that,” said Wedegaertner.

The gene-silencing technology is contained in an RNAi “cassette,” which Rathore would splice into cotton cells. Those cells grow in petri dishes until they’re developed enough to plant in a greenhouse – a process that can take years.

The average gossypol level in a cottonseed is about 10,000 parts per million. Rathore’s goal was to lower that to 450 parts per million, which is a safe level for human consumption, according to the Food and Drug Administration.

Rathore experimented with how he introduced the cassette into the cotton plant, as well as with different varieties of cotton.

“Depending on where [the cassette] goes, the level of silencing will be different,” Rathore said. “So this is the reason why we have to produce, you know, 50 to 100 plants, and so then from those you choose which is the best one.”

If a certain line showed promise, Rathore would grow several more generations in a greenhouse to ensure the low-gossypol trait was stable. Then, he’d test it in field trials. Farmers in Texas, North Carolina and Mississippi received low-gossypol seeds made with Rathore’s technique.

“They looked at everything, all the agronomic characteristics to make sure that it was growing just as well as the normal cotton plant,” Rathore said.

The tests didn’t always yield perfect results. Around 2018, a USDA research scientist named Mike Dowd measured the levels in seeds Rathore was testing.

“I think there were 20 to 25 samples from individual plants. And every one of them came out ten times higher than what Keerti says they are,” said Dowd, who’s now retired.

Different varieties of cotton naturally have different levels of gossypol. The weather can also affect how much of the toxin is in the plant. But food products – even for livestock feed – have to be consistent, Dowd says.

“Lord help you if you say this stuff is 100 to 150 [parts per million] and you give it to a food company and they come out and decide that it’s 1,000. That kills it; it’s done,” said Dowd.

Rathore said there’s “no doubt” that the technology could be improved. But eventually, he produced a viable line of low-gossypol cotton, called TAM66274.

Trying to sell seeds

Rathore, with help from Cotton Inc., submitted documents to the FDA and the Department of Agriculture in 2017 showing that TAM66274 had low enough gossypol levels for humans and livestock to eat.

Both agencies okayed the seeds for consumption. The next step was to find a company that would market them.

“Our hope was that one or all the companies that are in this cottonseed business here in the U.S. will pick up the technology,” Rathore said. “And we tried; we knocked at each and every door.”

Rathore and Wedegaertner haven’t yet found a partner to sell the seeds. And it’s not because they haven’t made the pitch. Years of meeting with companies like Bayer and BASF yielded nothing.

“There was a lot of interest,” Wedegaertner said. “But I think once they really crunched the numbers … they backed away from it.”

Large parts of the globe won’t take any genetically modified foods, such as Rathore’s seeds. In countries that would accept them, the seeds would undergo a long, expensive regulatory review.

“So it was hard for [seed companies] to justify all the millions of dollars they would have to invest,” Wedegaertner said.

There’s no additional funding for future research, however. If low-gossypol cotton has a future, it may be outside the U.S.

Finding fertile ground

After getting told thanks but no thanks over and over by seed companies, Rathore is now trying a new strategy to get the low-gossypol cotton out in the world.

Cotton Inc. and Texas A&M University, which both own the rights to the technology, have agreed to share seeds with developing countries free of charge. The idea is that if it takes off, the companies that said ‘no’ earlier might give Rathore’s seeds a second look.

“We’ve very optimistic that if it can be demonstrated in these countries, then the U.S. planting seed companies will have an opportunity to look at that in a real-world environment,” said Kater Hake, a retired Cotton Inc. researcher who worked with Rathore on the project.

Uzbekistan is the first country to test out Rathore’s cottonseed. The country’s agriculture minister, Ibrokhim Abdurakhmonov, went to A&M for graduate school and worked in Rathore’s lab.

“Keerti’s technology, if applied correctly – that will be revolutionary for food security,” Abdurakhmonov said.

Uzbekistan is a major cotton producer and consumer. The country’s biggest soccer club, Pakhtakor FC, is nicknamed “the cotton growers.” Even though the country’s population is about one-tenth of the United States, Uzbeks use more cottonseed oil annually. Producing that oil with low-gossypol seeds would be more efficient.

Abdurakhmonov thinks the seeds could also boost the country’s poultry business. Right now, that industry largely relies on imported soybeans for chicken feed. Domestic cottonseeds could help mitigate that.

He also believes the seeds could serve as a nutritious human food down the road.

“Cottonseeds have a very high-quality protein,” he said. “So it’s a matter of time that people get used to it, you know, and eat it as sunflower seeds.”

This story was produced in partnership with Harvest Public Media, a collaboration of public media newsrooms in the Midwest and Great Plains. It reports on food systems, agriculture and rural issues.