According to the new study, which in appears in this week's issue of the Proceedings of the National Academy of Sciences, for every kilogram of fiber the cotton plant produces, it also provides 1.65 kilograms of seeds (22 percent of each seed consists of a high quality protein). But the gossypol's appearance in the seeds relegates them to feed for multistomached animals like cows. Previous attempts at blocking gossypol production in cotton targeted glands throughout the plant--but that allowed pests and viral visitors to overrun it, because the toxin's presence in the plant's floral tissues and foliage serve to protect it.
The Texas A&M team, led by plant biotechnologist Keerti Rathore, looked to RNAi technology to silence the gene responsible for the production of gossypol in the cottonseed. In RNAi, researchers insert a DNA construct into a genome, which when transcribed yields a double-stranded piece of RNA, a potent trigger for silencing a specific gene. This auxiliary strand of RNA gets cut into smaller pieces, which effectively cut off the production of a target protein. For the cotton plant, Rathore targeted the enzyme δ-cadinene synthase with a construct taken from a wild, upland cotton (Gossypium hirsutum) and a gene promoter that was highly specific to the seed.
In the first generation of transgenic plants, the researchers found as much as a 99 percent reduction in gossypol (from 10 micrograms per milligram to 0.1 microgram per milligram). In a test of how well this trait of low gossypol production in seeds is passed to progeny, Rathore tested toxin levels in the next generation of plants, finding an average gossypol value of 0.19 microgram per milligram. "RNAi does not knock out a gene's activity completely," Rathore explains. "But, we have been able to bring down gossypol below a level considered safe. (The United Nations Food and Agriculture Organization allows as much as 0.6 microgram per milligram in cottonseed products.) Rathore also notes that his team tested the levels of gossypol in other parts of the plant and found that they were at levels comparable with unaltered specimens. He adds that his team's method keeps the silencing effect from spreading to other tissues outside of the seed.
Vicki Vance, a biologist at the University of South Carolina, says she suspects that the reason the silencing signal was limited to the seed is probably because it can efficiently move upward in the plant (from root to stalk) but does not move well downward. "What [Rathore has] done is silence things in the seed," she says, "which is the final part of the plant." Rathore notes that his plants have not yet left the greenhouse--he has grown three generations and will cultivate a couple more before conducting field studies that will determine the effectiveness of the silencing.
Even if this technology does work, it is hard to believe people will want to eat cottonseeds. Vance says she would be game for eating the seeds, noting that their high protein contentmakes them more advantageous than commonly eaten seeds such as corn and some beans, which are missing certain amino acids, making them inferior protein sources. But how do they taste? Rathore has tried them and already has some recipes to offer. "Cottonseed kernels can be roasted and salted and eaten; they taste pretty good," he recommends. "Or, they can be crushed and mixed with wheat or maize or sorghum flour to make bread that now has high protein content."