The main ingredient of wood, cellulose, is one of the most abundant organic compounds on Earth and a dream source of renewable fuel. Now, bioengineers suggest that it could feed the hungry as well. In a new study, researchers have found a way to turn cellulose into starch, the most common carbohydrate in the human diet.
Ethanol is today's most common biofuel used to power vehicles. It's typically made using sugars from crop plants such as corn and sugar cane, a system critics decry is a waste of food. Enter cellulose. Plants generate as much as 180 billion tons of the substance globally per year. Companies around the globe are racing to produce biofuels from cellulose from inedible plants, such as switchgrass and poplar trees, grown on marginal land that requires little water, fertilizer, herbicide, and pesticides, or from the vast amount of scrap from crop and wood-based industries. For instance, every ton of harvested cereals is often accompanied by 2 to 3 tons of cellulose-rich scrap, most of which goes to waste.
Now, for the first time, it appears there may be a practical way that cellulose could also feed people, says bioprocess engineer Y.-H. Percival Zhang of the Virginia Polytechnic Institute and State University in Blacksburg. He credits his line of thought to his Chinese background. "Food security has always been the number one question for nearly 5000 years of Chinese history," Zhang says. "Without enough food, crises happened and dynasties shifted." For instance, famines spurred peasant rebellions that helped lead to the collapse of the Tang Dynasty in the 9th century and the Ming Dynasty in the 17th century.
Zhang and his colleagues focused on starch, which makes up as much as 40% of people's diets. The idea of turning cellulose into starch is one rooted in the similarities between the compounds: Cellulose is composed of hundreds to thousands of molecules of the sugar glucose, and starch compounds are made of glucose as well, although the sugar is bonded together in different ways.
To make the conversion, the researchers took genes from certain species of bacteria, soil fungi, and potatoes and then genetically modified Escherichia coli, a different bacterium and a common lab model, to produce the needed enzymes. One set of these enzymes break cellulose down into smaller molecules, while the other set builds these components into starch. Then they mixed this cocktail of enzymes with cellulose in a glass vessel.
Instead of breaking cellulose down entirely to its base components of glucose and reassembling them into starch, the first set of enzymes breaks cellulose down to cellobiose, a compound made from a pair of glucose molecules. The next set of enzymes then splits the cellobiose apart into an ordinary glucose molecule and a molecule known as G-1-P, which is a glucose with a phosphate molecule attached. This G-1-P serves as the building block of chains of the starch molecule amylose. "It's a simple but nice idea," says bioengineer Frances Arnold of the California Institute of Technology in Pasadena, who did not take part in the work.
So far, the process converts up to nearly a third of the cellulose into amylose, the team reports online today in the Proceedings of the National Academy of Sciences. That compound, which is a white powder when dried, can not only be added to food but can also be made into transparent, flexible, biodegradable plastics and store hydrogen for energy applications. Zhang and his colleagues even tried some of the final product: "No taste in the beginning," Zhang says. "After chewing for a while, it tasted slightly sweet."
The rest of the cellulose is turned into glucose, which can be fermented into biofuel. "No glucose released from the cellulose was wasted," Zhang says.
Though the process works, it's expensive. Zhang estimates that, given the current price tag of the enzymes that his team used, it would cost about $1 million to turn 200 kilograms of crude cellulose into 20 kilograms of starch, about enough to feed one person's carbohydrate needs for 80 days. Still, after 5 to 10 years of further research, Zhang says companies could do the same thing for just $0.50 per person per day. "We do not see big obstacles to the commercialization of this process."
Optimistically assuming 100 billion tons of cellulose is available per year, "we will have a potential of approximately 4.5 billion tons of starch, which is nearly twofold the annual production of cereal—that is, 2.3 billion tons per year now," Zhang says. That would provide up to 30% of the food that prior studies estimate is needed to feed the world in 2050.
Still, it remains uncertain whether the approach will be economically viable, says energy economist Wallace Tyner of Purdue University in West Lafayette, Indiana, who did not participate in the study. "I am not saying it will not be—just that it is so early there are and so many uncertainties that it appears to be a very long way from a commercial process," he says. "That is usually the case when new processes are introduced."
Arnold agrees that whether the process is economically feasible overall "is the big question, but that cannot be answered in a proof-of-concept study such as this one. I think the paper demonstrates an important conversion and overall idea—it's pretty cool."
*Update, 1:25 p.m., 16 April: This article originally reported that the researchers took genes from certain species of bacteria. In addition to taking genes from bacteria, the researchers also took genes from soil fungi and potatoes.