This is the second in a series about biotechnology. Click here for the first story.
During the early days of the Gulf of Mexico oil spill, Ponisseril Somasundaran, a chemical engineer at Columbia University, received an email from a colleague who needed help. Somasundaran is one of the world's foremost experts in surfactants, the essential ingredient in dispersants. The government was debating how to break up the oil. Could he come to Louisiana?
Somasundaran got on a plane. It's what you do when Lisa Jackson comes calling.
Before he flew south at the U.S. EPA administrator's request, though, Somasundaran had a call to make. For a couple of years, he had been working with a biotech startup near Boston, exploring ways to brew surfactants with genetically engineered microbes. There was a lot of noise about dispersant toxicity in the Gulf. All signs pointed to their surfactants being more benign. They had been targeting body washes, but the chemical could disperse crude oil, too.
Somasundaran reached his research partner, Kevin Jarrell, the CEO of Modular Genetics, during the Friday evening commute. Somasundaran was heading to Tulane University in New Orleans, where he serves on an advisory board with Jackson. Time was short. Should he mention the surfactant? It seemed their moment.
"We need to talk about a greener way of doing this," Somasundaran said.
Surfactants are typically derived from oil, or trees known to bring about rainforest destruction. They, instead, could use farm waste from soybean fields. Jarrell dashed together a slide show.
As it did for many, the Gulf crisis had turned into an opportunity.
"Important people wanted to hear these ideas, which we weren't expecting," Jarrell said.
There was no chance for their surfactant, early in development, to be used during the spill. But the concept was so well received that, by late August, the National Science Foundation had fast-tracked grants to Somasundaran, along with scientists at Louisiana State University and Iowa State University, to brew and test their green dispersants against Corexit 9500, the chemical widely used during the spill. Modular Genetics would lead the effort.
The team's study is not complete. Toxicity trials on young fish have only started at LSU. The Iowa scientists have brewed low amounts of the surfactants in their test rigs. Somasundaran is defining how the chemicals "dance" with oil. And genetic tinkering continues outside Boston.
Regardless of the group's results -- publication is expected later this year -- their work provides a small-scale example of how chemicals, developed through what many call synthetic biology, can begin replacing their oil-derived alternatives in the economy. Across the country, in private and public labs, for various chemicals and at varying sophistication, scientists are mirroring their work. It is a revolution bubbling just below the horizon.
Developing dispersants created without oil, especially if they hit a sweet spot between persistence and biodegradability, should be a priority, said Liz Kujawinski, a chemist at the Woods Hole Oceanographic Institution who is unaffiliated with the study. Earlier this year, Kujawinski published a study tracing the path taken during the BP spill by microscopic amounts of the surfactant used in Corexit (Greenwire, Jan. 27).
"Green chemistry is critically important," Kujawinski said. "Moving away from dependence on petroleum is all to be desired." But caution is needed shifting from petrochemicals, she added.
"You're fixing one problem," she said. "Are you certain you're not creating another one?"
There is no question that biosurfactants can compete, and often outdo, their oil-derived cousins, at least when it comes to performance, said Inge van Bogaert, a bioengineer at Belgium's Ghent University who has also developed microbe-based surfactants.
"The major issue will always be price," van Bogaert said. "Biosurfactants are still more expensive. This is no problem in specific applications like pharmaceuticals and cosmetics, or when the customer is prepared to pay more for environmentally friendly detergents."
Jarrell believes he can compete with oil products. But the dispersant work is about more than cost, he said. It is a high-profile opportunity to show that next-generation biotechnology can identify a need in the economy and respond, just like the petrochemical industry, where so many products of everyday life now have their origins.
"We want to show that we can use synthetic biology to meet this sort of need," Jarrell said. "To say, 'Here is a different formulation. It's totally green and it's safe and it's biodegradable.'"
'Now we have a molecule everybody loves'
To a large degree, understanding the team's challenges means understanding surfactants.
Surfactants are mongrel compounds. They resemble elongated keys, with a bulbous head that favors water and a skinny tail that loves plunging into oil and fats. When not made out of petroleum, their tails are stewed out of palm or coconut oil, hardly standard-bearers for sustainable chemistry, given their tendency to replace rainforest. Many microbes also create surfactants, but they tend not to dissolve well in water. For dispersants, that's a nonstarter.
Jarrell has long specialized in one of these bugs, a variety of Bacillus subtilis. While at Boston University, he designed tools that automated engineering of bacterial DNA, prompting Modular Genetics' creation. Knowing his microbe produced surfactant, he pitched the chemical to companies making shampoo and body wash.
The companies were unimpressed. Interesting, they said, but not water soluble enough. Try again.
Modular went back to basics. Taking a hard look at their surfactant's structure and its genetic basis, Jarrell's team guessed that it could starkly alter the chemical through simple subtraction. Its water-loving head was made of a chained loop of amino acids, the building blocks of proteins. Eliminating the DNA that provided the blueprint for these acids, except for the first few letters, might result in a soluble surfactant.
"We said, 'Well, gee, [could] we just use our synthetic biology ability to go in and site-specifically delete 25,000 base pairs of the Bacillus chromosome?'" Jarrell said. "Which is what we did. ... And now we have a molecule everybody loves."
They called their creation FA-Glu. It closely resembled a surfactant already on the market, derived from vegetable oils, that was used in contact lens solution and soaps. However, according to Somasundaran's testing, FA-Glu was 10 times more effective at reducing surface tension, the oil's ability to resist water and stay cohesive.
Surfactants are, at their heart, all about taking out surface tension. Add enough to a mix of naturally standoffish oil and water and the molecules, with their bipolar pedigree, begin pulling droplets out from oil's body, encircling the globules to maximize their exposure to the two liquids. The oil droplets come to resemble taupe, spherical pushpins, with the surfactants, heads sticking out and tails buried deep, playing the pins.
There are many uses for a chemical that can play this trick. In fact, surfactants are widely used in advanced oil drilling, freeing the crude from its geological constraints to boost production. Somasundaran had focused primarily on these industrial applications; only later did he begin to explore, with Jarrell, green variations. He had seen many surfactants deployed during his time and not all of them were harmless.
"Many surfactants are oil-based, and they're not really benign," he said.
One of the more promising developments out of the duo's FA-Glu work, which predated the oil spill, was when they fed the bioengineered bug solely soybean hulls, fibrous casings not easily digested by man or microbe. Typically, such a food source would require additional enzymes for the bugs to feed, but to their surprise the Bacillus happily grew on the hulls, producing FA-Glu, albeit at limited concentrations.
The potential of rerouting soybean hulls to surfactant production led Jarrell to a team of fermentation experts at Iowa State University, including Charles Glatz and Buddhi Lamsal. Snug in soybean country, Glatz and Lamsal have access to tons of soybean hulls, and massive steel tanks, up to 1,000 liters in volume, used to test microbial fermentation.
Much of the dispersants' "green" validation rests on the question of whether these hulls can be used to feed microbes at commercial volumes. Currently, the Iowa scientists have grown three variations of Modular's microbe in 25- and 50-liter tanks, with larger sizes to follow. The secreted surfactant gets to work immediately, a head of foam rising out of the foul-smelling, brown melange of water, hulls and byproducts.
"It looks a little bit like dirty apple cider," Lamsal said.
So far, production rates have not been chart topping. Depending on the strain, they are getting up to 3 to 4 grams of surfactant per liter. According to van Bogaert, the Belgian scientist, rates more like 50 grams per liter will be need to compete with oil-derived surfactants. Further genetic work could improve their yields, Lamsal added.
"Ideally, we would like much more productivity," he said.
From Iowa, the surfactants are sent back to Somasundaran at Columbia, where he puts them through their scientific paces, judging how few are needed to reach a threshold where the chemicals begin pulling droplets out -- their "dance" along the water-oil boundary, as he put it.
Other tests require a bit less finesse.
"You can also beat the hell out of it and see how long it will last," Somasundaran said.
The dispersants used in the Gulf of Mexico did not solely contain small, water-soluble surfactants. There were also larger surfactants, better able to break apart some of the heavier molecules in the crude mix. Solvents accompany these surfactants to smooth their way into the water; one solvent, 2-butoxyethanol, used early in the spill, has been known to carry toxicity concerns.
The team's dispersants will require a surfactant mix that can mimic the effectiveness of Corexit, a mission that became easier when EPA published Corexit's full ingredients last year, Jarrell said. However, all these surfactants, including the FA-Glu derivative being brewed in Iowa, would have sustainable origins, he added.
While Somasundaran is testing the dispersants at Columbia, Andy Nyman, an ecology professor at LSU, has begun toxicity trials of the chemical on tiny fish and, soon, worms.
Prior to the spill, Nyman conducted several influential toxicity studies with Corexit 9500. His research made him one of the many scientists plucked into the spotlight last year, leading to a torrent of companies trying to win his endorsement for this or that oil treatment. Among that crowd, Jarrell had reached out with questions, not requests.
Eventually, Nyman agreed to test the team's dispersant, if they found the money.
"I was very skeptical that this would go anywhere," he said.
Earlier this month, Nyman began testing the dispersant, after first establishing the toxicity of Corexit 9500. He uses week-old Gulf killifish, Atlantic residents about an eighth-of-an-inch long. The fish are placed in what look like miniature, rounded ice cube trays, each well about the depth of a pressed-in fingertip. The water is laced with dispersants running from 5,000 to 0.5 parts per million, and, after four days, the highest level that sees at least 50 percent of fish surviving provides the baseline for the next round.
"It sounds easy," Nyman said. "Just add some fish and wait for them to die."
While these toxicity trials, which follow common standards, are a blunt tool, they require deft hands and quick work, he said. Little water is held in the wells, and the fish are prolific waste producers. Early limits on the amount of available test dispersant mandated the small wells, which must be cleared out daily. The fish are moved with tools resembling eye droppers, and each day requires furious pipetting, Nyman said.
No results from the toxicity trials will be available before publication, but Nyman added that the study would be expanding past its initial goals. For the team's dispersants to be effective, they have to work in salt water, and Nyman is confident they can judge whether the salt changes the surfactant. If it does, the entire study could be undermined.
"We're going to go ahead and start scratching that surface," he said.
Nyman was irked by the toxicity exaggerations he saw during the spill, especially for Corexit 9500, the main dispersant applied by BP. (Corexit 9527, containing 2-butoxyethanol, was applied in limited amounts at the spill's beginning.) Surfactants are incredibly common chemicals, found in products people willingly rub all over their skin. Yet, when reporters came calling, searching for extreme quotes, Nyman had no comparisons to draw on, he said.
For perspective, then, Nyman has added a third chemical to the study.
Soon, his lab will have toxicity findings for Dawn soap.
'If you really believe in this field'
Research is unpredictable. There is a considerable chance that the team's work could fail to produce a suitable alternative to Corexit. Production might not work at large volumes -- often a challenge in fermentation -- or their suite of chemicals could fail when tested in the real world. The government's support could simply result in an incremental increase in knowledge, and the next oil spill could continue to be dispersed by oil-derived chemicals.
Jarrell is certain there is an appetite for FA-Glu somewhere in the $23 billion surfactant market. A patent is pending.
Modular is still collaborating with several personal-care companies on body washes or similar products, though Jarrell cannot disclose his partners. There is a lot of excess fermentation capacity out in the Midwest, empty tanks left by a bust in the ethanol market a few years back, and so much soybean, he said. Specialty chemical production can start from there.
"If you really believe in this field, and I do, right now the chemical industry makes 70,000 different products," he said. "We figure we can make about two-thirds of those from renewable materials. That's about a trillion dollar market. It's 50,000 different things."