New GM Crops Could Make Superweeds Even Stronger
Brandon Keim
Wired News, May 1 2012 
http://m.wired.com/wiredscience/2012/05/new-superweed-evolution/all/1

Herbicide-resistant superweeds threaten to overgrow U.S. fields, so agriculture companies have genetically engineered a new generation of plants to withstand heavy doses of multiple, extra-toxic weed-killing chemicals.

It’s a more intensive version of the same approach that made the resistant superweeds such a problem and some scientists think it will fuel the evolution of the worst superweeds yet.

These weeds may go a step further than merely being able to survive one or two or three specific weedkillers. The intense chemical pressure could cause them to evolve resistance that would apply to entire classes of chemicals.

"The kind of resistance we'll select for with these kinds of crops will be different from what we've seen in the past," said agroecologist Bruce Maxwell of Montana State University. "They'll select a kind of resistance that's more metabolism-based, and likely resistant to everything."

Next-generation biotech crops erupted into controversy with the U.S. Department of Agriculture’s ongoing review of Enlist, a Dow-manufactured corn variety endowed with genes that let it tolerate high doses of both glyphosate, an industry-standard herbicide better known as Roundup, and a decades-old herbicide called 2,4-D.

Back in the mid-1990s, when so-called Roundup Ready seed strains first allowed farmers to spray the herbicide directly onto fields without fear of damaging crops, 2,4-D and other older chemicals were already falling from favor. They were more toxic and less effective than glyphosate, and farmers gladly replaced them with a single all-purpose treatment.

Roundup Ready varieties now account for 90 percent of U.S. soybeans and 70 percent of corn and cotton, and the pipeline for new herbicidal chemicals is mostly empty. But reliance on a single chemical came at a price: Though industry scientists predicted that weeds wouldn’t become resistant to glyphosate, more than a dozen species have done exactly that.

These superweeds now infest 60 million acres of U.S. farmland, a fast-growing number that foreshadows a time when agriculture’s front-line weedkiller is largely useless. Enlist, which Dow estimates will save $4 billion in superweed-related farm losses by 2020, represents the industry’s main response to the problem: Bringing back old chemicals in new ways.

Of 20 genetically engineered crops under federal regulatory consideration, 13 are designed to resist multiple herbicides. They suggest a future in which more farmland is treated with more herbicides in ever-higher doses, and have been criticized by activists and researchers worried about possible chemical dangers to human and environmental health.

Largely overshadowed in the health furor, however, is the issue of new superweeds: If glyphosate-drenched Roundup Ready fields were evolutionary crucibles that favored the emergence of new, glyphosate-resistant weed strains that threaten multi-billion-dollar damage, what might new herbicide regimes create?

“Resistance happens, particularly when the selection pressure is largely from one or two tactics,” said weed ecologist David Mortensen of Penn State University. "Plants are wired to protect themselves from troublesome compounds in some interesting ways."

In January, Mortensen and Maxwell co-authored "Navigating a Critical Juncture for Sustainable Weed Management," a BioScience paper on superweed control that described two routes for evolving resistance to multiple herbicides.

The first, best-understood route is fairly straightforward: After genetic mutations help a weed's enzymes break down a specific herbicide or prevent the chemical from entering its cells, the plant breeds with another plant that's resistant to another herbicide. Its offspring inherit both defenses.

Some 108 strains in 38 weed species already resist multiple herbicides. Though troublesome, however, this type of stepwise, "site-of-action" enhancement can be countered at least for a while with chemicals to which the weeds haven’t yet adapted. But another type of adaptation is both possible, and far harder to handle, say Mortensen and Maxwell.

Plants can undergo changes in how they detoxify themselves or metabolize foreign materials, becoming more efficient at excreting toxins or sealing them inside cellular waste-containment structures called vacuoles. The resulting adaptations provide broad-spectrum defenses against many chemicals, not just those to which a few site-of-action mutations apply.

Improvements to these systems don’t occur easily, requiring many mutations and often coming at the expense of growth rates and general health. But Mortensen and Maxwell say that, paradoxically, increased simultaneous use of multiple herbicides will make them more likely.

In a field treated with a single herbicide, for example, weeds with a single site-of-action-based resistance will outcompete slower-growing weeds with improved metabolic resistance. But if a field is treated with multiple herbicides, those site-of-action weeds will be exterminated, leaving the metabolic mutants to compete only with each other. Extrapolate these dynamics across vast areas, year after year, and the implications are dire.

"This kind of resistance isn't very widespread. It usually has a fitness cost associated with it and doesn’t get to a high frequency in populations," Maxwell said. "But if we select very hard for that kind of mechanism, which I feel we are with this new approach, we'll find that fitness increases quite rapidly. It will become more and more frequent."

In Australia, researchers have identified strains of Lolium, a common weed also known as ryegrass, with detoxification adaptations that confer resistance to seven different modes of herbicidal action. These adaptations developed naturally rather than as consequence of herbicide use, but hint at what may be possible.

While this type of resistance hasn’t yet been connected to herbicide pressures, Maxwell thinks it may be undetected because the initial fitness costs leave weeds looking sickly. People looking for resistance naturally focus on healthy-looking strains.

“The mechanism has been ignored. When people go to sample these things, they’re often not scientists. They’re field people who want to know why plants didn’t die. They get the healthiest ones,” Maxwell said. “We’ve been missing, in my opinion, the other kinds of resistance that aren’t as convincing. These mutants are out there. They’re becoming more frequent. Pretty soon, if we start sampling correctly, we’ll find them more frequently.”

Among other weed ecologists and herbicide resistance experts, Maxwell and Mortensen’s predictions provoke a range of responses.

“It is  unlikely that the stacked herbicide resistance traits and associated herbicide use would push weed evolution towards a Lolium-type metabolic resistance,” said Stephen Powles, director of the Australian Herbicide Resistance Initiative.

Powles, widely regarded as one of the world’s leading experts on herbicide resistance, thinks the mutations necessary are too many and too complicated to survive the herbicide gauntlet and meet in one place.  But Pat Tranel of the University of Illinois, another leading herbicide resistance expert, said “it definitely seems plausible. This idea of broad resistance patterns has already been documented.”

“It’s really hard to predict what sort of cross-resistance you’re going to get if you use those crops wide-scale and use them every year, like people did with glyphosate-resistant crops, but you’re much more likely to get metabolic resistance than other types,” said USDA plant physiologist Stephen Duke. “It makes simple evolutionary sense that would happen.”

Duke cautioned that it’s difficult to say how long it will take for metabolic resistance to evolve. He also said the resulting weeds would still have some vulnerabilities, a point echoed by Mark Peterson, a Dow AgroSciences weed scientist and lead developer of the Enlist system.

“Just because a plant develops metabolic resistance doesn’t mean it’s automatically going to be resistant to all herbicides. There are still going to be differences in terms of the plant’s ability to deal” with different chemicals, Peterson said.

Peterson defended the approach used in Enlist, saying weed evolution is guided by how herbicides are used, regardless of whether crops are conventional or herbicide-resistant. “There are ways to utilize all the tools in the toolbox, whether it’s different herbicides or herbicides in combination with other cultural practices,” that will make it difficult for weeds to evolve resistance, Peterson said.

On this point, he and Enlist’s critics agree. Industrial agriculture reformers recommend what’s known as integrated weed management, a suite of tactics varied planting cycles, judicious herbicide use, locale-specific seed choices intended to control weeds with minimal environmental costs. Because weed control techniques would vary between years and seasons, resistances would evolve only slowly, if at all.

“As agronomists and farmers, we’ve forgotten the basic tenets of integrated pest control and management,” said James Gray, executive director of the agriculture industry-supported 2,4-D Task Force. According to Gray, the industry has learned the lessons of glyphosate over-reliance, and Dow will make integrated weed management a part of its Enlist system.

Historical and anticipated patterns in herbicide use on U.S. soybeans. Assumes new herbicide-resistant crops adopted at same rates as glyphosate-resistant crops. Image: Mortensen et al./BioScience

“There will be every effort to remind practitioners to figure out the best mechanisms on a field-by-field basis and not just rely on moving from one herbicide to another herbicide,” he said. “It’s important that we manage the appropriate use of these compounds.”

For the moment, however, non-industry researchers seem skeptical of this professed change in corporate heart.

Maxwell and Mortensen note that agriculture industry research is heavily biased towards herbicide-based solutions, with a predictable influence on academic and government researchers who often rely on industry support. The researchers’ survey of weed scientists found two working on herbicides for every one studying integrated approaches.

“Frankly, they don’t get money from pushing integrated strategies,” said Tranel. “Monsanto or Syngenta doesn’t make money from a farmer growing a cover crop. Our current paradigm is so heavily entrenched in herbicides. That’s the tool we use, that’s most cost-effective right now. It’s like anything else: Until the problem is there for you to deal with, you’re not going to switch.”

Pressure doesn’t only come from herbicide makers, either, but is intrinsic to the modern agricultural system: Farmers are pressured by large corporate buyers to grow as cheaply and efficiently as possible, and to be wary of non-herbicidal solutions that would raise short-term costs. That makes research on integrated solutions all the more important, said Mortensen.

The USDA’s period for public comment ended Apr. 27, with more than 365,000 citizens opposing Enlist’s approval. An official decision is expected later this year.

“This is not a solution to the resistance problem. It’s a step in the creation of a resistance problem of a different order of magnitude than what we’ve had,” said Margaret Mellon, an agriculture policy expert at the non-profit Union of Concerned Scientists.

“We need to decide whether we’re going to resolve weed control issues by using more and more herbicides, or by trying more sophisticated, systems-based approaches,” Mellon said. “People often say that we’re at a crossroads. We really are.”