Weeds' herbicide resistance loophole exposed

The quest for potent new herbicides with highly selective activity against plant enzyme systems may be having the paradoxical effect of accelerating the emergence of herbicide-tolerant weeds, according to an Australian weeds expert.

Dr Chris Preston, of the Cooperative Research Centre for Australian Weed Management in Adelaide, believes modern, high-throughput screening techniques may miss promising compounds that could provide more durable control of crop weeds.

And Preston says the international agrochemical regulatory environment may inadvertently be speeding the emergence of resistant weeds through increasing requirements that new herbicides exhibit narrow, well-defined biochemical activity.

Preston's concerns arise from his Weeds CRC research project to determine why some herbicides succumb to resistance more rapidly than others. "The assumption has always been that the herbicides that create resistance quicker kill more of the weeds, but I've never believed that is the only factor," Preston said.

The study sought to measure the background frequency of natural mutants tolerant to sulfonylurea herbicides like chlrosulfuron, which inhibit acetolactate synthase (ALS), an enzyme essential for amino acid biosynthesis.

The studies involved spraying native populations of annual ryegrass (Lolium rigidum) from south-eastern South Australia and adjacent south-western Victoria, and Buchan weed, Hirschfeldia incana, a dicotyledon, common in SA's Barossa Valley.

The frequency of natural mutants turns out to be surprisingly high, and consistent across species: around 1 in 17,000 for annual ryegrass, and 1 in 13,000 for Buchan weed.

At this frequency, says Preston, models predict that regular spraying with a particular herbicide will typically see resistance emerge within about four years.

"The project and other information has given us a good handle on why we get resistance and which herbicides are more likely to succumb," he says. "We've produced an information sheet for farmers, with a rule-of-thumb guide to how many times they can use a particular herbicide before they're likely to see resistance."

The remedy, says Preston, involves rotating herbicides so that the few weeds that survive exposure to one class of herbicide will be mopped up by another. Modelling suggests that rotating herbicides with different modes of action will delay the emergence of resistance.

One of the surprises from the study was that herbicide-resistance studies in the laboratory may not reliably predict what will happen in the field.

At the recent ComBio 2004 conference in Melbourne, Preston described how intense selection pressure with triazine herbicides under field conditions yielded resistant plants that, in all cases, featured a serine/glycine mutation in the D1 protein of Photosystem II, whereas in laboratory studies a number of different mutants can be found.

"The fact that many of the early examples of herbicide resistant weeds carried this single mutation has coloured the way we think about the way resistance will emerge," Preston says. That is, that resistance will be caused by single target enzyme changes.

A similar story applies to the widely used herbicide glyphosate (Roundup). Early on it was thought that glyphosate resistance would not occur because glyphosate was not metabolised by plants and it was very difficult to create viable mutants enzymes in the lab. Instead, when resistance did evolve in annual ryegrass it was caused by a reduction in herbicide movement to critical parts of the plant.

Preston says glyphosate resistance is very rare in the field, because the mutation that confers resistance to the herbicide in annual ryegrass significantly modifies plant metabolism. Preston has just completed a study showing that glyphosate-resistant annual ryegrass produces between 40 and 80 per cent fewer seeds than normal types.

Far from spawning hybrid 'superweeds', Roundup Ready canolas that hybridise with brassica relatives like wild radish (Raphanus) are likely to produce "wimpy weeds" that cannot compete with wild-type plants or canola itself.

Preston says resistance to herbicides like 2,4-D is rare despite more than 50 years of use, because they interfere with multiple activities in the plant, a very desirable property in a herbicide.

The current use of high-throughput screening techniques to identify promising new herbicidal compounds is identifying molecules that inhibit single enzymes with narrow activity. These 'single-gene' herbicides are more likely to succumb rapidly to resistance than those that target enzymes involved in multiple synthetic pathways.

Preston says some regulatory agencies are requiring companies to provide increasing amounts of documentation on the precise mode of activity of herbicides, including their target enzymes.

"In some cases, we still don't know precisely how some of the very effective herbicidal compounds introduced more than 30 years ago work," he says. They would struggle to make it through some current regulatory systems.

"Economic factors and the regulatory environment are leading to a situation where we may see fewer herbicide choices for farmers. Companies are looking for compounds that will win a major market share -- they're generally less interested in chemicals that might be only minor players."