New research shows that exposure to low levels of the popular herbicide atrazine feminizes male northern leopard frogs, Rana pipiens, both in the laboratory and in the field and “raises concerns about the effects of atrazine on amphibians in general,” according to Tyrone Hayes and his colleagues at the University of California at Berkeley’s Laboratory for Integrative Studies in Amphibian Biology. The study is being published at a crucial time because the U.S. EPA is under a court deadline to publish its preliminary reassessment of the risk posed by atrazine in January.

Last week, Hayes’s research was published on Environmental Health Perspectives’ Web site, and a brief communication about the research is being published in Nature today. The paired studies are the first peer-reviewed research to show effects on both wild and laboratory-raised specimens of R. pipiens at levels of atrazine found in the environment, down to 0.1 parts-per-billion (ppb)—which is 30 times lower than the current U.S. drinking water standard, Hayes says. The Berkeley researchers found atrazine contamination at all of the eight sites where they collected leopard frogs, including sites chosen as controls because of low reported atrazine usage.

To date, most of the published studies that show endocrine disruption in frogs have been laboratory studies focused on the African clawed frog, Xenopus laevis, which is native to sub-Saharan Africa. Last year, Hayes presented research linking exposure to 0.1 ppb of atrazine in the lab to gonadal alterations in male X. laevis frogs (Environ. Sci. Technol. 2002, 36, 55A-56A). Other researchers have thus far been unable to duplicate Hayes’s findings, although their attempts to repeat his laboratory work were complicated by the fact that they were unaware of the actual experimental conditions Hayes used when they started their efforts, says Keith Solomon, director of the Centre for Toxicology at Canada’s Guelph University. A report on this research is due out in December.

The northern leopard frog was once the most widespread frog in North America, but it has been disappearing from or decreasing in many of its previous habitats, according to the provincial government of Alberta, Canada, one of the places the species is “red-listed”. Hayes says he picked R. pipiens, together with a third frog on which he has yet to publish data, the Pacific tree frog, with the goal of amassing data on three very different species of frogs. Such data will help indicate whether all amphibians could be subject to endocrine-disrupting effects from exposure to atrazine, he says.

The U.S. EPA was sued by the Natural Resources Defense Council for failing to reassess the risk posed by a number of older pesticides, including atrazine, that were originally registered prior to 1984 as part of a lawsuit. The agency expects to issue its final risk assessment by October 2003, says Dave Deegan, an EPA spokesperson. Atrazine is manufactured by a number of companies including Syngenta, a Swiss firm, and it is the most popular herbicide in the United States and perhaps the world, according to EPA.

Among the most controversial findings in both Hayes’s current research with R. pipiens and his earlier X. laevis work is that lower concentrations of atrazine, on the order of 0.1 ppb, appear to be more likely to induce male frogs to develop feminine characteristics than higher doses of 25 ppb. Both studies imply that there is a parabolic, or inverted-U-shaped, dose–response relationship for this hermaphroditism, Hayes writes in the Environmental Health Perspectives paper.

“This is a topic that a lot of ecotoxicologists are interested in because we know that organisms have detoxification systems but they seem to only be turned on at a threshold level of contamination, so below that, the hypothesis is that the chemical can be having some sort of effect which is not seen at higher concentrations because of induction of some metabolic protection,” says Stanly Dodson, a freshwater ecologist at the University of Wisconsin–Madison.

However, “it’s difficult to build a dose–response relationship for two concentrations—you’re talking about a 250-times difference between the concentrations,” objects Jim Carr, an associate professor of biology and endocrinology at Texas Tech University, who is one of the scientists trying to repeat Hayes’s X. laevis findings. “I’d really like to see more data points,” Solomon adds, noting that some endocrine-disrupting chemicals have been reported to show such an unusual dose–response curve, but a simpler explanation needs to be ruled out.

Hayes agrees that he has not yet published enough data to prove this parabolic dose–response relationship, but he claims to have more unpublished data to support his contention. He also stresses that he was surprised by the finding. “We didn’t set out to figure out if there was a dose–response…. We were doing a screen to determine if multiple populations [of different species] showed the effect. So we weren’t trying to go through and figure out threshold effects. We’re doing that now—we’ve already looked at five populations [of R. pipiens], looking at everything up from 0.1 ppb all the way up to parts per million,” he says.

Solomon points out that the concentrations of atrazine reported in Hayes’s field studies don’t actually tell much about the concentrations of the chemical to which the animals were exposed, a fact that Hayes acknowledges in his paper. The weekly atrazine concentration data that Solomon has been collecting in South Africa in conjunction with Louis du Preez, an associate professor of zoology in the School of Environmental Science and Development at South Africa’s Potchefstroom University for Christian Higher Education, show how wildly levels can fluctuate.

“Depending upon rainfall, we have numbers going from 1 or 2 to about 11 ppb,” says Solomon, adding that the U. S. Geological Survey data for atrazine levels in the United States show similar fluctuations. “We have no idea what [the frogs Hayes examined] were exposed to during their metamorphosis,” which is a key time for sexual development, Solomon adds.

Hayes agrees that it is impossible to correlate the number of hermaphrodites with his atrazine measurements. “It’s a one-time measure, and all it shows is where you have hermaphrodites, you have atrazine—it doesn’t tell you anything about the level of exposure,” he says, noting that he has only been able to make limited atrazine measurements because of the expense involved. This lack of correlation between the measured levels of atrazine and the percentage of frogs with gonadal abnormalities explains why two of Hayes’s sites that were observed to contain 0.2 ppb of atrazine had dramatically different populations, he says. At one site, less than 10% of the frogs examined by the researchers had abnormalities, but more than 90% of the frogs at the second site were hermaphroditic.

“In studies that I am undertaking in South Africa, I looked at frogs (Xenopus laevis, Afrana (Rana) angolensis, and Afrana fuscigula) from corn production areas where atrazine is being used extensively. I have not observed the intersex cases that Hayes et al. [are] referring to. There are a number of factors that can and do affect differentiation of gonads in amphibians and reptiles,” says du Preez.

For example, Carr says that wild intersex Rana frogs were reported for decades before atrazine’s use began in the 1950s. Hayes’s paper takes pains to explain how the sex changes his team observed are different than those described previously. “They’re completely unrelated phenomenon… . In other Rana [species], there’s been a natural intersexuality described that’s presumably normal. That anatomy is a mixture of testes and ovaries. What we’re showing is that animals with testes have eggs in their testes—they don’t have ovarian tissue.”

Carr remains unconvinced. “There [are] a lot of reasons why you might see something like [oocytes in testes],” he says. Hermaphroditism can be caused by environmental factors other than atrazine, such as temperature fluctuations and the concentrations of nutrients like calcium and magnesium in the water, Solomon adds.

The debate over whether or not atrazine affects frogs—or other amphibians—is unlikely to end soon, as Hayes questions how closely the research intended to repeat his results has followed his experiments. To make matters even more interesting, he and his fellow researchers are beginning to evaluate the effects of some of the mixtures that they have found in natural settings, including combinations of 10 different herbicides, insecticides, and fungicides. “It’s going to blow this current thing out of the water: thresholds change, the effects are enhanced in ways you would not predict,” Hayes says.