||Hayes, TB, P Case, S Chui, D Chung, C Haefele, K Haston, M Lee, VP Mai, Y Marjuoa, J Parker and M Tsui 2006. Pesticide mixtures, endocrine disruption, and amphibian declines: Are we underestimating the impact? Environmental Health Perspectives, in press.
Recent studies relevant to the effects of mixtures:
More on mixtures
Studies used by the EPA and FDA to set health standards experiment with one chemical at a time. In this research publication, Hayes et al. demonstrate that this approach can lead to profound underestimates of the risks of chemical exposures.
Hayes et al. report that tadpoles suffered heavy mortality when exposed to a mixture of 9 pesticides, even though mortality rates were low when tadpoles were exposed to any one of the pesticides by itself . They discovered that most of the animals treated with the mixture developed a bacterial disease not observed in any of the other treatment groups.
This study is highly unusual for two reasons, both of which contribute to the realism and real-world relevance of the research:
These results may help explain widespread extinctions of frogs, which have appeared to be the result of fungal infections. They also raise profound questions about health standards that have been established through experiments that use only one chemical at a time.
- First, most of the pesticides in the experiments had previously only been studied at levels 10,000 times or more above the levels used in these experiments.
- Second, no study of pesticides' impacts on tadpoles has ever been published on a mixture of more than a few chemicals simultaneously.
What did they do? Hayes et al. used information about pesticide applications on fields in Nebraska to create laboratory simulations of mixtures that would be encountered by frogs living nearby. They studied the impact first of chemicals one-at-a-time, and then what happened when those same chemicals were applied in mixtures typical of what real frogs would experience. The pesticides used for lab experiments were obtained from the manufacturer or comparable sources and mixed according to the experimental design.
In the first set of experiments, Hayes et al. examined the effect of each chemical by itself at a concentration of 0.1 ppb. They then created mixtures of pesticides, one mixture with each chemical at 0.1 ppb, the other at 10 ppb.
In the experiments, animals were exposed for most of their larval life, from 2 days post hatching until their tails were completely resorbed. Each treatment was repeated three times, with 30 tadpoles per treatment. Aquaria were color-coded with treatments so that people measuring the effect of the treatment did not know what treatment a specific tadpole had received.
Two experiments were conducted separately with atrazine:
- atrazine + S-metolachlor, in a mixture with each pesticide at either 0.1 or 10 ppb;
- Bicep II Magnum, a commercial preparation of atrazine that contains 33% atrazine plus other ingredients, including S-metolachlor. This mixture was delivered in two concentrations, calculated to contain 0.1 and 10 ppb atrazine.
Hayes et al. measured several endpoints in the tadpoles after treatment to identify effects of exposure:
- tadpole mortality
- weight and size of each animal at metamorphosis
- timing and duration of metamorphosis
- histological examination of gonads and the thymus. Examination of the thymus was added once it became apparent that many treated animals were dying from infections.
After observing deaths due to infections, Hayes et al. added an additional experiment, testing the effect of the 9-pesticide mixture, each pesticide at 0.1 ppm, on plasma corticosterone levels in adult male African Clawed Frogs Xenopus laevis. They used a different species in this experiment because leopard frogs are too small at metamorphosis to provide repeated blood samples, and because Xenopus is available year-round for research. These adult males were exposed to the pesticide mixture for 27 days, after which corticosterone levels were measured in their blood.
What did they find?
Effects of pesticide exposures on mortality:
- Average mortality of tadpoles exposed to individual pesticides at 0.1 ppb was 4%, ranging from 0 to 7.8% for different pesticides.
- Tadpoles exposed to the two atrazine mixtures experienced less than 10% mortality.
- 35% of tadpoles exposed to the mixture of 9 pesticides, each at 0.1 ppb, died.
- 100% of tadpoles exposed to the mixture of 9 pesticides, each at 10 ppb, died.
Effects on timing of metamorphosis:
By itself, only propiconazole affected the timing of tadpole metamorphosis: animals exposed to propiconazole began and completed metamorphosis later than control animals. This is reflected in the graph below, which shows the number of days since hatching that animals in different treatment groups completed metamorphosis.
The atrazine mixtures (atrazine with S-metolachlor or in Bicep, the commercial preparation) did not affect timing of tadpole metamorphosis, either.
However, the mixture of 9 pesticides delayed initiation and completion of metamorphosis.
Effects on size at metamorphosis:
Atrazine, cyfluthrin and tebupirimphos reduced one measurement of frog size at metamorphosis, snout-vent length. Atrazine and tebupirimphos decreased body weight, also.
Of the mixtures, atrazine plus S-metolachlor and the 9-pesticide mixture decreased snout-vent length, while only atrazine plus S-metolachlor decreased body weight.
Effects on interaction between timing and size:
Normally, the longer a tadpole takes to complete metamorphosis, the larger it will be upon completion. This is reflected by a positive correlation between time to completion and size at completion. For the most part, this relationship was maintained when the tadpoles were exposed to single pesticides. However, in tadpoles exposed to the 9 pesticide mixture, animals that took longer to complete metamorphosis weighed less upon completion.
In previous work with leopard frogs from another source population, Hayes et al. have shown that low levels of atrazine cause some males to become hermaphroditic. In this set of experiments, the frogs used did not complete sexual maturation by the end of the experiment, so effects on gonadal differentiation could not be ascertained. Hayes et al. attribute this to differences among populations.
Of the animals exposed to the 9-pesticide mixture that survived to metamorphosis, 70% developed a suite of symptoms that included an inability to sit upright, meningitis, otitis interna, and septicemia due to the water-borne bacteria, Chryseobacterium (Flavobacterium) menigosepticum.
None of the animals in the control group, nor those exposed to individual pesticides nor any of the atrazine-mixture treated animals developed the syndrome.
However, Hayes et al. found that all treatment groups, including the controls, tested positively in assays for the bacteria. So the bacteria were universally present; they caused disease, however,only in animals exposed to the mixture of 9 pesticides.
|The frog to the upper right is newly metamorphosed frog from the control group. The lower frog (same age) was exposed to the 9-pesticide mixture. It shows symptoms typical of that treatment group.
||Because of the marked effect on the health of the animals treated with the 9-pesticide mixture, Hayes et al. dissected the thymus of animals from different treatment groups and calculated the number of thymic plaques in dissected animals. They found none in the control animals. The highest percentage was in animals treated with the 9 pesticide mixture. This difference was statistically highly significant (p<0.001).
Effect of mixture on corticosterone levels:
Corticosterone levels in Xenopus males treated with the 9-pesticide mixture (each pesticide at 0.1 ppb) were almost 4-times higher than in control Xenopus (p < 0.05). Increases in corticosterone have been shown to cause a wide array of effects in amphibians, including: retarted growth, slower development and immune system suppression, i.e., the responses seen in these experiments to the pesticide mixture.
What does it mean? This study is unique. No other published experiment has examined the impact on tadpoles of extended exposure to a mixture of pesticides at environmentally-relevant levels. Indeed, several of the pesticides used in this experiment had never been tested on amphibians, and with one exception, those that had been tested had only been examined at exposure levels 10,000 times higher than those used here, and only for the effects of acute exposure. The exception is atrazine, which has received considerable attention at low doses beginning with Hayes's prior work. Only a few published studies have examined the effects of mixtures on amphibians, but they exposed animals to much higher concentrations of pesticides (references listed in Hayes et al.)
These results indicate that low-level, widespread exposures to pesticide mixtures can impair immune system function in tadpoles, leaving them vulnerable to infection by ubiquitous bacteria. They also show that traditional approaches to assessing pesticide toxicity-- one chemical at a time-- completely miss major sources of harm.
These startling results published by Hayes et al. raise a series of questions, first about frogs, then about the ways that health standards are established.
There is widespread consensus that frogs and other amphibians are experiencing widespread extinctions, globally. Climate change, fungal infections, habitat loss, UV exposure and chemicals have been identified as potential contributors. Recent work on Central American frogs, for example, has implicated climate change as an agent, because warming creates conditions favorable to a fungal agent that kills frogs.
This paper demonstrates that environmentally-relevant exposures to mixtures of pesticides undermine the immune system of developing leopard frogs, rendering them vulnerable to bacterial infections they otherwise would have resisted. Notably, all treatment groups in this experiment tested positive for the bacteria; only those exposed to the pesticide mixture developed the syndrome of conditions.
Such an effect could compound the vulnerability of frogs to the climate-fungus interaction cited above, rendering frogs even more vulnerable to the impact of the fungus. In studies of the fungal infection, it has been suggested that this is a new fungus, or that its range is spreading into new areas. Hayes et al. offer a different interpretation: what has been spreading has been frogs' inability to fight an infectious agent that was already present.
Earlier work has indicated pesticide impacts on immune system function increase vulnerability to frog deformities, also.
Why are changes in timing of metamorphosis important? Average time to completion of metamorphosis was increased by over 2 weeks in surviving animals exposed to the 0.1 ppb pesticide mixture. This can be a life-and-death difference for frogs, because they often breed in ephemeral habitats that dry up as spring moves into summer. Longer metamorphosis also extends the period of vulnerability to predators: an adult frog, fully capable of hopping, is less vulnerable than an animal whose tail has not yet regressed completely.
The research team also noted that animals maturing later were smaller than animals maturing earlier, contrary to the normal pattern (longer time to maturation means larger animals). This adds another risk: Smaller animals face tighter limits on diet because they can't ingest larger objects.
Hence the impacts of the mixture not only lead to death from infection, it also heightened vulnerability to habitat loss (ponds drying), predation and food shortage.
According to Hayes (personal communication) the rapid death of the animals exposed to the mixture composed of pesticides at 10 ppb was not due to bacterial infection. Something else, as yet unidentified, is happening at that contamination level.
Setting health standards: None of the tadpoles exposed to one pesticide at a time came down with bacterial infections. No regulatory tests of pesticide safety have ever been carried out in a way that would reveal this effect.
Regulatory tests are conducted one chemical at a time, even though in the real world, frogs and people are exposed to hundreds of chemicals simultaneously.
This gap between real world circumstances and artificial conditions in testing laboratories is likely to have led to significant underestimates in risks of chemical exposures.