Northern Prairie Wildlife Research Center
Pesticides are an established agricultural tool in the northern Great Plains. Grue et al. (1986) estimated that 80% - 90% of row-crop acreage is treated with herbicides. In a survey of water quality at streamflow-gaging stations throughout the corn and soybean belt of the United States, Thurman et al. (1992) detected atrazine in 98% of postplanting samples; 55% of these detections were above the maximum contaminant level (MCL) set by the U. S. Environmental Protection Agency. Spikes in concentrations of triazines during postplanting runoff events could reach an order of magnitude higher than the MCL (Thurman et al. 1992). Concentrations as high as 2.3 mg/L have been recorded (referenced in Morgan et al. ) although they rarely exceed 20 µg/L in Midwestern watersheds (Solomon et al. 1996). Results from previous work in the prairie pothole region (D. L. Larson, unpublished data) indicated sediment atrazine concentrations as high as 12.2 µg/g in prairie wetlands. Atrazine persists in the environment for relatively long periods, with a half-life of 244 d (at 25°C and pH 4; Solomon et al. 1996). Although atrazine is not the most lethal herbicide for either vertebrates or invertebrates (Solomon et al. 1996), sublethal effects are commonly observed in aquatic organisms (Dewey 1986; Kross et al. 1992; Douglas et al. 1993; Davies et al. 1994; Fairchild et al. 1994), including Xenopus embryos (Morgan et al. 1996). The combination of widespread use, relative persistence, and known sublethal effects suggests that atrazine may affect animal communities in wetlands within agricultural areas in the Great Plains.
As a group, amphibians are generally considered to be more sensitive to aquatic contaminants than other aquatic vertebrates (Boyer and Grue 1995). In particular, larval northern leopard frogs (Rana pipiens) and American toads (Bufo americanus) were found to be more sensitive to atrazine than either channel catfish (Ictalurus punctalus) or rainbow trout (Onchorhunchus mykiss) in acute toxicity tests (Howe et al. 1998).
Tiger salamanders (Ambystoma tigrinum) occur throughout the prairie pothole region of the northern Great Plains (Conant and Collins 1991). Their ubiquity and complex and variable life history make them ideal model organisms to explore potential effects of agricultural practices on prairie wetland communities. Larson (in press) has proposed that agricultural pesticides and fertilizers that enter prairie wetlands may influence aspects of salamander life history via interactions between stress hormones (e.g., corticosterone) and thyroid hormones (e.g., thyroxine) that promote differentiation associated with the onset of metamorphic climax. Developmental plasticity that has been adaptive for tiger salamanders over evolutionary time scales in variable environments typical of these wetland habitats (Whiteman 1994) could be less adaptive in the current agricultural environment. Novel stimuli such as agricultural chemicals could activate stress responses that influence speed and progress of metamorphosis, uncoupling the link between environmental variability and developmental plasticity.
Plasma corticosterone concentrations rise in response to stressors in the environment (Axelrod and Reisine 1984; Sutanto and De Kloet 1994). Likewise, thyroxine concentrations have been shown to increase in neotenic tiger salamanders under captivity stress (Norris 1978). Corticosterone accelerates metamorphosis by promoting conversion of thyroxine to the more potent triiodothyronine in larvae in which circulating thyroid hormones are already elevated (Rosenkilde 1985; Burggren and Just 1992; Denver 1996). Thyroxine not only promotes differentiation but also retards growth in tiger salamanders (Norman et al. 1987). Plasma concentrations of both thyroxine and corticosterone typically rise during metamorphic climax in tiger salamander larvae (Larras-Regard et al. 1981; Norman et al. 1987; Carr and Norris 1988). If larvae have not reached the beginning of metamorphic climax, elevated corticosterone concentrations have not been shown to influence metamorphic progress (Burggren and Just 1992). Although corticosteroids almost invariably rise in response to acute stressors, for example, short-term confinement (Wingfield et al. 1997), there is evidence from studies of fish and mammals that the response may become exhausted under chronic stress from contaminated environments (Hontela et al. 1992; Fairbrother 1994).
Amphibian larvae often have the ability to accelerate metamorphosis in response to deteriorating conditions in larval habitat (Rose and Armentrout 1976; Newman 1994). A link between the neuroendocrine stress pathway, involving both corticosterone and thyroid hormones, and changes in metamorphic progress in response to habitat desiccation has been described recently for an anuran, Scaphiopus hammondii (Denver 1997). In this study, we asked if exposure to sublethal concentrations of the herbicide atrazine results in changes in thyroxine and corticosterone concentrations in plasma of larval tiger salamanders and if these endocrine changes are associated with changes in growth rate and metamorphic progress. If tiger salamander larvae respond to atrazine contamination as they would to habitat desiccation (i.e., if both constitute stressors that promote metamorphic transformation), we would predict elevated thyroxine and corticosterone concentrations in atrazine-exposed larvae. Concurrent with elevated thyroxine, we also would predict slowed growth, but accelerated metamorphosis. If corticosterone remains elevated, it should further accelerate metamorphic progress, provided thyroid hormone levels are already high. However, if the stress response is exhausted by chronic exposure, metamorphosis may be slowed by the lack of corticosterone-mediated synergism in the conversion of thyroxine to triiodothyronine.