Diverse sources for endocrine disruption in the wild

ABSTRACT

We are increasingly realizing that diverse human activities are associated with endocrine disruption in wildlife populations. Most field research on endocrine disruption tends to narrowly target particular endocrine disrupting chemicals (EDCs) and contaminant sources as the dominant culprit of observed endocrine disruption. Recently, we and our colleagues assessed patterns of endocrine disruption in metamorphosing frogs in ponds along a forest-suburban land use gradient. In conjunction, we assayed for a diversity of putative of EDCs from an array of possible sources. We found that endocrine disruption in suburban frog populations was associated with a cocktail of different EDCs including classical estrogens, phytoestrogens, and metalloestrogens. These results indicate that endocrine disruption in suburban frogs is likely to be the product of multiple chemicals and sources. Our findings have implications for other systems where endocrine disruption is attributed to a limited set of contaminants or sources without necessarily exploring the contributions of other EDC pathways. Here, we discuss our findings in the context of other field endocrine disruption research and encourage future studies to assess the broader EDC landscape which may be impacting wildlife populations.

KEYWORDS:

• Estrogen
• frog
• land use gradient
• metalloestrogen
• phytoestrogen
• suburban

Commentary

During the last 2 decades, endocrine disruption in wildlife has evolved from a troubling hypothesis with relatively little empirical support to a well-developed arena of study. Controlled laboratory exposure experiments have been joined by numerous field studies, particularly on fish and amphibians, revealing that endocrine disruption is a widespread phenomenon among wild populations inhabiting human-altered habitats. Studies of fish living in rivers subject to inputs from waste water treatment facilities (WWTF) uniformly show that the frequency of abnormal sexual development diminishes with distance from outflow.^1-8 Among studies of pond dwelling frogs, frequencies of sexual abnormalities in agricultural environments are consistently elevated relative to frogs from reference sites.^9-13 While contaminants were not usually characterized in many of the studies just referenced, the diversity of study locations implies at least some variety of contaminant types. In spite of whatever heterogeneity in contaminants, observed pathologies can show remarkable consistency across geography and taxonomy. In particular, testicular oocytes have been studied or noted in animals collected from impacted sites discussed in nine of the 13 studies listed above.^{1,3,4,7-10,12,13}

While most work on freshwater species has focused on riverine fish and frogs from agricultural settings, our lab has concentrated on the endocrine disruption in amphibians from suburban environments. This focus emerged from a study in which the testicular oocyte frequencies in male green frogs (Rana clamitans) from suburban landscapes were much higher than those observed in frogs from agricultural contexts.¹⁴ A follow up study of 28 suburban ponds confirmed that on the order of 1 in 8 adult male green frogs shows evidence of intersex and that intersex has been found in virtually all suburban ponds evaluated.¹⁵ The presence of testicular oocytes in suburban frogs raises important questions regarding the chemical composition of these ponds as well as the potential sources of endocrine disrupting chemicals (EDCs). While wastewater effluent and agricultural pesticides are well-studied putative sources of EDCs, suburban neighborhoods are complex, heterogeneous landscapes, making it challenging to target specific potential EDCs.

To develop a more comprehensive understanding of the potential causes of endocrine disruption in suburban environments, we carried out a study combining an array of chemical analyses with evaluations of sex ratios of wild green frog metamorphs along a forest-suburban land use gradient. Our findings show that there is a wide range of potential sources for endocrine disruptors in suburban landscapes (Fig. 1).¹⁶ While we assayed water samples for evidence of well-studied EDCs including pharmaceuticals such as 17α-ethynyl estradiol (EE2), a common contraceptive, or pesticides such as atrazine, we did not detect these chemicals in any suburban pond. Instead, the EDCs we detected included classical steroidal estrogens (e.g., estrone) as well as phytoestrogens (e.g.,, daidzein) and inorganic trace elements that may act as estrogens, so called “metalloestrogens."¹⁶ While phytoestrogens have been found in surface waters previously,^17-19 they have not been studied in conjunction with evaluations of endocrine disruption in wildlife. While many trace elements have been studied in the laboratory where they do exhibit estrogenic properties,²⁰ and a variety of studies show that human land uses are associated with increased loads of trace elements,^21-23 metalloestrogens are rarely studied as putative EDCs. Collectively, our chemical findings implicate several potential EDC sources including septic- or sanitary sewer-derived wastewater, plants inhabiting landscaped yards, and runoff from roads or buildings (Fig. 1). This picture of endocrine disruption contrasts with most prior studies which have tended to focus on a single source or a single chemical compound.

Figure 1. Conceptual diagram illustrating differences in chemical diversity and loading between forested ponds (left) and suburban ponds (right). Arrow colors represent different classes of endocrine disrupting chemicals (EDCs) while arrow width represents relative differences in the average concentrations of each chemical class between forested and suburban ponds. Note that phytoestrogens were not detected in forested ponds. Human land use leads to both a higher diversity of EDCs present in a waterbody as well as higher overall concentrations of EDCs.

Biologically, there is little reason to believe that endocrine disruption will be the result of just a single contaminant. Even so, most work on endocrine disruption in the wild has a priori targeted particular chemicals and sources without considering other potential sources. As one example, most research on endocrine disruption in amphibians has been carried out in agricultural environments and in virtually every case, the presumed agent has been the pesticide atrazine.^9-11,13 Leaving aside continued uncertainty about the effects of atrazine on sexual development in amphibians,^13,24 it is remarkable that not a single study has considered the potential role of other estrogenic agents, notably phytoestrogens. Crop plants are a defining feature of agricultural landscapes. Both soy (Glycine max) and alfalfa (Medicago sativa) are common crops also known to produce estrogenic EDCs^25,26 that can be found in agricultural waters.^17,18,27 Similarly, fields of corn (Zea mays) or wheat (Triticum spp.) contaminated with the fungus Fusarium can produce estrogenic mycotoxin EDCs that are known to contaminate agricultural surface waters.¹⁸ Recent experimental evidence showed that exposure to root exudates of red clover (Trifolium pratense) can cause sex ratio biases in metamorphosing froglets,²⁸ yielding effects that mimic those seen in response to pesticide exposure. Red clover is another species common in agricultural environments as well as within lawns. While phytoestrogens and mycotoxins only represent one set of possible EDCs in agricultural settings, a more comprehensive understanding of agricultural endocrine disruption would assay for these compounds in addition to atrazine and other contaminants to better understand the possible diversity of chemicals contributing to sexual abnormalities.

Our recent work indicates not only that the diversity of contaminants and sources may be greater than is generally appreciated but also that certain EDCs may have multiple possible sources.¹⁶ This message is complemented by prior work indicating that attributing the source of certain chemicals can be challenging. A survey of 11 lakes surrounded by varying degrees of agricultural, urban, and residential land uses found that endocrine disruption in fish was prevalent.²⁹ Chemical analyses revealed that the lakes were contaminated with steroidal hormones, alkylphenols, and bisphenol A. All of these chemicals are typically associated with WWTF discharge,⁴ yet none of the study lakes received direct WWTF effluent.²⁹ Recent work within streams located in heterogeneous landscapes showed, unsurprisingly, that chemicals like the pesticides atrazine and metachlor were predominantly associated with agricultural land use while the pharmaceuticals carbamazepine and sulfamethoxazole as well as the insect repellant DEET were predominantly associated with urban/residential land use.¹⁹ However, chemicals like the pharmaceutical acetaminophen, the stimulant caffeine, and the fungicide iprodione were common throughout agricultural, golf course, and urban/residential land uses indicating that chemicals often considered to be tracers of agricultural practices (iprodione) or urban wastewater (caffeine and acetaminophen) may actually have a diversity of sources, some of which may be unknown.¹⁹ Taken together this body of work shows that a given chemical can have multiple sources and that using individual chemicals as tracers of particular sources, as is commonly done, may have greater limitations than previously understood.

Our studies in the suburbs provide evidence that endocrine disruption in frogs is likely the result of multiple chemical agents and anthropogenic sources. These sources may include runoff from buildings and roads, effluent from septic systems and sanitary sewers, and even in vivo production by the plants growing in our lawns.¹⁶ While suburban endocrine disruption appears to be a complex phenomenon, we have little reason to expect that suburban systems are unique in this respect. As we continue to deepen our understanding of endocrine disruption, we suspect that we will learn that organisms are responding to a multitude of chemicals from a diversity of sources. As a corollary, in many systems we may be unable to point to a single source or contaminant acting as a “smoking gun."

Endocrine disruption in wildlife is now well established empirically. We know that it can be frequent in some populations, particularly those impacted by human activities. Research in suburban ponds,¹⁶ and other systems,²⁹ makes it clear that we cannot discount the role of multiple chemical agents and sources when studying how the landscape contributes to endocrine disruption in wildlife. An important next step will be to more fully characterize the EDC landscape in different environments harboring wildlife showing symptoms of endocrine disruption. Studies adopting approaches that will allow the characterization of multiple sources and types of agents will be critical in developing a more complete understanding of the causes of endocrine disruption as well as the best ways to mitigate its effects.

Disclosure of potential conflicts of interests

No potential conflicts of interest were disclosed.

Acknowledgments

Discussions with colleagues including L. Barber, G. Giller, and A. Vajda contributed to our thinking about sources of endocrine disruptors. Fig. 1 was designed using freely-available vector images from the Integration and Application Network (IAN) at the University of Maryland Center for Environmental Science (UMCES).