Ecotoxicology of organochlorine chemicals in birds of the great lakes


Silent Spring 1 was fulfilled in the United States with passage of environmental legislation such as the Clean Water Act, the Federal Insecticide, Fungicide, and Rodenticide Act, and the Toxic Substance Control Act in the 1970s. Carson's writings, television interviews, and testimony before Congress alerted a nation and the world to the unintended effects of persistent, bioaccumulative chemicals on populations of fish, wildlife, and possibly humans. Her writings in the popular press brought attention to scientific findings that declines in populations of a variety of birds were directly linked to the widespread use of dichlorodiphenyltrichloroethane (DDT) in agriculture, public health, and horticulture. By the 1970s, DDT and other persistent organic pollutants (POPs) were being banned or phased out, and the intent of these regulatory acts became apparent in a number of locations across the United States, including the Great Lakes. Concentrations of DDT and its major product of transformation, dichlorodiphenylchloroethane (DDE), were decreasing in top predators, such as bald eagles (Haliaeetus leucocephalus), osprey (Pandion haliaetus), colonial waterbirds, and other fish-eating wildlife 2. Eggshell thinning and the associated mortality of bird embryos caused by DDE had decreased in the Great Lakes and elsewhere by the early 1980s 3.

Nevertheless, populations of certain species of birds in the Great Lakes were not recovering as quickly as expected. Hatching rates increased in a number of bird species; yet, fledging rates in certain populations remained low, and deformities of the bills, feet, head, and eyes were observed at elevated rates compared with populations from areas not associated with the Great Lakes or from more remote areas. Collectively, the deformities observed in a variety of fish-eating species of birds were described as Great Lakes embryo mortality and edema disease syndrome (GLEMEDS) 4. The Canadian Wildlife Service initiated a herring gull (Larus argentatus) monitoring program in 1973 2, partially to monitor the status and trends of concentrations or persistent residues in eggs but also to monitor biological outcomes in the birds. Meanwhile, in the United States, the U.S. Fish and Wildlife Service was also observing deformities in colonial waterbirds in certain populations they monitored, particularly in Green Bay (Lake Michigan) and Saginaw Bay (Lake Huron) (e.g., Kubiak et al. 5). Additionally, a family of avid birders routinely visited several islands to observe and enumerate the breeding of colonial waterbirds 6. In all cases, these monitoring efforts demonstrated that populations of most species of birds increased in the 1980s to the 1990s; but in certain locations, and particularly among colonial waterbirds, GLEMEDS continued to be observed.

The causal agent(s) for the embryo deformities in select populations of Great Lakes birds had not been confirmed, but the toxicity of polychlorinated biphenyls (PCBs), chlorinated dioxins, and chlorinated furans, collectively known as dioxin-like compounds (DLCs), had been hypothesized. Discovery of the aryl hydrocarbon receptor (AhR) and then characterization of responses of several classes of chemicals that activate the AhR signal-transduction pathway in vertebrates revealed two important facts about agonists of this receptor 7. First, similar signs of wasting syndrome (reduced weight gain), activation of hepatic oxidative enzymes (e.g., cytochrome P4501A [CYP1A]), immune suppression, and developmental anomalies (cardiovascular defects, cranial and pericardial edema, hemorrhage) were consistently observed in mammals, birds, and fish when exposed to DLCs 8. Although sensitivity among species varies, similar AhR-mediated effects were observed following exposure to DLCs. Second, the potency of different AhR agonists to induce signs of dioxin-like toxicity followed quantitative structure–activity relationships among the different toxicological responses, leading to the development of relative toxic potency factors for DLC congeners compared to the standard, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) 8. The potency of individual dioxin-like congeners to induce CYP1A activity through a ligand-activated pathway and transcriptional activation of the “AhR gene battery” was proportionally related to the potency of that same congener to cause other pathological signs such as reduced weight gain or immune suppression 9.

These findings led to better tools for evaluating both the toxicity of DLCs and vertebrate exposure to DLCs. For example, the ability of DLCs to induce CYP1A activity in vitro could be used to evaluate the sensitivity of a particular species or to assess the toxic potency of a mixture of DLCs. Complex mixtures of DLCs could be tested and evaluated for overall toxic potency to elicit AhR-related effects using cell bioassays and responses such as ethoxyresorufin O-deethylase induction 10. Advances in understanding the molecular signaling pathways of DLCs were critical events that allowed development of exposure-assessment tools. The subsequent marriage of these new technologies for exposure and toxicity assessment of DLCs together with intensive on-site monitoring of waterbird colonies in the Great Lakes proved to be critical in the determination of causal linkages between GLEMEDS observed in populations of Great Lakes fish-eating birds and DLCs.

Colonial waterbirds exhibited classic signs of DLC-related toxicity—including teratogenicity, decreased fledging rates, altered biochemical metabolism, and reproductive problems—in populations nesting near industrial areas of the Great Lakes, such as Saginaw Bay (Lake Huron), Green Bay (Lake Michigan), Maumee Bay (Lake Erie), and Hamilton Harbor (Lake Ontario) 2, 4, 5. Initial monitoring by the Canadian Wildlife Service focused on the herring gull, with status and trends evaluated through the measurement of contaminant concentrations in eggs collected annually from 13 designated colonies 2. In an effort to properly interpret these measurements, develop biomagnification factors, and better understand ecosystem chemical dynamics, the Canadian Wildlife Service created a toxicokinetic model for the intercompartmental movement of a series of POPs in herring gulls 11, 12. This model turned out to be important for the evaluation of colonial waterbirds in the Great Lakes and ultimately other locations around the world where fish-eating birds were contaminated with organochlorine chemicals. Indeed, the publication by Braune and Norstrom 13 was one of the most cited papers in Environmental Toxicology and Chemistry 14 and has been used in numerous modeling efforts to evaluate the exposure of birds to organochlorine chemicals.

The identification of individual compounds and a more complete understanding of their distribution in fish-eating birds was a critical step in testing the hypothesis that DLCs were causing the deformities, edema, wasting syndrome, and reproductive problems observed in certain breeding colonies of Great Lakes waterbirds. However, a single compound or class of compounds had not been identified and linked with these emerging effects, collectively referred to as GLEMEDS 4. In laboratory exposure, PCBs were known to cause GLEMEDS in birds. At about the same time, polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) had been identified as the agents responsible for chick edema disease 15.

The hundreds of PCB, PCDD, and PCDF congeners, all with different potencies, were found to have varying profiles (i.e., relative proportions among congeners), based on different sources and distribution patterns. Consequently, the toxicity of these complex mixtures could not be rapidly assessed. Also, there was only limited information on the actual environmental concentrations of these DLCs in fish and wildlife due to the challenges of analyzing very low concentrations (picograms to nanograms per gram range). Chemical methods for DLCs were experimental, available only in limited laboratories, and costly. This led to the need for an analytical method to quantify low concentrations of DLCs and evaluate the potency of those mixtures in environmental samples, including fish and wildlife 10.

The advances in knowledge regarding the mechanistic aspects of the AhR, signal-transduction pathways, specific gene responses, and quantitative structure–activity relationships among end points provided the opportunity for the development of bioassay approaches to measure DLCs. In particular, the H4IIE rat hepatoma cell line was found to contain an intact AhR pathway with inducible CYP1A activity, measured as ethoxyresorufin O-deethylase 10. The H4IIE bioassay proved to be a useful screening tool for integrating the cumulative potencies of DLC mixtures in environmental samples, and it cost much less than chemical analysis 16. This bioassay represented a means of rapid assessment that led to the ability to quantify complex mixtures of DLCs, particularly in eggs of fish-eating waterbirds. Dose–response relationships between bioassay-derived potencies of DLCs in eggs of colonial waterbirds and signs of GLEMEDS in distinct populations of Caspian terns (Hydroprogne caspia), double-crested cormorants (Phalacrocorax auritus), Forster's terns (Sterna forsteri), and herring gulls provided strong evidence that DLCs were causing GLEMEDS in a variety of nesting bird populations in the lower Great Lakes 16–20. This piece of evidence proved to be critical for the determination of causality 4. Moreover, the use of bioassay systems, like the H4IIE bioassay, for measurement of DLCs and their potencies in environmental samples grew after this period and became an extremely useful approach for environmental assessments 21. The linkages of GLEMEDS in colonial waterbirds to DLCs also served as an important conceptual model for assessment of these chemical mixtures in other locations around the world.

Today, advances in molecular biology have allowed sequencing of the AhR and discovery of the basis for sensitivity differences toward DLCs in birds. It has been known for some time that the sensitivity of birds to embryotoxic effects of AhR agonists varies among species 8. The most sensitive avian species to DLC-related effects has consistently been the domestic chicken (Gallus gallus domesticus), while other species have sensitivities toward DLCs that are hundreds (Japanese quail, Coturnix japonica) to thousands (double-crested cormorant or herring gull) of times less than that observed in chickens. The reason for the differential sensitivity among avian species was unknown until recently, when it was discovered that amino acid sequences in the ligand binding domain of AhR1 portend the relative sensitivity of a species to DLC-induced embryotoxicity 22. Subsequent studies have reaffirmed the importance of two amino acids in AhR1 and demonstrated that the prototypical DLC, TCDD, is not always the most potent congener for eliciting dioxin-related toxicity in birds 23. Indeed, there appear to be three distinct categories of avian sensitivity based on the amino acid substitution patterns at these two residues sites (residues 324 and 380) in the binding domain of AhR1 24.

The ability to predict species' relative sensitivities based on the sequence of AhR1 is an important tool for evaluating risk from these chemicals, particularly for threatened and endangered species of birds in which toxicological testing is largely impractical. The new understanding of the relative potencies of various DLCs compared to TCDD will also be important for risk assessments and understanding the relative importance of different classes of DLCs. As DLCs continue to elicit responses in select avian populations of the Great Lakes, better toxicological evaluations of risk or recovery will allow better natural resource management actions.


Table S1 (49 KB PDF).