Effects of In Ovo Exposure to 3,3 0,4,4 0 -tetrachlorobiphenyl (PCB 77) on Heart Development in Tree Swallow (Tachycineta bicolor)

Transcription

1 Environmental Toxicology and Chemistry Volume 37, Number 1 pp , Received: 29 April 2017 Revised: 2 June 2017 Accepted: 31 July 2017 Environmental Toxicology Effects of In Ovo Exposure to 3,3 0,4,4 0 -tetrachlorobiphenyl (PCB 77) on Heart Development in Tree Swallow (Tachycineta bicolor) Tiffany Carro, a Mary K. Walker, b Karen M. Dean, c and Mary Ann Ottinger a, * a Department of Animal & Avian Sciences, University of Maryland, College Park, Maryland, USA b Division of Pharmaceutical Sciences, University of New Mexico, Albuquerque, New Mexico, USA c Department of Neuroscience, University of Lethbridge, Lethbridge, Alberta, Canada Abstract: Tree swallow (Tachycineta bicolor) eggs from 2 uncontaminated sites, the Patuxent Research Refuge (Laurel, MD, USA) and the Cobleskill Reservoir (Cobleskill, NY, USA) were dosed with polychlorinated biphenyl (PCB) 77 to evaluate effects on the developing cardiovascular system. To ensure embryonic viability, treatments were administered into the air cell at embryonic day 2.5 including: untreated (control), vehicle (filtered sterilized fatty acid mixture), 100 ng/g and 1000 ng/g egg. Eggs were dosed in the field with 0.2 ml/egg, returned to the nest, collected at embryonic day 13, hatched in the laboratory, and necropsied. The PCB 77-treated hatchlings were compared with uninjected, vehicle-injected, and environmentally exposed hatchlings collected from a PCB-contaminated Upper Hudson River (NY, USA) site. The PCB 77-treated embryos showed no effects on hatching success or hatchling mortality, heart index, or morphological measures of 4 distinct heart layers (heart width, length, septal thickness, total and ventricular cavity area) compared with controls. Hatchlings that had received PCB 77 exhibited increased incidence of a cardiomyopathy and absence of the ventricular heart wall compact layer (Chi square test;p < 0.001); environmentally exposed embryos showed no apparent effects. The compact layer is essential in development and overall heart function for ventricular cardiomyocyte proliferation and normal heart contraction. The finding that in ovo exposure to PCB 77 resulted in distinct cardiomyopathy has implications for long-term individual fitness. Environ Toxicol Chem 2018;37: C 2017 SETAC Keywords: Polychlorinated biphenyls; PCB 77; Heart morphology; Tree swallow; Compact layer INTRODUCTION Polychlorinated biphenyls (PCBs) are synthetic compounds composed of a biphenyl molecule with 1 to 10 chlorine substitutions and 209 known congeners. They were used extensively in manufacturing in the United States from early in the 20th century until their manufacture in the United States was banned in On the Hudson River (NY, USA), beginning in the 1940s, General Electric used PCBs in its capacitor manufacturing operations at its Hudson Falls and Fort Edward, New York (USA) facilities, releasing PCBs from these plants into the Hudson River [1 3]. Avian species are variable in their sensitivity to PCB congeners, but it has been well established that coplanar dioxin-like congeners such as PCBs 77, 81, and 126 induce toxicity at higher * Address correspondence to maotting@central.uh.edu Published online 2 August 2017 in Wiley Online Library (wileyonlinelibrary.com). DOI: /etc.3933 rates than non-dioxin-like congeners [4 8]. Recent contaminant analysis has shown that concentrations of PCBs 77, 81, and 126 account for approximately 86, 9, and 3% respectively, of the total PCB toxic equivalents (TEQs) in tree swallow eggs collected along the Upper Hudson River in 2004 [9]. These data, along with contaminant analysis showing 2000 pg TEQ/g egg for PCB 77, indicate that this congener is a significant contributor to summed PCB TEQs in tree swallow eggs on the Hudson River [1,2,10]. Environmental exposure to PCBs has been linked to diverse adverse effects in birds and other species, including cancer [11,12], immune system dysfunction [13,14], decreased reproduction [15], endocrine disruption [16,17], and neurological and cognitive dysfunction [18]. In addition, environmental exposure to PCBs has been linked to effects on cardiovascular development in wild birds [19]. Previous research has also demonstrated that exposure of birds to dioxins and PCBs, including PCB 126, resulted in increased incidence of ventricular dilation, thinned ventricular wall, elongated apical heart region, increased overall heart weight, and decreased chronotropic responsiveness [20 22]. C 2017 SETAC wileyonlinelibrary.com/etc

2 PCB 77 effects on heart development in the tree swallow Environmental Toxicology and Chemistry, 2018;37: The effects of environmental exposure to PCBs on tree swallow (Tachycineta bicolor) embryos and nestlings have been well documented at the Upper Hudson River and other locations across the country [9,23 31]. Moreover, contaminant analysis of bird eggs and nestlings collected from the Upper Hudson River showed that PCBs were detectable; in one report, geometric means of concentrations were 6.8 mg/g wet weight in tree swallows, 11.7 mg/g wet weight in kingfisher, and 12.6 mg/g wet weight in sandpiper; however, the PCB TEQs were greater in the tree swallows because of high PCB 77 concentrations [1,9,10,25,30,31]. Dioxin-like characteristics are found in PCB 77, as well as a relatively high toxicity equivalency factor (TEF) of 0.05 in chickens (TEF factors for PCB 77 in birds range from to 0.05), which suggests that PCB 77 may be a principal congener linked to adverse health effects mediated by the aryl hydrocarbon receptor [5,32]. Thus the high relative concentration of PCB 77 measured in eggs collected from the Hudson River environs and the high TEF of this PCB would predict adverse effects, especially with environmental exposure or injected treatment during embryonic development. Tree swallow eggs were collected from 2 uncontaminated sites, the Patuxent Research Refuge (Laurel, MD, USA) and the Cobleskill Reservoir (Cobleskill, NY, USA), and one contaminated site on the Upper Hudson River. Eggs from the Upper Hudson River were not dosed and were thus treated as environmentally exposed embryos. Eggs collected from the Patuxent Research Refuge and Cobleskill were dosed with 1 of 2 concentrations of PCB 77 and were necropsied at hatch. Sample sizes for the birds we studied were maximized while adhering to the bird collection permit. The purpose of the present study was to determine whether PCB 77 increased mortality in tree swallows eggs and to assess other impacts of this congener, particularly heart defects in the developing embryo. MATERIALS AND METHODS Field studies and egg injection All field studies and experimental procedures were conducted under an approved Institutional Animal Care and Use Committee protocol from the University of Maryland, College Park (MD, USA). Tree swallow eggs were collected under permits from the Patuxent Research Refuge (US Fish and Wildlife Service), US Fish and Wildlife Service (Migratory Birds), and the Maryland Department of Natural Resources for the studies at the Patuxent Research Refuge; and from the New York State Department of Environmental Conservation and US Fish and Wildlife Service (Migratory Birds) for the Cobleskill Reservoir (hereafter Cobleskill) and Upper Hudson River. Tree swallow eggs originated from several field sites; all studies were conducted during the 2008 field season. A total of 256 tree swallow eggs was used for determination of mortality and in ovo developmental effects of PCB 77 (Patuxent Research Refuge: n ¼ 125; Cobleskill: n ¼ 92; Upper Hudson River: n ¼ 39). Additional eggs (both developing and not developing) were collected from these sites for contaminant analysis (Patuxent Research Refuge: 22; Cobleskill: 31). The total numbers of tree swallow eggs at the Patuxent Research Refuge and Cobleskill were 356 and 245 from 71 and 45 clutches, respectively. Of these clutches, 65 at the Patuxent Research Refuge and 43 at Cobleskill contained more than 3 eggs and could be included in the present study. Clutches were initiated from the 22 April 2008 until 24 June 2008 at the Patuxent Research Refuge. Full nest activity was not recorded at Cobleskill; however, the first clutch was initiated on 9 May 2008, and the last clutch collected for egg injections was initiated on 30 May The Upper Hudson River was considered contaminated based on previous studies and measurements of PCBs in eggs collected from nests in that region [29 31]. Embryos in these eggs were considered environmentally exposed and were not dosed with PCB 77 treatments. Eggs transferred from Cobleskill were collected at the same time point (day 10) as at the Patuxent Research Refuge and then stored in a Koolatron heated portable incubator set at a temperature suitable to maintain the eggs for an 8 to 10-h drive from New York to the University of Maryland. Similarly, viable eggs collected from the Upper Hudson River at approximately embryonic day 13, as estimated by lay date and relative embryo size assessed by candling were transported to the University of Maryland for incubation and hatching as described later in this section, with samples collected at hatch. Tree swallow eggs used in the injection study were collected from 2 additional clean reference field sites: the Patuxent Research Refuge and Cobleskill. Tree swallow eggs were injected at both the Patuxent Research Refuge and Cobleskill. Eastern bluebird egg injections were only conducted at the Patuxent Research Refuge. At both locations nests were checked twice a week throughout the nesting season and more frequently (every 1 2 d) when egg laying was in progress within a nestbox. Thus eggs were candled at least 3 times in the field following injection. If any eggs were determined to be dead at the time of candling, they were removed from the nest and transported back to the University of Maryland. Within the first week of development the egg contents were removed and archived. Untreated or vehicleinjected eggs could be added to the pool of samples for contaminant analysis. If embryos were in the second week of development, they were evaluated for stage of development and deformities. If abnormal embryos were found, they were photographed, preserved, and archived. Eggs that failed to hatch were opened and staged, and the condition of the embryo was noted. Deformities were scored for the presence or absence of crossed bill, shortened upper bill, missing or deformed eyes, edema of the neck and head area, incomplete ossification of skull (brain not enclosed in skull), gastroschisis in post stage 45 embryos, malformed or clubbed feet, asymmetrical body form, malposition in the egg, and any other abnormal appearances were noted. Embryos from eggs collected in the field and incubated in the laboratory were dissected within 24 h of hatching. Hatchlings were euthanized by cervical dislocation and necropsied immediately to collect tissue. Body and organ weight measurements were taken during dissection. Weights were measured to the nearest 0.01mg. wileyonlinelibrary.com/etc C 2017 SETAC

3 118 Environmental Toxicology and Chemistry, 2018;37: T. Carro et al. The PCB 77 solutions were prepared in a fatty acid mixture that represented ratios of fatty acids in egg yolk, with the fatty acid mixture composed of palmitic, stearic, oleic, and linoleic acids (Nuchekprep) in a ratio of 10:5:55:30. All solutions were filter-sterilized with a 0.2-mm syringe-driven filter unit before use [33]. The PCB 77 treatments were solubilized in the filtered sterilized fatty acid mixture at 2 concentrations: 100 ng/g egg and 1000 ng/g egg. The selection of doses was based on reported concentrations of PCBs in tree swallow eggs of 18 ng/g to as high as ng/egg, and the observed concentrations of PCB 77 as part of this mixture in the Hudson River area of concern [3,9,10,25]. Eggs were injected as described below. Nests were checked daily and egg laying was monitored. Eggs were candled, and injections were administered at approximately 18% embryonic development or embryonic day 2.5 to standardize embryonic stage at treatment. Individual eggs were randomly assigned the following treatments: untreated, 0 mg/g (vehicle), 100 ng/g, and 1000 ng/g. Treatments were administered in the field. All eggs received the same volume of dosing solutions; therefore, treatment concentrations per egg are approximate. Notably, body weight of tree swallow hatchlings from both Patuxent Research Refuge (df ¼ 3, 52, F ¼ 0.997, p ¼ 0.402) and Cobleskill (df ¼ 3, 54, F ¼ 0.667, p ¼ 0.576) did not differ across treatment groups, and there was no significant difference between untreated groups (df ¼ 2, 60, F ¼ 0.324, p ¼ 0.725). Mean body weights of hatchlings from both the Patuxent Research Refuge and Cobleskill in 2007 were approximately 1250 mg, which is similar to the average weight of untreated hatchlings from the Upper Hudson River ( mg), Patuxent Research Refuge ( mg), and Coblekskill ( mg). Thus embryos in each treatment group would be predicted to receive a similar dosage. A small hole was drilled (with a Dremel drill) into the blunt side of the egg, and a total injection volume of 0.2 ml/egg was deposited into the air cell using a reverse displacement pipettor (Rainin Instrument). The injection site was sealed with wax, and eggs were immediately returned to nests for incubation, with egg handling time averaging less than 5 min. The number of eggs injected for each site and treatment were: Patuxent Research Refuge n ¼ 25, 29, 32, 39 and Cobleskill n ¼ 18, 21, 24, 29, respectively. At approximately embryonic day 13, eggs from all treatment groups were collected, transported to the University of Maryland, placed in hatching trays and maintained at 37 8C and 60% humidity until hatch. Heart collection Hatchlings were euthanized by cervical dislocation within 12 h of hatch; the heart was dissected while beating and placed in icecold 25 nm KCl until beating ceased ( 30 s). The heart was then rinsed thoroughly in 1 phosphate-buffered saline (PBS; 1, ph 7.4, osmolarity mosm/kg; Gibco, ThermoFisher Scientific), blotted dry, weighed, and immersed in ice-cold 10% neutral-buffered formalin for storage at 4 8C for 48 h. Following fixation, hearts were removed from fixative and immersed in 1 PBS solution for 30 min (twice) at 4 8C and transferred to a 5% sucrose solution overnight at 4 8C. The heart was dehydrated by a stepwise immersion into ethanol (EtOH; Pharmco-Asper) at the following increments: 50% EtOH, 70% EtOH, 80% EtOH, 95% EtOH, 100% EtOH (twice) for 60 min each. The heart was immersed in a 1:1 mixture of the clearing agent Hemo-De (Meridian Bioscience) and EtOH for 30 min at room temperature, transferred to 100% Hemo-De (for 15 min at 58 8C, followed by 1:1 Hemo-De:paraffin (Triangle Biomedical Science) for 60 min at 58 8C, and then immersed in 100% paraffin for 30 min at 58 8C (repeated 3 times). Following the last immersion, the heart was transferred to a 100% paraffin-filled mold, oriented uniformly, and allowed to solidify for 24 h. Heart sectioning and staining Hearts were sectioned transversely, from apex to cranial region (base) at 10 mm, and wet-mounted onto glass slides. Once dried, sections were taken through a standard H & E staining procedure. Briefly, slides were dewaxed in HemoD for 3 min (3) and rehydrated using reverse stepwise increments of EtOH: 100% EtOH, 95% EtOH, 70% EtOH, and 1 PBS for 3 min each; nuclear staining using Gill s hematoxylin (<5 s) and cytoplasmic staining using 0.5% eosin solution (6 min). Tissue was dehydrated again for 30 s at 70% EtOH and 100% EtOH, immersed in HemoD for 3 min, coverslipped, and dried for 24 h before storage. Analysis of heart indices and morphological measurements Heart wet weight and indices including heart weight/body weight percentage were recorded and calculated for each individual. Detailed morphological analysis was conducted on a subset of samples to examine more specific morphology of the heart. A total of 25 tree swallow hearts were analyzed, representing the untreated, 0 mg/g, 100 ng/g egg, and 1000 ng/g egg treatment groups and the environmentally exposed Upper Hudson River nondosed group. This subset of hearts analyzed was selected randomly, and all subsequent imaging and measurements were conducted blind to treatments. Hearts were examined for septal defects and any additional gross anatomical defects. Images of sections were captured using the 4 achroplan Zeiss objective, Zeiss microscope (model , Carl Zeiss), and a Photometrics Coolsnap fx camera (Photometrics). Four layers were evaluated as defined by morphometric structures as follows: layer 1: most caudal section, identified by the left ventricular wall and cavity (95% of the section), and the very beginning of the right ventricular opening; layer 2: contains both left and right ventricles, with both clearly prevalent, with no evidence of papillary muscles or valves; layer 3: identified by the upper portion of both ventricles with emerging aorta present in the center of the section, left ventricle papillary muscles emerging, and evidence of the right ventricle papillary muscle; layer 4: identified by the emergence of the pulmonary artery without the emergence of the aortic valve, and was the most cranial section analyzed. Images were captured using IP Lab 3.6 software (Biovision Technologies) and analyzed using ImageJ C 2017 SETAC wileyonlinelibrary.com/etc

4 PCB 77 effects on heart development in the tree swallow Environmental Toxicology and Chemistry, 2018;37: software (National Institute of Health). Treated groups were compared with controls with all measures taken in millimeters (mm). Measurements taken for layer 1 were width, depth, left ventricle compact layer, left ventricle ventral compact layer, total area, and left ventricle cavity area. Layer 2 measurements were width, depth, left ventricle dorsal compact layer, left ventricle ventral compact layer, and total area. Layer 3 measurements were width, depth, left ventricle dorsal compact layer, right ventricle ventral compact layer, right ventricle ventral compact layer, septal thickness, and total area. Layer 4 measurements were width, depth, and total area (methodology developed by M.K. Walker, University of New Mexico, Albuquerque, NM, USA). Compact layer analysis Because the measurements taken indicated that PCB 77 exposure affected development of the compact layer in hatchlings, additional hearts were sectioned, and layers 1 and 2 were analyzed in detail. These hearts, as well as hearts used for morphological measurements, were scored to determine whether the compact layer was present or absent in layers 1 and 2, along the dorsal and ventral walls of the sectioned samples. Differentiation between the compact layer and the trabeculated heart layer were clearly identified and visualized using the 4 objective. Sample sizes varied because of hatch success rates as follows: untreated (n ¼ 15), vehicle (n ¼ 11), 100 ng/g (n ¼ 14), 1000 ng/g (n ¼ 15), and environmentally exposed (n ¼ 8). Histology and imaging were completed as described previously, and the analyses were performed blind [34,35]. Statistical analysis Treatment effects on embryo survival and heart weight were tested by one-way analysis of variance (ANOVA), using a generalized linear design model (JMP 8; SAS). Normality of the data was determined prior to proceeding with the statistical analysis, with appropriate correction, if needed. Each heart layer was analyzed to discern specific effects on development of distinct portions of the heart. The analyses were conducted on the measurements taken for each individual and for each heart layer. Prior to running the statistic, there was a check for normality of the data set. A one-way ANOVA was used to test for effects of treatment for each treatment layer separately, with the significance criteria of p < 0.05 for each heart layer, using JMP 8. A separate comparison was made to determine whether treatment affected the development of the compact layer by Chi square statistic. Specifically, samples were categorized for either the presence or absence of the compact layer, and then Chi square analysis was used to detect treatment related differences. Statistical analysis was conducted using JMP 8 software. TABLE 1: Hatching success (%) of untreated tree swallow eggs across experimental groups Group Sample size Mean SE Cobleskill controls PRR population PRR controls Upper Hudson Cobleskill population PRR ¼ Patuxent Research Refuge; SE ¼ standard error. nests from which sample eggs had been collected. The mortality of both untreated experimental eggs and that of local populations were considered for the present study (Table 1). Mortality was found to be statistically different between untreated experimental eggs at Cobleskill and both the untreated control and population of untreated eggs at Patuxent Research Refuge (df ¼ 4, 430; F ¼ 3.73, p ¼ 0.005). The increased sample sizes at the Patuxent Research Refuge and Cobleskill illustrated that a random subsection of the population enabled detection of this difference. The PCB 77 treatments did not affect hatching success rate or survivability to hatch (Figure 1). There was a significant difference between the 1.0 mg/ml injection group and the untreated group at Cobleskill (df 3, 88, F ¼ , p ¼ 0.018), with the untreated controls having greater hatching success. Hatching success of the untreated Cobleskill groups was higher than that observed for the unsampled Cobleskill group (Table 1 and Figure 1; p < 0.018). At the Patuxent Research Refuge, there was no significant difference among any of the groups (df 3, 123, F ¼ 0.051, p ¼ 0.985). Combining the Cobleskill untreated control group with the larger data from the rest of the unsampled Cobleskill population eliminated this difference (df 3, 232, F ¼ 1.182, p ¼ 0.317) and reduced hatching success for untreated eggs from % (n ¼ 18) to (n ¼ 162). Thus there were no differences between the untreated and treatment groups in hatching success or survivability to hatch. Significant gross deformities or abnormalities were not observed in embryos that died prior to hatch (data not shown). Tree swallow hatchling hearts from the 3 collection sites showed no difference in wet weight (Table 2) or heart index (heart weight/body weight %; based on wet weight; Table 2; RESULTS Hatching success, heart weight, and indices Mortality of untreated control samples in our experimental group was comparable to hatching success of eggs left in the FIGURE 1: 2008 tree swallow hatching success following polychlorinated biphenyl (PCB) 77 egg injection. Data show means standard error. CoblNY ¼ Cobleskill; UHR ¼ Upper Hudson River; PRR ¼ Patuxent Research Refuge. wileyonlinelibrary.com/etc C 2017 SETAC

5 120 Environmental Toxicology and Chemistry, 2018;37: T. Carro et al. TABLE 2: Heart weights and index for polychlorinated biphenyl (PCB) 77-treated eggs from uncontaminated sites (the Patuxent Research Refuge and Cobleskill) and environmentally exposed eggs from the PCB-contaminated site Upper Hudson River a Treatment PRR Cobleskill UHR Heart weight (mg) Untreated (12) (17) Vehicle (12) (12) 100 ng/g egg (15) (12) 1000 ng/g egg (17) (17) UHR e.e (33) Heart index (heart weight/body weight %) Untreated (12) (17) Vehicle (12) (12) 100 ng/g egg (15) (12) 1000 ng/g egg (17) (17) UHR e.e (33) a Mean standard error (SE), with numbers in parentheses of heart weight (mg) and index (heart weight/body wt %) for 100 ng/g egg, 1000 ng/g egg, untreated, and vehicletreated tree swallow hatchlings from the Patuxent Research Refuge and Cobleskill. Eggs from the Upper Hudson River were environmentally exposed to PCB levels and not dosed with PCB 77. Heart indices were compared among uncontaminated, injected, and Upper Hudson River eggs. PRR ¼ Patuxent Research Refuge; UHR ¼ Upper Hudson River; e.e. ¼ environmentally exposed. p ¼ 0.543). Two samples from the vehicle group at the Patuxent Research Refuge were eliminated because of improper collection techniques. Morphological measurement analysis of any heart layer was not affected by PCB 77 (Table 3; p > 0.05). Compact layer deformities significantly increased with PCB 77 exposure (Table 4; p < 0.001). Heart morphological measurements Exposure to 100 and 1000 ng/g egg weight concentrations of PCB 77 did not affect heart width, depth, length of dorsal and ventral compact layers (when present), total area, septal thickness, or ventricular cavity area compared with controls or environmentally exposed hearts (Table 3; p > 0.05). Heart morphological structures, with the exception of the ventricular compact layer (Figure 2; layers 1 3), did not differ between PCB 77-treated and control swallows. Compared with the control group, the compact layers (layers 1 3) were absent in the highdose treatment (1000 ng/g; Table 3). Layer 1 was absent (p ¼ 0.03) in the 1000 ng/g PCB 77-treated hatchlings of the dorsal and ventral compact layers. Layers 2 and 3 ventral right ventricular compact layers were absent (p ¼ 0.01 and p ¼ 0.04, respectively; Figure 2) in hatchlings in the 1000 ng/g PCB 77-treated hatchlings. There was no difference in the thickness of the compact layer of animals from PCB-treated groups when the compact layer was intact (Table 3; p > 0.05), and the trabeculated ventricular layers were present in all of the animals sampled. There was no significant difference in measurements and morphology between control samples and environmentally exposed hearts from the Upper Hudson River. There was no morphological effect on atrium formation in the PCB 77-treated birds compared with control groups (data not shown). Given the response observed in layers 1 and 2, additional hearts were examined from the Patuxent Research Refuge, Cobleskill, and Upper Hudson River to specifically focus on compact layer alternations. Analyses showed an increase in compact layer deformities for embryos exposed at both PCB 77 treatment levels. Analysis determined that PCB 77-treated embryos at both 100 and 1000 ng/g dose concentrations had CL deformities (Chi square, p < 0.001; Table 4). There was no difference between control hearts and environmentally exposed hearts from the Upper Hudson River. DISCUSSION Exposure to PCB 77 did not affect survivability It has been well established that many environmental chemicals are maternally deposited into eggs [36 39]. However, despite the aryl hydrocarbon receptor (AhR) activity and TEQ of PCB 77, no adverse effects had been observed on embryonic survival in treated or environmentally exposed tree swallows. This lack of lethality may be ascribed to interspecific differences of PCB congeners on the activation of the AhR; according to these data, tree swallows have lower sensitivity to PCBs than other avian species [5,7,8,40]. Similarly, the potency of dioxin-like PCBs with similar chemical structures to dioxins (e.g., 2,3,7,8-tetrachlorodibenzodioxin) is toxicologically categorized by TEFs [41,42]. Thus PCB 77 (dioxin-like PCB congener), with a TEF of in kestrels (Falco sparverius), in turkeys (Meleagris gallopavo), and 0.05 in chickens (Gallus gallus domesticus), should likely have significant adverse effects [6,14,43]. Previous dosing studies with PCB 77 showed no effects on herring gulls and mallards at concentrations of 1000 to 5000 ng/g [42]. Conversely, Hoffman and colleagues [44] reported in ovo effects of PCB 77 in American kestrels, with a median lethal dose (LD50) of 316 ng/g accompanied by overt malformations and edema at 100 and 1000 ng/g; chickens had a LD50 of 2.6 ng/g with overt malformations at 6 and 12 ng/g. Our study used the TEF of PCB 77 and data from the literature to determine which dosing concentrations to use. Based on the TEQ of PCB 77, the prediction would have been elevated tree swallow embryo mortality, especially when the observed TEQ-related effects documented across a range of species are considered [32,41]. However, this outcome was not observed. Therefore, these data suggest the limitation of TEQs when studying wild bird populations and furthermore, even if 2 species are classified as being in the same general category of sensitivity C 2017 SETAC wileyonlinelibrary.com/etc

6 PCB 77 effects on heart development in the tree swallow Environmental Toxicology and Chemistry, 2018;37: TABLE 3: Spatial measurements of morphological heart structures at layers 1, 2, 3, and 4 in tree swallow hatchlings dosed with polychlorinated biphenyl (PCB) 77 during early embryonic development, control groups, and environmentally exposed (Upper Hudson River) embryos a Width (mm) Depth (mm) D LV CL (mm) VLV CL (mm) Total area (mm 2 ) LV area (mm 2 ) Septum (mm) Treatment Average Range Average Range Average Range Average Range Average Range Average Range Average Range Layer 1 PRR untreated PRR 100 ng/g PRR 1000 ng/g UHR e.e Layer 2 PRR untreated PRR 100 ng/g PRR 1000 ng/g UHR e.e Layer 3 PRR untreated PRR 100 ng/g PRR 1000 ng/g UHR e.e Layer 4 PRR untreated PRR 100 ng/g PRR 1000 ng/g UHR e.e a Control (untreated data presented; there was no difference from vehicle control group), 100 ng/g, and 1000 ng/g PCB 77-treated hearts were randomly selected from the PRR uncontaminated site. The Upper Hudson River environmentally exposed hearts were randomly selected for a contaminated site reference. Layer identification was determined using morphological landmarks, described in the Materials and Methods section. There was no difference across treatments in morphology measurements, including compact layer depth when compact layer was present. Averages and ranges are expressed in millimeters or mm 2. No compact layer present in the samples. No difference across treatments when compact layer was present; when compact layer was not present, a value of 0.0 mm was recorded. D ¼ dorsal; V ¼ ventral; LV ¼ ventral left ventricle; RV ¼ right ventricle; CL ¼ compact layer; PRR ¼ Patuxent Research Refuge; UHR e.e. ¼ Upper Hudson River environmentally exposed. wileyonlinelibrary.com/etc C 2017 SETAC

7 122 Environmental Toxicology and Chemistry, 2018;37: T. Carro et al. TABLE 4: Compact layer detailed analysis in tree swallow hatchling hearts exhibiting compact layer deformities/absence following in ovo exposure to polychlorinated biphenyl (PCB) 77 Treatment a,b No. No. of hearts with intact CL c % Hearts with CL deformities Untreated Vehicle ng/g ng/g UHR e.e. d a Untreated, vehicle, 100 ng/g, and 1000 ng/g hatchling hearts were randomly sampled from the Patuxent Research Refuge and Cobleskill treatment groups. b Chi square analysis determined significant differences between PCB 77-treated hearts and control samples (p < 0.001), indicated by an asterisk ( ). c Compact layer analysis (presence or absence of this morphological structure) was determined on the dorsal and ventral ventricular wall regions in layers 1 and 2. Hearts with compact layer absence in any location were classified as compact layer deformed hearts. d The environmentally exposed group, collected from the Upper Hudson River, was not significantly different from control groups. CL ¼ compact layer; UHR e.e. ¼ Upper Hudson River environmentally exposed. to embryo mortality based on AhR, there can be variation within this category in sensitivity to different apical outcomes. PCB 77 affects heart development in tree swallow hatchlings It has been shown in vitro that PCB 77 induces proinflammatory cytokines that may be disruptive to developmental pathways [45]. The present study focused on studying PCB 77 because it is a primary component of the PCB mixtures found in the Hudson River environs. The findings demonstrated that mixtures containing significant concentrations of PCB 77 should have the capacity to adversely affect heart development in wild birds. However, the Upper Hudson River environmentally exposed samples did not show effects on heart development. A possible explanation would be regional differences in exposure, including low relative exposure to PCB 77 in the mixture in that region. FIGURE 2: A comparison of tree swallow hatchling hearts treated in ovo with polychlorinated biphenyl (PCB) 77 (1000 ng/g) with untreated hearts, analyzed at layers 1, 2, 3, and 4 for morphological heart defects. Tree swallow hatchling heart samples shown are representative of untreated and 1000 ng/g PCB 77-treated embryos collected from the Patuxent Research Refuge. There was no significant difference for all morphological measurements taken across treatments. The PCB 77-dosed hatchlings (100 ng/g and 1000 ng/g concentrations) showed a significant increase in compact layer absence compared with the control. Layer 1: (A) control, (B) PCB 77 treated. Layer 2: (C) control, (D) PCB 77 treated. Layer 3: (E) control, (F) PCB 77 treated. Layer 4: (G) control, (H) PCB 77 treated. Layer identification is described in the Materials and Methods section. LV ¼ left ventricle; RV ¼ right ventricle; D CL ¼ dorsal compact layer; V CL ¼ ventral compact layer; TL ¼ trabeculated layer; S ¼ septum; A ¼ aorta; PM ¼ pulmonary muscles; M ¼ mitral valve; PA ¼ pulmonary artery; ( ) ¼ CL absence. C 2017 SETAC wileyonlinelibrary.com/etc

8 PCB 77 effects on heart development in the tree swallow Environmental Toxicology and Chemistry, 2018;37: It must be noted that only viable eggs from the Upper Hudson River (environmentally exposed) were collected on embryonic days 12 to 13 for hatching in the laboratory. Thus it is possible that the selection of viable eggs favored a sampling pool of embryos that were more tolerant to overall PCB exposure, and the difference in timing of exposure to PCB between the Upper Hudson River and experimentally injected eggs may also have led to variation in the adverse outcomes. Finally, the lack of effects for Upper Hudson River eggs could also be because of antagonistic interactions between PCB 77 and less planar PCBs in the mixture. Other studies in which hearts were examined from nestling house wrens (Troglodytes aedon), tree swallow, and Carolina chickadee (Parus carolinesis) exposed to environmental mixtures of PCBs showed heart defects including reduced heartto-body ratio, deformities, microsurface roughness, apical and ventricular deformities, and thinning of the ventricular wall [46]. In laboratory studies, domestic chick embryos exposed to environmentally relevant PCB mixtures, dioxin-like PCB congeners, and dioxins had ventricular dilation, thinning of the ventricular wall, septal wall malformation, absent trabeculated layer, absent compact layer, and reduced chronotropic responsiveness [20 22,34,35,47 49]. Cardiomyopathy of the ventricular wall, specifically the absence of the compact layer, was also observed in chicken embryos exposed to a Upper Hudson River environmentally relevant PCB mixture [34,35]. Moreover, it was found that early embryonic exposure to the PCB mixture interfered with cardiomyocyte proliferation. The compact layer is a portion of the myocardial wall that is formed throughout incubation by proliferation of the cardiomyocytes. The myocardial wall is comprised of a highly proliferative region, the compact layer, and a highly trabeculated inner zone [50,51]. The compact layer is absolutely necessary for myocardial growth and proper overall heart growth, resulting from cardiomyocyte proliferation, and proliferates at high rates through hatch in avian species, with the major increase in ventricular wall thickness occurring between embryonic days 8 and 14 [52]. The high volume of cardiomyocytes in the compact layer is essential for normal muscle contraction and cell turnover in the organ [53]. Absence of such a vital portion of the ventricular wall in PCB 77-treated tree swallow hearts would lead to decreased fitness in these animals, especially in stressful periods such as migration. Furthermore, absence of the compact layer has been associated with compromised heart contraction and reduced myocardial proliferation in other species, which would contribute to decreased overall fitness [54]. In some species, even minor localized absences of the compact muscle layer can result in cardiomyopathies, heart failure, and even cardiac death [55,56]. Although these are clinically observed cardiomyopathies, the effects of these abnormalities of the compact muscle layer provide insight into potential impacts on the organism. All avian species, especially those that migrate, have extensive cardiac requirements as they feed and protect their nests and during migration; therefore, any compromise of the heart morphology could disadvantage an individual. Furthermore, in humans, exposure to environmental contaminants in utero increases the possibility of an underdeveloped left ventricle [57], a similar phenotype to the thinning ventricular walls seen in PCB-exposed birds. It has been estimated that 30% of congenital cardiovascular malformations in human infants could be linked to environmental risk factors [58]. The link between environmental contaminants and congenital cardiovascular malformations in humans is reminiscent of some observations in wild birds. Our data confirm that PCB 77 adversely affects the proper formation of the compact layer; this suggests that although PCB 77 is not lethal at the doses tested in the present study, there is a potential for impacting cardiac function in affected individuals. While the present study demonstrated increased compact layer defects with PCB 77 exposure, some of the other adverse heart effects seen in previous avian studies were not observed. One possible explanation is that treatments were administered at a later stage of development (18% of incubation). Thus embryonic mortality was minimized by injecting after development had progressed [45]. While this approach minimized mortality and verified the presence of a developing embryo, the timing of injection occurred after the initiation of heart development. In chickens, embryonic heart development begins at approximately Hamburger Hamilton stage 7, with the emergence of the cardiogenic mesoderm at approximately 25 h of incubation [59]. In tree swallows, which have an 18-d incubation period, heart formation would be predicted to occur at 21 h of incubation, based on a relative Hamburger Hamilton staging. This timing is approximately 40 h earlier than the time of PCB exposure in the present experiment. Therefore, the embryos in the present study were treated in ovo to PCBs well after the initiation of heart development [60]. Studies that have linked PCB exposure to altered heart weight and dilated cavities have exposed embryos prior to incubation, which might account for the differences relative to heart size and weight [19 22]. This suggests that we could have observed a greater effect, including impact on heart weight, morphological measurements, and size, if in ovo dose treatments had been administered earlier. Acknowledgment The authors thank L. Baltos and A. Marcell for their valuable assistance in fieldwork. Our sincere thanks go to M. Bohannon, A. Tschiffely, and K. Davani for all their assistance. The PCB 77 and fatty acid mixture was provided by D.E. Tillitt, Columbia Environmental Research Center, US Geological Survey, Columbia, OH. Morphological analysis was developed by M.K. Walker, University of New Mexico. The present study is in partial fulfillment of the requirements for the PhD degree (of T.Carro) at the University of Maryland. Disclaimer The present study was supported by the US Fish and Wildlife Service and the Hudson River Natural Resource Trustees. The conclusions and opinions presented are those of the authors and do not represent the official position of the funding agency, the Hudson River Trustees, or the United States Government. Data Availability Please contact the corresponding author, Mary Ann Ottinger (maotting@central.uh.edu), who will forward the request to US Fish and Wildlife Service and the Hudson River Natural Resource Trustees. wileyonlinelibrary.com/etc C 2017 SETAC

9 124 Environmental Toxicology and Chemistry, 2018;37: T. Carro et al. REFERENCES [1] Hudson River Natural Resource Trustees Design and preparation of a custom 58-congener PCB mixture dosing solution for avian egg injection studies. Hudson River Natural Resource Damage Assessment. Report dated December 20, Final. Public Release Version. US Department of Commerce, National Oceanic and Atmospheric Administration, Hudson River NRDA, Silver Spring, MD. [2] Hudson River Natural Resource Trustees Congener-specific analysis of polychlorinated biphenyl residues in tree swallow chicks, eggs and other biota from the Hudson River. W.U Report dated November 25, Final Laboratory Report FY US Department of Commerce, National Oceanic and Atmospheric Administration, Hudson River NRDA, Silver Spring, MD. [3] Hudson River Natural Resource Trustees Preparation of individual and custom PCB mixture dosing solutions for avian egg injection studies associated with injury determination under the Hudson River NRDA. USGS Biochemistry & Physiology Branch Laboratory report FY 2011, dated March 14, Final. Public Release Version. August 1, US Department of Commerce, National Oceanic and Atmospheric Administration, Hudson River NRDA, Silver Spring, MD. [4] McKinney JD, Waller CL Polychlorinated biphenyls as hormonally active structural analogues. Environ Health Perspect 102: [5] Kennedy SW, Lorenzen A, Jones SP, Hahn ME, Stegeman JJ Cytochrome P4501A induction in avian hepatocyte cultures: A promising approach for predicting the sensitivity of avian species to toxic effects of halogenated aromatic hydrocarbons. Toxicol Appl Pharmacol 141: [6] Elliott JE, Wilson LK, Langelier KW, Norstrom RJ Bald eagle mortality and chlorinated hydrocarbon contaminants in livers from British Columbia, Canada, Environ Pollut 94:9 18. [7] Head JA, Hahn ME, Kennedy SW Key amino acids in the aryl hydrocarbon receptor predict dioxin sensitivity in avian species. Environ Sci Technol 42: [8] Herve JC, Crump D, Jones SP, Mundy LJ, Giesy JP, Zwernick MJ, Bursian SJ, Jones PD, Wiseman SB, Wan Y, Kennedy SW Cytochrome P450A induction by 2,3,7,8-tetrachlorodibenzo-p-dioxin and 2 chlorinated dibenzofurans in primary hepatocyte cultures of three avian species. Toxicol Sci 113: [9] Custer CM, Custer TW, Dummer PM Patterns of organic contaminants in eggs of an insectivorous, an omnivorous, and a piscivorous bird nesting on the Upper Hudson River, New York, USA. Environ Toxicol Chem 29: [10] Secord AL, McCarty JP, Echols KR, Meadows JC, Gale RW, Tillitt DE Polychlorinated biphenyls and 2,3,7,8-tetrachlorodibenzo-pdioxin equivalents in tree swallows from Upper Hudson River, New York State, USA. Environ Toxicol Chem 18: [11] Safe SH Polychlorinated biphenyls (PCBs). Environmental impact, biochemical and toxic responses, and implications for risk assessment. Crit Rev Toxicol 24: [12] Silberhorn EM, Glauert HP, Robertson LW Carcinogenicity of polyhalogenated biphenyls: PCBs and PBBs. Crit Rev Toxicol 20: [13] McNabb FM Biomarkers for the assessment of avian thyroid disruption by chemical contaminants. Avian Poult Biol Rev 16:3 10. [14] Lavoie ET, Grasman KA Effects of in ovo exposure to PCBs 126 and 77 on mortality, deformities, and post-hatch immune function in chickens. J Toxicol Environ Health A 70: [15] Golub MS, Donald JM, Reyes JA Reproductive toxicity of commercial PCB mixtures: LOAELs and NOAELs from animal studies. Environ Health Perspect 94: [16] Van den Berg KJ Interaction of chlorinated phenols with thyroxine binding sites of human transthyretin, albumin, and thyroid binding globulin. Chem Biol Interact 76: [17] Gould JC, Cooper KR, Scanes CG Effects of polychlorinated biphenyls on thyroid hormones and liver type I monodeiodinase in the chick embryo. Ecotoxicol Environ Saf 43: [18] Jacobson JL, Jacobson SW, Humphrey HEB Effects of in utero exposure to polychlorinated biphenyls and related contaminants on cognitive functioning in young children. J Pediatr 116: [19] DeWitt JC, Meyer EB, Henshel DS Environmental toxicity studies using chickens as surrogates for wildlife: Effects of injection day. Arch Environ Contam Toxicol 48: [20] Walker MK, Pollenz RS, Smith SM Expression of the aryl hydrocarbon receptor (AhR) and AhR nuclear translocator during chick cardiogenesis is consistent with tetrachlorodibenzo-p-dioxin induced heart defects. Toxicol Appl Pharmacol 107: [21] Walker MK, Catron TF Characterization of cardiotoxicity induced by 2,3,7,8-tetrachlorodibenzo-p-dioxin and related chemicals during early embryo development. Toxicol Appl Pharmacol 167: [22] Kopf PG, Walker MK Overview of developmental heart defects by dioxins, PCBs, and pesticides. J Environ Sci Health 27: [23] Custer TW, Heinz GH Reproductive success and nest attentiveness of mallard ducks fed Aroclor Environ Pollut 21:313. [24] Bishop CA, Koster MD, Chek AA, Hussell DJ, Jock K Chlorinated hydrocarbons and mercury in sediments, red winged blackbirds (Agelaius phoeniceus) and tree swallows (Tachycineta bicolor) from wetlands in Great Lakes, St. Lawrence River basin. Environ Toxicol Chem 14: [25] McCarty JP, Secord AL Reproductive ecology of tree swallows (Tachychineta bicolor) with high levels of polychlorinated biphenyl contamination. Environ Toxicol Chem 18: [26] Echols KR, Tillitt DE, Nichols JW, Secord AL, McCarty JP Accumulation of PCB congeners in nestling tree swallows (Tachycineta bicolor) on the Hudson River, New York. Environ Sci Tech 38: [27] Custer TW, Custer CM, Hines RK Dioxins and congener specific polychlorinated biphenyls in three avian species from Wisconsin River, Wisconsin. Environ Pollut 119: [28] Custer CM, Custer TW, Dummer PM, Munney KL Exposure and effects of chemical contaminants on tree swallows nesting along the Housatonic River, Berkshire County, Massachusetts, USA, Environ Toxicol Chem 22: [29] Custer CM, Gray R, Custer TW Effects of egg order on organic and inorganic element concentrations and egg characteristics in tree swallow (Tachycineta bicolor). Environ Toxicol Chem 29: [30] Custer TW, Gray BR, Custer CM. 2010a. Polychlorinated biphenyls, dioxins, furans, and organochlorine pesticides in spotted sandpiper eggs from the upper Hudson River basin, New York. Ecotoxicology 19: [31] Custer TW, Custer CM, Gray BR. 2010b. Polychlorinated biphenyls, dioxins, furans, and organochlorine pesticides in belted kingfisher eggs from Upper Hudson River Basin, New York, USA. Environ Toxicol Chem 29: [32] Karchner SI, Franks DG, Kennedy SW, Hahn ME The molecular basis for differential dioxin sensitivity in birds: Role of the aryl hydrocarbon receptor. Proc Natl Acad Sci USA 103: [33] Dean KM, Baltos LD, Carro T, Iwaniuk AN, Bohannon MEB, Ottinger MA Comparison of vehicle mortality following in ovo exposure of Japanese quail (Coturnix japonica) eggs to corn oil, triolein and a fatty acid mix. Endocrine Disruptors 4. [cited 2017 November 6]. Available from: [34] Carro T, Taneyhill LA, Ottinger MA The effects of an environmentally relevant 58 congener polychlorinated biphenyl (PCB) mixture on cardiac development in the chick embryo. Environ Toxicol Chem 32: [35] Carro T, Dean K, Ottinger MA Effects of an environmentally relevant polychlorinated biphenyl (PCB) mixture on embryonic survival and cardiac development in the domestic chicken. Envir Toxicol Chem 32: [36] Hill EF, Hoffman DJ Avian models for toxicity testing. J Am Coll Toxicol 3:357. [37] Lin F, Wu J, Abdelnabi MA, Ottinger MA, Giusti MM Effects of dose and glycosylation on the transfer of genistein into the eggs of Japanese quail (Coturnix japonica). J Agric Food Chem 52: [38] Ottinger MA, Wu JM, Hazelton JL, Abdelnabi MA, Thompson N, Quinn MJ Jr, Donoghue D, Schenk F, Ruscio M, Beavers J, Jaber M Assessing the consequences of the pesticide methoxychlor: Neuroendocrine and behavioral measures as indicators of biological impact of an estrogenic environmental chemical. Brain Res Bull 65: [39] Rufer ES, Hacker TA, Flentke GR, Drake VJ, Brody MJ, Lough J, Smith SM Altered cardiac function and ventricular septal defect in avian embryos exposed to low-dose trichloroethylene. Toxicol Sci 113: C 2017 SETAC wileyonlinelibrary.com/etc