Environmental Pollution 195 (2014) 109e114

Contents lists available at ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

Incubation stage and polychlorinated biphenyl (PCB) congener patterns in an altricial and precocial bird species Christine M. Custer a, *, Thomas W. Custer a, Stefan Thyen b, Peter H. Becker c a

USGS, Upper Midwest Environmental Sciences Center, 2630 Fanta Reed Rd., La Crosse, WI 54603, USA Schlossstrasse 30, D-53115 Bonn, Germany c Institut für Vogelforschung “Vogelwarte Helgoland”, An der Vogelwarte 21, D-26386 Wilhelmshaven, Germany b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 8 May 2014 Received in revised form 8 July 2014 Accepted 11 August 2014 Available online 13 September 2014

The composition of polychlorinated biphenyl (PCB) congeners was compared between non-incubated and embryonated eggs of tree swallows (Tachycineta bicolor) and little terns (Sterna albifrons) to determine if measurable changes in PCB congeners occurred during the embryonic period. There was no indication of changes in PCB congener patterns over the incubation period in tree swallows in 1999 and 2000 at a site with very high PCB exposure or a site with more modest PCB exposure. Additionally, congeners known to be either quickly metabolized or conserved based on experimental studies did not generally respond as predicted. Similarly, PCB congener patterns in eggs of little terns from Bottsand, Schleswig-Holstein, Germany, did not differ between non-incubated and embryonated eggs. The results from both species suggest that the stage of incubation is not an important consideration when evaluating PCB congener patterns; comparisons and assessments can be made with eggs collected at all stages of incubation. Published by Elsevier Ltd.

Keywords: Little tern PCB congener profile PCBs Polychlorinated biphenyls Tree swallow

1. Introduction Polychlorinated biphenyl (PCB) congeners can be grouped, based on chemical structure, to indicate degree of possible metabolism or biotransformation. Metabolism of PCB congeners is determined by the presence or absence of chlorine atoms at specific locations on the two phenyl rings (Borlakoglu et al., 1990; Frame, 1997). For example, congeners that have chlorine atoms on two neighboring carbon atoms, in both the ortho-meta and meta-para positions (e.g., congener 153 and 180), undergo little metabolism. Other congeners, which include many of the Ah-active congeners, 77, 114, 118, 126, and 156, have vicinal chlorine atoms in the meta-para positions, but not in the ortho-meta positions, and are more readily metabolized. Congeners that have opposite positioning with the chlorine atoms in the ortho-meta positions are also more readily metabolized. Congeners with three or fewer chlorine atoms are usually rapidly metabolized unless both para positions contain chlorine atoms, which reduces metabolism (Borlakoglu et al., 1988, 1990).

* Corresponding author. E-mail addresses: [email protected] (C.M. Custer), [email protected] (T.W. Custer), [email protected] (S. Thyen), [email protected] (P.H. Becker). http://dx.doi.org/10.1016/j.envpol.2014.08.010 0269-7491/Published by Elsevier Ltd.

Birds, both adult and juveniles, are able to clear PCBs from their bodies with the clearance rate varying by congener (deFreitus and Norstrom, 1974; Drouillard et al., 2001, 2007). Part of the elimination or change may be from metabolism or biotransformation, but excretion may also play a role. Data from field studies which began in the 1970s (Borlakoglu et al., 1988; Borlakoglu and Walker, 1989) corroborated these experimental results. Metabolic processes are occurring in avian embryos and some of those processes are disrupted (enhanced or reduced) when exposed to contaminants. For example, basal plasma corticosterone concentrations in herring gull (Larus argentatus) embryos were negatively associated with PCBs and polychlorinated dibenzo-p-dioxin and polychlorinated dibenzofuran (PCDDs/PCDFs) concentrations, probably as a result of the disruption of the hypothalamo-pituitary-adrenal axis (Lorenzen et al., 1999). Two of four intermediary metabolic enzymes, however, did not differ between sites with different exposure levels. Phosphoenolpyruvate carboxykinase activity showed a negative association with contaminant exposure; however, only malic enzyme activity differed from the reference site at 3 of the 4 contaminated locations. This decline in activity was related to PCDD/PCDF exposure, not PCBs (Lorenzen et al., 1999). Antioxidant defenses in embryos can also be activated upon exposure to PCBs, however there are species differences (Jin et al., 2001). For instance 7-ethoxyresorufin-0-deethylase (EROD) activity and thiobarbituric

110

C.M. Custer et al. / Environmental Pollution 195 (2014) 109e114

acid reactive substances were induced in chickens (Gallus domesticus), but not Pekin ducks (Anas platyrhynchos), when exposed to PCB 126 via egg injection (Jin et al., 2001). We do not know, however, whether these processes transform or degrade PCB congeners, and none of these papers addressed the issue of changes in congener composition during the incubation period. By logistical necessity, most field studies that involve egg collections for contaminant analysis collect eggs at different stages of incubation, so it is important to know if stage of incubation is an important factor that needs to be controlled or accounted for. Now that all PCB congeners are more frequently quantified, rather than just a small subset of 10e15 specific congeners, it is more important now than in the past to understand whether changes in the PCB congener profile is occurring in ovo. Although theoretically possible that the PCB congener profile in avian embryos changes because of physiological responses to contaminant exposure as illustrated above, there is a dearth of empirical evidence, for or against embryonic metabolism of PCB congeners, although it has frequently been postulated. For example, Thyen et al. (2000) suggested that lower chlorinated PCB congeners may be metabolized during embryogenesis of little tern (Sterna albifrons) eggs; however, their results do not strongly support this conclusion. Custer et al. (1997) did not see evidence for a change in composition pattern of 13 PCB congeners between sibling fresh and full-term double-crested cormorant (Phalacrocorax auritus) eggs; however, the tri-chloro PCBs, those that might be rapidly metabolized, were not represented in their sample. There is no information on changes in PCB congener profile during the embryonic period in tree swallows. Embryonic metabolism may not be as prominent as thought because liver tissue is not present or entirely functional for a large proportion of the embryonic period. Also, for some altricial species, such as the tree swallow (Tachycineta bicolor), the incubation period is quite short (e.g., ~12e13 days in the case of tree swallows [Robertson et al., 1992]) further reducing the possibility of seeing a change in the profile of PCB congeners. Our objective was to use tree swallow data from the Housatonic River, MA (Custer et al., 2003) and to reanalyze little tern data from Bottsand, Schleswig-Holstein, Germany (Thyen et al., 2000) to determine whether there was any evidence for changes in the PCB congener pattern across the incubation period. The little tern is a precocial species with a longer incubation period (23 days [Ratcliffe et al., 2008]), so it offered a contrast to the altricial tree swallow. Our approach was to examine differences in PCB congeners between fresh and embryonated eggs in tree swallows and in little terns using multivariate analyses similar to the methods of Custer et al. (1997). Additionally, for tree swallows we looked for evidence for congener pattern changes for the nine Ah-active congeners and the eight rapidly metabolized congeners as identified experimentally by Drouillard et al. (2007). Tree swallows were chosen as a model species because of the large amount of PCB data available and its wide use in avian ecotoxicology studies (Custer, 2011). The little tern data had not been evaluated using a multivariate approach (Thyen et al., 2000). This study is not a test of PCB metabolism per se, because metabolites of PCB congeners were not specifically quantified, but this approach examines whether there is sufficient change in congener composition in field samples to make this a concern for interpretation and use of field data. 2. Methods and materials For the tree swallows, in-depth field and analytical chemistry methods are presented in Custer et al. (2003) and Custer and Read (2006). In brief, tree swallow egg samples were collected from nest boxes located in the Housatonic River basin in Massachusetts in 1999 and 2000 near their expected hatching dates. Eggs were opened, the stage of incubation noted, the contents placed in chemically clean jars, and the jars frozen in a standard freezer at 20  C until the contents were chemically analyzed (within ~6 months of the date of collection). This storage time is

normal for field-collected samples. The contaminant analyses were done by Geochemical and Environmental Research Group, College Station, TX following standard methodology and quality assurance practices. Tissues were homogenized and extracted with Na2SO4 and methylene chloride and then purified by silica/ alumina column chromatography and high-performance liquid chromatography. Quantitative analyses were performed by capillary gas chromatography. Detection limits for PCB congeners were 0.2e0.4 ng/g wet weight. Blanks, duplicates, and spikes were run on at least 5% of total number of samples and met quality assurance standards. Reference material was analyzed concurrently, and 5% of samples were confirmed by gas chromatography/mass spectrometry. Specifics can be found in Custer et al. (2003). Details of the little tern field and analytical chemistry methods can be found in Thyen et al. (2000), but in brief samples were dried, purified with a silica gel column, and analyzed with gas chromatography/mass spectrometry. Detection limits were 0.23e0.68 ng/g wet weight. Congener numbers for PCBs follow the International Union of Pure and Applied Chemistry (IUPAC) numbering scheme. For the statistical analysis of tree swallows, a subset of egg samples, those that showed no development (n ¼ 11) and those that contained full term embryos (n ¼ 31), was selected. Although it would be desirable to compare non-incubated to full term eggs within the same clutch as Custer et al. (1997) did, those data were not available, so instead we used eggs with no development compared to eggs with full term embryos. Because the congener profile within a site is quite consistent (Custer and Read, 2006), we felt this approach provided an adequate test. There were two locations, Roaring Brook (42 2101800 N, 73130 2200 W) near Lenox, MA with 5 egg samples in the no development group and 18 in the full term embryo group, and Threemile Pond (42 080 4900 N, 73180 0500 W) near Great Barrington, MA, with 6 and 13 samples in those two groups (no development and full term embryo group). The two sites differed in contaminant exposure; Roaring Brook had very high PCB exposure (1999 geometric mean ¼ 100.9 mg/g wet wt. [94.0e108.3 ¼ 95% CI] and 2000 ¼ 67.5 mg/g wet wt. [61.3e74.5]) and Threemile Pond had much lower PCB contamination (6.0 mg/g wet wt. [5.02e7.28]) (Custer et al., 2003). These differences allowed us to assess whether the extent of exposure might affect our ability to detect changes in congener composition, if it was occurring. The two sites also differed in congener composition (Custer and Read, 2006). Because of the extremely high concentrations of PCBs at Roaring Brook and the elevated concentrations at Threemile Pond, there were few congener concentrations below the detection limit that might have confounded the statistical analyses. Comparisons were made on PCBs expressed on a wet weight basis because that is the normal basis used in most tree swallow studies (Custer, 2011). The eggs, regardless of developmental stage, had similar moisture content, 79.5% and 80.2% (no development and full term, P ¼ 0.79, t-test), indicating no loss of moisture. Because the statistical analyses were conducted on a percent that each congener comprised of total PCBs (see below), any moisture loss was immaterial. For the statistical analysis of little terns, we compared individual eggs with no development (n ¼ 8) with eggs with embryonic development (n ¼ 12). Each egg was from a different female. The eggs were collected from Bottsand, Schleswig-Holstein, Germany, during the period 1978e1996 (Thyen et al., 2000). Total PCB concentrations in fresh eggs averaged 7.7 mg/g dry weight (~2.4 mg/g wet wt., using 75% moisture content) (Thyen et al., 2000) and 6.6 mg/g dry weight (~1.9 mg/g wet wt.) in embryonated eggs. We used multivariate nonmetric multidimensional scaling plots (NMDS) and analysis of similarity (ANOSIM) analyses on (1) standardized data (% of total PCBs that each congener comprised) which tends to weight more heavily congeners which comprise a higher percentage of the total, and (2) normalized (mean ¼ 0 and standard deviation ¼ 1) concentration data that weights each congener between 2 and þ2 so that each congener is weighted approximately equally. For tree swallows, analyses were done on all 82 congeners that were detected in those samples, on nine Ah-active congeners (PCB 77, 81, 126, 105, 118, 156, 167, 169, and 189), and eight rapidly metabolized congeners (PCB 16/32, 26, 52, 44, 41/64, 84, 136, 95/80) as identified by Drouillard et al. (2007). For the Roaring Brook site we had two years of data so we nested stage of incubation within year. For Threemile Pond we only had one year (2000) so we did a 1-way ANOSIM. For the little tern data, analyses were done on 50 individual or combined congeners. A total of 62 congeners were analyzed, but 11 congeners coeluted with other congeners; PCB 169 was not detected in any sample and was not included in the multivariate analyses. The multivariate analyses followed the procedure outlined in Custer et al. (2010). In brief, if R is 0.4 there are clear differences in PCB patterns among the groups. There is some support for pattern differences when R is 0.3 to

Incubation stage and polychlorinated biphenyl (PCB) congener patterns in an altricial and precocial bird species.

The composition of polychlorinated biphenyl (PCB) congeners was compared between non-incubated and embryonated eggs of tree swallows (Tachycineta bico...
602KB Sizes 0 Downloads 4 Views