Clemson University
TigerPrints
All Dissertations
Dissertations
12-2012
Spider Mediation of Polychlorinated Biphenyl
Transport and Transformation Across Riparian
Ecotones
Diana Delach
Clemson University, [email protected]
Follow this and additional works at: http://tigerprints.clemson.edu/all_dissertations
Part of the Environmental Sciences Commons
Recommended Citation
Delach, Diana, "Spider Mediation of Polychlorinated Biphenyl Transport and Transformation Across Riparian Ecotones" (2012). All
Dissertations. Paper 1051.
This Dissertation is brought to you for free and open access by the Dissertations at TigerPrints. It has been accepted for inclusion in All Dissertations by
an authorized administrator of TigerPrints. For more information, please contact [email protected].
SPIDER MEDIATION OF POLYCHLORINATED BIPHENYL TRANSPORT AND TRANSFORMATION ACROSS RIPARIAN ECOTONES A Dissertation Presented to the Graduate School of Clemson University In Partial Fulfillment of the Requirements for the Degree Doctors of Philosophy Environmental Toxicology by Diana Delach December 2012 Accepted by: Dr. Cindy M. Lee, Committee Chair Dr. John T. Coates Dr. Peter van den Hurk Dr. David M. Walters ABSTRACT Polychlorinated biphenyls (PCBs) contaminate the sediment of the Twelvemile Creek / Lake Hartwell Superfund Site, and are known to be transported throughout the resident biota via trophic transport. Riparian spiders have recently become of interest because they are terrestrial organisms that have significant PCB exposures derived from aquatic sources. Many riparian spiders primarily consume insects emerging from contaminated aquatic systems, and these spiders can have a body burden as high as 2900 ng/g lipid. These emergent insects carry contaminants out of the river and into the riparian zone where they are captured by spiders, which effectively directs the contamination towards arachnivorous predators such as lizards, frogs, and birds. The enantiomeric fraction (EF) was measured for chiral congeners to investigate the role of biological systems on transport of PCBs between trophic levels. The EF values varied between spider species, and indicate that foraging behavior may influence those parameters. Tetragnathid and basilica spiders were most similar, whereas both were different from araneid spiders despite all three spiders belonging to the same order of spiders. All spider taxa were significantly different from the aquatic prey source Chironomidae. Two approaches were used to confirm that spiders have the capacity to metabolize their PCB body burdens. Tetragnathidae spiders were collected along Twelvemile Creek, their enzymes isolated, and exposed to individual non-­‐planar and co-­‐planar congeners. PCBs 88 and 149 were incubated with S9 fractions (extracts ii
containing microsomal and cytosolic enzymes) from the spiders and qualitatively assessed for evidence of biotransformation. Tandem mass spectroscopy provided evidence to support the hypothesis that spiders have the capacity to biotransform PCBs. Additionally, PCB 61 was incubated with S9 fractions for quantitative analysis of a planar congener. Numerous compounds were detected after exposure, but OH-­‐
PCB 61 was measured at 1.63 (±0.35 SD) ng/g lipid at the Reese Mill sampling site for enzymes obtained with liquid nitrogen, thus indicating that spiders have the capacity to metabolize their PCB body burden. In the second approach mass spectroscopy of whole spider extracts of spiders obtained along the Twelvemile Creek arm of Lake Hartwell provided structural evidence that spiders can transform their body burden of PCBs to OH-­‐PCBs for congeners with six or fewer chlorines. Lastly, webs are hypothesized to play a protective role in spider ecotoxicology. Tetragnathid spiders are able to recycle approximately 90% of their web material without metabolizing it, thus creating an opportunity for web material to act as a storage location external to the body. Concentrations in webs ranged between 154 and 356 ppm, whereas concentrations for spiders at the same sampling locations ranged from 284 ng/g lipid to 2900 ng/g lipid. The enantiomeric fraction was also utilized to determine if storage in webs is an enantioselective process. Results indicate that web storage is enantioselective for PCB 149, with the (-­‐) enantiomer being preferentially retained in web materials. This differs from that seen in spider samples, where the EF is approximately racemic. iii
These investigations examined the exposure and toxicological model for spiders, with the intent of aiding understanding the role spiders play in mediating transport and transformation across riparian ecotones. Results indicate that spiders may use a variety of strategies to manage their PCB body burdens ranging from enantioselective uptake of parent compounds, metabolism to hydroxylated metabolites, and transfer to web material. Understanding spider mediation of PCB transport and transformation can help development of strategies that both manage and mitigate the risks posed to the environment by PCBs at Twelvemile Creek and Lake Hartwell. iv
DEDICATION I would be remiss if I didn’t note the contribution on the part of Henry, Barbara, and Alyssa Delach to this body of work. Their emails, phone calls, and cards were a constant source of encouragement and support, without which the completion of this research would have been far more taxing. They provided comic relief when it was most necessary! v
ACKNOWLEDGEMENTS First I must acknowledge the contribution Dr. Cindy Lee has made to this work. She took on not only the role of research advisor, but also mentor for myself, my labmates, and many others across the graduate school. I hope that my future endeavors will serve to support your already superior reputation. My committee members had a great impact on this project as well, being sure to point out not only the points that needed deletion, repeating, or revising, but also the successes along the way. I learned much from their expertise and kindness. My labmates– for teaching me everything from how to order supplies to sample preparation to winning strategies in the faculty-­‐student soccer games. The rewards of working with such fun, intelligent people cannot be underestimated. In particular, Viet Dang’s experience and patient teaching made this work possible. I hope that we all have walked away with at least some good stories from our time spent constructing spider habitats, prowling the river at night for bugs, and spending long days/evenings/nights in the lab together. My departmental colleagues in both Environmental Toxicology and Environmental Engineering have also enriched this work. Your support in and out of the lab has served to enhance my research experiences. In particular, thank you to Andrea Hicks, Tim Sattler, Jessica Dahle, Lee Stevens, Yogendra Kanitkar, Kay Millerick, and Meric Selbes. Anne Cummings’ dedication to maintaining smooth operation in the labs and with the instruments facilitated the completion of this work, as well as the work of the other graduate students in the Rich Laboratories. vi
TABLE OF CONTENTS Page TITLE PAGE……………………………………………………………………………………………….. i ABSTRACT...………………………………………………………………………………………………. ii DEDICATION………………………………………………………………………………………………. v ACKNOWLEDGMENTS……………………………………………………………………….……….. vi LIST OF TABLES………………………………………………………………………………………… ix LIST OF FIGURES………………………………………………………………………..………………. x CHAPTER 1. Literature Review Introduction…………………………………………………………..………………………... 1 Compound background………………………………………………………………........ 3 Study area…………………………………………...…………………………………….…….. 4 Arachnid role at Twelvemile Creek / Lake Hartwell Superfund site……..………………………………………………………………… 7 Ecological role of spiders………………………………………...……………………...… 8 The Spider Family Araneidae…………..………………………….……………..….......12 Exposure…………………………………………………………..…………………….............. 15 Excretion…………………………………………………………………….…………………… 19 Webs…………………………………………………………………..…………………………… 21 PCB Transformation ……………………………………..……………………………......... 25 PCB Metabolism by Spiders…………………………………………………………….... 32 Summary……………………………………………..…………………………………………... 33 References………………………………………………………………………………...…….. 34 2. Chiral Signatures of PCBs in Riparian Spiders Introduction…………………………………………………………………………………….. 47 Materials and Methods……………………………………………………………………... 50 Results…………………………………………………………………………………………….. 54 Discussion…………………..…………………………………………………………………… 60 Conclusions……………………………………………………………………………………… 64 References……………………………………………………………………………………….. 65 vii
Page 3. Spider Metabolism of PCBs: in vivo and in vitro Investigative Approaches Introduction……………………………………………………………………………………..68 Materials and Methods…………………………………………………………………….. 70 Results…………………………………………………………………………………………….. 78 Discussion……………………………………………………………………………………….. 90 Conclusions……………………………………………………………………………………… 99 References……………………………………………………………………………………….. 100 4. Webs as Substrate for PCB Partitioning Introduction…………………………………………………………………………………….. 106 Materials and Methods……………………………………………………………………... 107 Results…………………………………………………………………………………………….. 111 Discussion……………………………………………………………………….………………. 119 Conclusions……………………………………………………………………………………… 125 References…………………………………………………………………………………..…… 126 5. Conclusions and Recommendations Conclusions…………………………………………………………………………………….. 129 Recommendations for future work…………………………………………………… 130 APPENDICES A: Supplemental Figures for Chapter 2……………………………………………... 134 B: Supplemental Tables for Chapter 2………………………………………………. 136 C: Supplemental Table for Chapter 3…………………………………………..…….. 137
viii
LIST OF TABLES Table Page 1.1 Comparison of spider species ecological roles………………………… 12 2.1 GPS coordinates for EPA sampling sites………………………………….. 52 4.1 Web sampling site locations and distances…...………………………… 108 4.2 ΣPCB content in spiders and webs at each sampling location………………………………………………….…….. 112 4.3 Homolog distributions in spiders and webs at the Reese Mill sampling site………………………………………… 113 B.1 Environmentally stable chiral congener substitution patterns ………………………………………………….. 136 B.2 Spider pooled-­‐sample sizes for each sampling location……………………………………………………………………….136 C.1 SIM paramenters used to screen for PCB metabolites……………… 137
ix
LIST OF FIGURES Figure Page 1.1
1.2 1.3 1.4 1.5 1.6 2.1 2.2 2.3 Map of Twelvemile Creek / Lake Hartwell Superfund Site………..…………………………………………………… 6 Tetragnathidae exposure model…………………………………………….. 9 Spider anatomy……………………………………………………………………... 20 Different types and functions of silks produced by orb weaver spiders…..…………………………………………….. 23 Metabolism of PCBs…………………………………………………………..…… 27 Thyroid hormone regulation………………………………………………….. 31 Map of sampling locations……………………………………………………… 51 Measurements of EF for three spider species and Chironomidae averaged across all analyzed sites…..………..………………………………………………...................... 56 NMS plots for both spiders and midges…………………………………... 59 3.1 3.2 3.3 Map of sampling locations……………………………………………………… 73 Tetragnathidae in vivo analysis, SIM settings for m/z = 324………………………….……………….. 80 Tetragnathidae in vivo analysis, SIM settings for m/z = 353……..……..…………………………….. 81 3.4 3.5 Chromatogram for PCB 88 enzyme incubations……..……………….. 83 Chromatograms for enzymes with liquid nitrogen and incubated with PCB 88 at Reese Mill and Robinson Bridge…………….………………………………..…… 85 x
Figure Page 3.6 Chromatograms and spectra for enzyme incubation with PCBs 88 and 149 at Robinson Bridge……………………. 86 3.7 Chromatogram for positive control………………………………………… 87 3.8 Chromatogram for incubation with PCB 61…………………………….. 88 3.9 Quantification of 2-­‐OH-­‐PCB 61 after incubation with S9 fraction…………………………………………………………... 89 3.10 PCB 88 and potential biotransformation pathways…………………. 92 3.11 PCB 149 and potential biotransformation pathways ………………. 94 3.12 PCB 61 and potential biotransformation pathways…………………. 95 4.1 Map of web sampling locations………………………………………………. 109 4.2 Σ PCB concentration for spiders and webs…………………………....... 112 4.3 Homolog patterns of PCB concentrations in spiders and webs...…………….......................................................... 114 4.4 EF values for spiders and webs……………………………………………..... 117 4.5 A.1 A.2 NMS plots for spiders and webs……………………………………………… 118 A.3 Basilica EF by site for all congeners…………………………………………135 A.4 Chironomidae EF by site for all congeners………………………………. 135 Tetragnathidae EF by site for all congeners…………………………….. 134 Araneidae EF by site for all congeners……………………………………. 134 xi
CHAPTER ONE LITERATURE REVIEW Introduction Polychlorinated biphenyls (PCBs) contaminate the sediment of the Twelvemile Creek / Lake Hartwell Superfund Site (US EPA, 2004), and are known to be transported throughout the resident biota via trophic transport (Walters et al., 2008). Riparian spiders have recently become of interest because they are terrestrial organisms that have significant PCB exposures derived from aquatic sources. Many riparian spiders primarily consume insects emerging from contaminated aquatic systems, and these spiders can have body burdens greater than 5000 ng/g lipid (Walters et al., 2010). These emergent insects carry contaminants out of the river and into the riparian zone where they are captured by spiders, which effectively directs the contamination towards arachnivorous predators such as lizards, frogs, and birds. A better understanding of how spiders function as biovectors will elucidate how exposure to PCBs is mediated as the contaminants cross ecotones, and can inform development of policies aimed at mitigating the risks from those contaminants to humans and wildlife. In particular, investigating PCB uptake, biotransformation, and storage by spiders can clarify understanding of PCB fate and transport in the riparian zone at Twelvemile Creek / Lake Hartwell., as well as other 1
aquatic systems contaminated with persistent, bioaccumulative pollutants, to inform remediation strategy and policy decisions. Analysis of chiral congeners can indicate biotransformation of PCBs in the food web. The enantiomeric fraction (EF) is the parameter that is commonly used to quantify chiral character (See Chapter 2 for a detailed explanation of how the EF is calculated). The EF can change when the fate and transport parameters change. One process that can change EF is metabolism, which has been indicated as a source for enantiomeric enrichment in some species; for example, Buckman et al. (2006) determined that metabolism in fish produced OH-­‐PCBs. Biochemical assays can also help to elucidate any metabolic pathways that may exist. If enzymatic activity in PCB-­‐exposed spider can be quantified, an increase in activity may indicate metabolism of PCBs. Similarly, enzymes isolated from spider tissue and incubated with PCBs may yield metabolites. If these metabolites can be quantified and their structure confirmed, this, too, can provide the first evidence of spider detoxification capabilities. There may be other mechanisms by which spiders regulate an increasing body burden. Spider webs may function as an outer-­‐body tissue, providing a place to sequester contaminants from sensitive organs. Comparing the PCB content and congener patterns of web material can begin to describe molecular toxin management strategy by spiders. 2
Compound Background Polychlorinated biphenyls are synthetic compounds with a varying number of chlorines oriented about two biphenyl rings. There are 209 congeners that can be sorted into different homolog groups, ranging from one chlorine (mono) to ten chlorine atoms (deca) oriented about the biphenyl rings. Each of the homolog groups exhibits different fate and transport trends, as the physiochemical properties are affected by both the chlorine content and orientation. For example, PCB 2, with an ortho-­‐situated chlorine has a log Ciw of -­‐4.54; whereas, PCB 4 has a para-­‐chlorine and a log Ciw value of -­‐5.19, where Ciw is the solubility of the contaminant in water (Schwarzenbach et al., 2003). While chemical composition is identical, the change in substitution pattern between o-­‐Cl to p-­‐Cl decreases the solubility in water. As chlorine content increases, the tendency to partition into lipophilic media increases. All PCBs have low aqueous solubility, high thermal stability, and excellent dielectric properties. Their prolific use led to significant exposure in the environment. The stability that makes PCBs suitable in industrial processes is the same that make them legacy compounds in the environment that experience both long-­‐range transport and large-­‐scale distribution (Tanabe et al., 1983). Production of PCBs began in the 1920s and continued in the United States until public and scientific concern ceased their manufacture in the 1970s. Fifty years of production yielded approximately 150,000 metric tons of PCBs (deVoogt and Brinkman, 1989). Discovery of their human health hazards led to the ban of their use in 1976 under the Toxic Substances Control Act (U.S. EPA, 1976). PCBs 3
have been detected in all environmental compartments, and matrices ranging from sediment to fish tissue to human blood. Trophic transfer is one of the main routes of PCB transport, as their high lipophilicity creates a tendency to both bioaccumulate and biomagnify through the food chain. PCB congeners can cause a variety of physiological problems, ranging from thymus, spleen, and liver hypertrophy, as well as increased liver lipids (Safe, 1990). They may also be responsible for alteration of Ca2+ homeostasis in neuronal transmission (Kodavanti et al., 1995). There is evidence that ortho-­‐substituted PCBs can influence dopamine concentrations via inhibition of the tyrosine hydroxylase (Seegal et al., 1990; Shain et al., 1991). Study Area The Sangamo-­‐Weston Corporation manufactured capacitors using PCBs as dielectric fluid in Pickens, SC, between 1955 and 1977 (Figure 1.1) (Brenner et al., 2004). Production using PCBs was ceased because of the impending 1976 ban on its use in the United States, as it had been linked to cancer and other human health risks. Improper disposal of production waste led to the contamination of the sediment in Town Creek, a tributary of Twelvemile Creek and then Lake Hartwell. A fish advisory was established by the EPA in 1976, as the concentrations of PCBs in fish greatly exceeded the limit set by the Food and Drug Administration for commercial seafood (U.S. EPA, 1980; SCDHEC, 2006). The Lake Hartwell / Twelvemile Creek site was placed on the National Priority List in 1991, and was later named a Superfund site. The Record of Decision 4
designated two operational units: OU-­‐1 addressed the land sources of contamination, namely the Sangamo-­‐Weston plant and six satellite disposal areas, and OU-­‐2 addressed the pollution starting with sediment and surface water contamination up to fish and other contaminated biota (U.S. EPA, 1994). OU-­‐2 includes aquatic habitats in Town Creek, Twelvemile Creek, and Lake Hartwell (Figure 1.1). 5
Figure 1.1 Map of the Twelvemile Creek / Lake Hartwell Superfund site. Adapted from Brenner et al. (2004). 6
Arachnid Role at Twelvemile Creek / Lake Hartwell Superfund Site Recent efforts to monitor the site have determined that the contamination is moving from the aquatic compartment to the terrestrial compartment of the landscape (Walters et al., 2008, 2010). Contaminated sediments provide habitat space for larval aquatic insects. Upon maturation, these invertebrate metamorphose into winged adults capable of transporting allochthonous carbon, nitrogen, and PCBs from the contaminated sediment to the neighboring terrestrial system. Such insects are known as subsidy insects. These emergent insects then become prey for terrestrial predators. Walters et al. (2010) calculated that avian wildlife in the Twelvemile Creek watershed, both resident and transient, exhibit PCB concentrations that exceed threshold limits for negative effects. Spiders are of particular interest, as they consume primarily insects, and some families can be considered to be interface specialists, consuming only aquatic species. Arachnivorous predators, such as lizards, amphibians, and birds, consume spiders. It is believed that spider-­‐mediated transport across the ecotone is a dominant route into the terrestrial compartment (Walters et al., 2008, 2010; Raikow et al., 2011), though the mechanisms behind this transport require further investigation. To date, biochemical pathways of contaminant metabolism in spiders have yet to be described in the literature by either entomologists or biochemists. Their ecological function has been described, as has their physiology and diet, but there are lingering questions about their ability to detoxify and metabolize toxicant body 7
burden. Characterization of present metabolites via tandem mass spectroscopy, or identification of enzymatic pathways using biomarker suites may contribute to a more complete understanding of how this process may take place. Ecological Role of Spiders Understanding the ecological roles fulfilled by spiders provides insight into how they may acquire contaminants, how their body burden may be transformed, or how they facilitate transport of PCBs to other trophic levels. Spiders fulfill a variety of roles dependent upon where they reside in the ecosystem. Spider families and species are typically defined by their predation type (Foelix 1982), but will also be classified by their location within the ecosystem for the purposes of this review. The foraging habits of spiders directly influence their exposure. Riparian spiders prey upon aquatic insects, and consume only soft tissues, so capturing the insects prior to the hardening of the exoskeleton results in more food for spiders without exerting extra energy to consume more prey. When prey is unavailable and stored energy (lipids) are utilized the concentrations of contaminants will be affected. Mobilization of the lipid tissue re-­‐exposes the spider to the contaminants, which may be an important route of exposure due to the “wait and see” foraging strategy. The slow metabolism of spiders coupled with a stationary predation technique allows the spider to live for long periods of time without food. Spiders are consumed by a variety of predators, including but not limited to arachnivorous birds such as the Carolina wren, lizards, and even other spiders (Figure 1.2). The 8
role of webs in toxicology of spiders has not been previously investigated, but further discussion of relevant hypotheses is included under the “Excretion” heading below. Figure 1.2 Tetragnathidae exposure model. Most tetragnathids are obligate riparian species and specialized predators of aquatic insects. The competitive nature of spiders has been much debated. Tretzel (1955) developed a theory of “intragenerische isolation,” or niche partitioning within 9
genera. He postulated that differences in reproductive seasons, daily patterns of activity, and horizontal habitat utilization are the result of evolved differences. This concept builds upon the work of Bristowe (1939, 1941), who stated that different habitats house different species because they can provide a variety of niches, resulting in species that specialize by insect size, insect temporal patterns of activity, and insect modes of transportation. This has been expanded upon by Łuczak (1963) who found that as the weather patterns changed the niches expanded and contracted accordingly, resulting in a shifted pattern of community structure. This interspecific competition theory has been widely accepted. It explains the evolutionary activity that gave rise to the differences between the three spider taxa examined in this dissertation: Tetragnathidae, Araneidae, and basilica. Tetragnathid spiders are specialists of aquatic insects, so they are an ideal model taxa for investigating aquatic-­‐to-­‐terrestrial exposure and for making comparisons to other spiders with different predation tactics. Spiders have evolved under food-­‐limited conditions (Foelix, 1982). Generalist feeding patterns are reflective of this, as are their low metabolic rates (Carrel and Heathcote, 1976), and food storage capacity within the digestive system (Foelix, 1982). As sit-­‐and-­‐wait foragers, spiders have evolved to expend little energy seeking prey, and to most efficiently utilize the food intake they do have. Spiders have evolved to fast for long periods of time, and are disinclined to change foraging behavior due to increases in prey availability, (Greenstone, 1979). Terrestrial spiders have lived up to 200 days without feeding (Kaston, 1970). 10
Understanding of placement of spiders in the food web at the Twelvemile Creek / Lake Hartwell Superfund site has been elucidated in numerous ways. First, an assessment of the prey captured by webs indicates that spiders predate on aquatic insects (Walters et al. 2008; personal observation). The variation in insect contribution has been confirmed by carbon and nitrogen isotope ratios due to the fact that gut contents are not applicable due to their external digestion (Akamatzu et al., 2004). Studies also demonstrate that emergent insects constitute a large proportion of the spider diet (Walters et al., 2008). Walters et al. (2010) determined that the PCB concentration patterns of chironomids and other emergent insects did not correlate well with those seen in spiders. The poor correlation could be due to the fact that the insects move too fluidly up and down river to be good indicator species at each sampling site; or, they may not represent a large enough proportion of the spider’s diet to have nearly identical patterns. Spiders have not been the subjects of toxicology tests to date; rather the toxins they produce have been of more interest. However, when spiders are exposed to contamination through the food web they become a part of the ecotoxicological processes of the ecosystem. Trophic transfer of contaminants will expose the spiders to both parent compounds and their metabolites. When exposure of invertebrates can be defined qualitatively and quantitatively we can use that information to investigate the biovectors of the system and to identify remedy options to manage the risks of contaminants to wildlife. 11
The Spider Order Araneae Many spiders are found in the interior of forest patches, spinning webs in the various trees and shrubbery (Table 1.1). The three spider types investigated herein belong to the same order, Araneae. One of the investigated families of spiders is Araneidae, which ranges from riparian ecotones to upland terrestrial areas. They are solitary spiders that spin orb shaped webs, which are maintained on a near daily basis (Eisoldt et al., 2011, 2012). Araneid spiders will consume the web from the previous day and reconstruct a new assembly. The angle of the web is vertical, which allows for the capture of flying terrestrial insects. Araneidae spiders emerge in the spring and are active through late fall (Foelix, 1982). During the winter they retreat underground where they suspend function until warmer temperatures facilitate motility and predation. Table 1.1 Comparison of Spider Ecological Roles Feature Habitat Location Solitary or communal Web orientation Web type Web maintenance Invertebrate types in diet Araneidae Tetragnathidae Facultative riparian Obligate riparian Solitary Solitary Horizontal Vertical Orb Orb Individual, nightly Individual, nightly Terrestrial, or mix of Aquatic terrestrial & aquatic Basilica Shoreline shrubbery Communal Messy Orb Communal over year Mix of terrestrial & aquatic Tetragnathidae, which is the main focus of these investigations, is a family separate from that of Araneidae, however they both fall in the same order, Araneae. Tetragnathids are obligate riparian predators, and as such are important terrestrial 12
predators of aquatic emergent insects (Baxter et al., 2005). This has been confirmed at the Twelvemile Creek Superfund site through use of stable isotope analysis (Walters et al., 2008). A variety of processes contribute to riparian spider exposure, including but not limited to contaminant uptake by benthic, larval insects, maturation, and predator-­‐prey interactions (Walters et al., 2010). Tetragnathidae spiders begin to emerge in March and are ubiquitous along Twelvemile Creek until about September. They will feed opportunistically upon any insects that entangle themselves in their orb-­‐shaped webs. Their web structure and proximity secures them in this riparian niche (Gillespie, 1987). Their webs are angled so as to capture the insects as they emerge from the water column as adults, but also are spun with silk that is likely too weak to capture terrestrial insects (Foelix, 1982). To obtain tangled insects the spider travels across the web, subdues the prey by wrapping it in silk, and then moves it with the checliare (the outer jaws) to the outer part of the web where it consumes the insect. Wildlife values to assess risk were calculated to gauge the importance of tetragnathids in the diets of arachnivorous birds (Walters et al., 2008, 2010). The analysis indicated that the spiders are a viable and important pathway of contaminant transfer to birds at the Twelvemile Creek / Lake Hartwell Superfund site. Patterns in activity are distinctive and correlate with habitat type. Tetragnathids have been able to survive in many habitat types, but they thrive near open areas of water with many branches and twigs. This habitat type may be present in sheltered stream areas, or open lake areas. Gillespie and Caraco (1987) 13
found that spiders around lakes were more prone to relocating webs, whereas spiders along streams were more stationary. They attributed these patterns to differences in risk; stationary behavior is indicative of low risk, enhanced by the amount of vegetation presenting hiding places from arachnivorous predators. Exposed lake areas also do not have the same amount of shade, which results in higher temperatures and faster web desiccation, resulting in higher energy expenditures for the spiders to maintain the webs. Basilica spiders (Mecynogea lemniscata) belong to the Araneidae family, but have some unique characteristics distinct from other family members. Basilica weave communal webs that are maintained over the course of the year (Foelix, 1982). These webs are usually constructed in shoreline shrubbery to allow predation of both aquatic and terrestrial prey. Over the course of the seasons, the web becomes messier and messier as the commune continually patches any damage done to the web procuring prey. These spiders may exhibit a different contamination profile due to their varied prey sources. Basilica spiders are similar to Tetragnathidae spiders in that their diet is largely constituted of aquatic prey due to their proximity and specialization to the riparian ecotone. PCBs at Twelvemile Creek are transfered between environmental compartments due to biotransportation by emerged aquatic insects, and are deposited across the ecotone by these mechanisms. The net flux is out of the aquatic compartment and into the terrestrial; however, there is also some transfer in the opposite direction when spiders die and fall from their webs back into the river. 14
While the transfer to the terrestrial ecosystems may reduce the PCB load in Twelve Mile Creek over time, the problem of PCB contamination is not so much solved as it is distributed and diluted over a larger area. These deposits can return to the river in time if they percolate into the ground water, or physical forces such as erosion reintroduce contaminated soil and plant material to the river. An alternate transport pathway may see PCB toxicity expressed in the terrestrial system by oligochetes and other organisms that rework the soil, plant matter, and spider litter. Exposure Riparian spiders are exposed to PCB contamination at the Twelvemile Creek / Lake Hartwell site via trophic transfer. Spiders consume contaminated emergent insects and redirect contaminant cycling away from the aquatic zone. Biological processes such as selective protein binding, tissue transport, and metabolism can affect and change the contamination profile to which spiders are exposed. For example, these processes can produce enantiomeric enrichment of chiral congeners, which results in spiders having different EF values than observed in their prey. In vivo and in vitro studies can be completed to more fully understand the mechanisms and effects of a given compound. Such studies must be done so as to complement each other, too, rather than just one or the other. In vivo assays may yield information about the biotransformation rate, whereas in vitro parameters can more fully describe the enzymatic activity resulting from field exposure (U.S. EPA, 15
2004). Conducting these studies in tandem allows for collection of complementary information. Life stage contributes to the varying accumulation of contaminants in ecosystem components. Toxicants vary in their targets, but also when those targets are susceptible to exposure. Organisms are typically most sensitive to toxicant exposure when in early development stages, particularly during organogenesis when the organs develop while in utero, which is when metabolic rates are high, and all nutrients are utilized for either energy production or growth (Klassen, 2008). Once an organism reaches full maturity and the metabolism slows and growth stops the potential for preventative sequestration of toxicants into lipophilic tissues is greater, which protects the organism from toxicant action. A toxicant body burden is not inherently dangerous –contaminants pose a threat only when they interact with target molecules (Walker et al., 2006). It is protective when toxicants are sequestered in lipid tissue away from the target sites. The route of uptake for spiders at Twelvemile Creek, is trophic transfer rather than uptake from the environment (Walters et al., 2008). The contamination of the Lake Hartwell/Twelve Mile Creek Superfund site is confined to the sediments in the aquatic system. PCBs are also unlikely to partition into chitin, the material that makes up the exoskeleton of spiders, so exposure from contact with the terrestrial environment is negligible. Spiders initiate digestion outside of the body by regurgitating digestive fluid onto the prey. The spider then ingests a drop of the predigested fluid and repeats until the prey is mostly consumed. The chelicae are 16
used to break apart the prey such that the soft tissue may be accessed. After feeding, all that remains is a small ball of cuticular parts (Foelix, 1982). Metabolism continues in the sucking stomach. The aquatic insects they consume in this manner transfer contamination from the sediments of the river system to the spiders. Due to the exposure through trophic transfer, it is important to first determine the exposure on the part of those emergent insects. Their life cycle places them in the sediments for the first few instars of their development, which results in high levels of exposure to lipophilic contaminants that have sorbed onto the sediment during the most susceptible time of their development. They pump oxygenated water across their gills to breathe. They may be exposed to contaminants in the water column this way, but can also export contamination into the water column with each “exhale.” We can infer that the biotic contaminant concentration lies somewhere between what would be predicted by equilibrium with the water and equilibrium with the sediment; typically greater than would be predicted by water exposure, but less than sediment exposure predictions. A theoretical concentration at equilibrium for PCBs needs to relate organic carbon content and octanol partitioning to each other, which, for PCBs, results in the following relationship: 𝐾!"!#$% = 2.3 × 𝐾!" !.!" (Equation 1) where Kilipoc is the partitioning coefficient for lipids and organic carbon, and Kow is the partitioning constant for octanol and water (Schwarzenbach et al., 2003). While these predictive equations for Kilipoc and Kilipw are useful, the measured quantity 17
known as the biota-­‐sediment bioaccumulation factor (BSAF) is defined in Equation 2 as: 𝐵𝑆𝐴𝐹! =
!!"#$
!!"
(Equation 2) where Ciorg is the concentration in the organism, and Cis is the concentration in the sediment. BSAF can be experimentally determined via field samples. BSAF values are seen to increase with increasing Kiow values for the contaminants, until the Kiow increases above a value of 107, because compounds with a log Kiow above 7 are usually too large and lipophilic to partition past the polar head groups in a membrane. Predictions for systems at equilibrium prove challenging enough; however, it must be noted that river systems are open systems, and there may be net flux of volatile contaminants in or out of the system. For PCBs this flux has been proven to be both highly temperature dependent and significant. It is believed that PCBs volatilize out of warm systems and are transported through the atmosphere until the temperatures are reduced such that net flux is into water. Such behavior has resulted in PCBs appearing in snow in Antarctica despite the absence of direct exposure on that continent (Tanabe et al., 1983). At the Twelvemile Creek / Lake Hartwell system, volatilization may be an important source of PCBs to tetragnathid webs, which are suspended directly over the water column. If such behavior is occurring, then spider exposure models need to incorporate this exposure pathway. 18
Excretion Excretion completes detoxification of toxicants. Lipophilic contaminants are removed from the organism via solid waste, whereas more hydrophilic contaminants are excreted with liquid waste. PCB excretion by spiders is complex in part because they only excrete solid waste. Water must be sourced from prey, so waste is typically concentrated and dehydrated in the cells where the diverticula of the midgut extends into the coaxe of the spider’s legs (see Figure 1.3 for anatomy), though some spiders have a developed midgut area that penetrates the area between the eyes (Foelix, 1982). Underlying interstitial tissue may also store waste as crystals once the capacity for storage is exhausted. Guanocytes also serve to store metabolites as crystals, and may form a cell layer under the hypodermis. The midgut widens to form the stercoral pocket, or cloacal chamber. On either side of the stercoral pocket are two thin evaginations called Malpighian tubes. Metabolites and substances to be removed from the body are concentrated and stored before they are expelled into the lumen of the Malpighian tubile. Guanine, adenine, hypoxanthine, and uric acid are the main excretory products, all of which are concentrated as crystals. The Malpighian tubules feed into the stercoral pocket, which connects to the anus. Excrement stored in the stercoral pocket periodically exits the body via the anus. Evolution of orb-­‐weavers like Tetragnathidae and Araneidae has evolved such that the coaxe of the walking legs no longer function to collect and then excrete waste through the coaxal gland duct (Foelix, 1982). It may be possible for the PCBs to be crystallized as part of this waste, and sequestered 19
away from the sites of action due to their low solubility in water. Spiders may also be able to further metabolize OH-­‐PCBs and excrete them as quinones or catechols. Figure 1.3 Spider Anatomy. Adapted from Foelix (1982). Exoskeleton molts are not expected to influence these concentrations or function as an excretion pathway due to their high chitin concentration. Chitin is a biotic macromolecule with functionality similar to both carbohydrates and proteins. They have high heteroatom content, ring structures, and many functional groups. The compound does not have much lipophilic character; thus, PCBs are unlikely to partition into the medium. Being more hydrophilic, the metabolites may partition into the exoskeleton, but this has yet to be investigated. Studies on other invertebrates found that when insects shed their exoskeletons the contaminant load is not reduced; rather, it is concentrated within the soft body tissue (Bartrons et al., 20
2007). When aquatic insects have just emerged, their exoskeleton has not hardened. During this so-­‐called teneral stage, spiders could have greater exposure to PCBs and their metabolites. The absence of a thick, chitin-­‐rich exoskeleton makes the soft tissue of those organisms more available to spiders, and facilitates the trophic transfer of PCBs. Ecdysone is the molting hormone in arthropods (Chang, 1993) and belongs to the same gene family as the vertebrate thyroid hormone (Laudet, 1997). Crabs exposed to oil spills were observed to exhibit different molting patterns as a result of exposure to thyroid disrupting compounds (Oberdörster et al., 1999), and supported hypotheses that PAHs from the oil may influence the expression of ecdysone. Additionally, PCBs were investigated as they have been demonstrated to interfere with the thyroid hormone system in vertebrates (Cheek et al., 1999;Goldey et al., 1995;Koopman-­‐Esseboom et al., 1994). In those studies the PCBs were not found to induce molting in arthropods. These studies indicate that species exposed to PCBs would not molt more than would be expected in clean habitats. Webs Another potential excretion pathway is web production. Web structure and composition has been extensively studied (Herberstein et al., 2000; Higgins and Rankin, 1999; Opell, 1998). Orb-­‐weaving spiders, which are the subject of this dissertation, rely heavily on silk throughout their entire life cycle (Foelix, 1982). Spiderlings will send up a small string of silk and parachute on the breeze to 21
disperse from the rest of the brood. Webs are used by adults to both capture prey and as a location for their mating habits. Silk is also used to create a nest for fertilized eggs until the spiderlings emerge. Additionally, the webs provide a surface for pheromone deposition. Males are attracted to females when they come into contact with pheromones that are embedded in the lipid matrix of the web (Schulz and Toft, 1993). Orb-­‐weaving spiders produce and enlist up to seven different types of silk (Figure 1.4), each with unique properties and functions (Vollrath and Selden, 2007). Silk is composed of some proteins, carbohydrates, and lipids so as to resist degradation from ambient humidity. In addition, the sticky substance on the web is acidic, approximately pH 4, such that the web is protected from bacterial degradation (Foelix, 1982). 22
Figure 1.4 Different types and functions of silks produced by orb weaver spiders. Adapted from Eisoldt et al. 2011. Orb-­‐webs are constructed such that they use a minimum of resources – just 0.1-­‐0.5mg of silk, and a maximum of thirty minutes for construction (Foelix, 1982). Tetragnathidae web orientation is slightly tilted from vertical, allowing it to efficiently capture aquatic subsidy insects (Gillespie, 1987). Tetragnathidae webs are orb-­‐shaped where there is ample space, but silk tangles are present all along the bare branches wherever room is available, thus maximizing the opportunity to 23
capture prey. The close proximity to the water increases probability of capture partly because emergent insects do not need to travel far before encountering the webs, but also because the insects that emerge are just barely mature. Overall, spider silk is more hydrophobic than hydrophilic to help webs withstand precipitation, humidity, and aridity (Foelix, 1982). The hydrophobicity is accomplished by glycoproteins and lipids, which coat the silk to prevent desiccation and enable prey capture (Opell and Bond, 2000). Lipids, and perhaps glycoproteins, favor the accumulation of hydrophobic contaminants such as PCBs. The lipophilicity of PCB congeners and their metabolites may cause them to partition into these lipids, effectively removing them from the body cavity. Out of body sequestration of contaminants could serve a protective factor, as up to 90% of the web making materials are recycled without metabolism for creation of the new web (Peakall, 1971). The alternative possibility is that cyclic exposure to the same contaminants and metabolites may occur. Female and immature tetragnathids rebuild their webs nightly, consuming the silk material and using it for the next web (Gillespie, 1987; Opell, 1998). The contaminant concentration in the webs could increase over time as body burden increases due to sustained predation of contaminated prey. Webs may also create a new exposure pathway if they are able to collect volatile PCBs and OH-­‐PCBs. The likelihood of volatilization of OH-­‐PCBs is considerably less than that of PCBs because their capability to form hydrogen bonds with water increases their concentration in water, and hence decreases their fugacity. The contaminant concentration in the web may then increase independently from spider diet. 24
PCB Transformation Metabolism of toxicants has various toxicological roles in organisms. It primarily functions to make compounds more hydrophilic such that they can be readily removed from the body. In this case, biotransformation leads to enhanced elimination, detoxification, sequestration, and redistribution. Alternatively, metabolism can be problematic if the toxicants are instead activated to more toxic forms. In the case of hydroxylated PCBs, the new physio-­‐chemical properties of the metabolites resulted in greater toxicity than their parent compounds in rats (Purkey et al., 2004). Similar toxicity trends have yet to be confirmed in invertebrates, but may have similar behavior. Our understanding of PCB metabolism comes from studies with non-­‐spider organisms; however, such studies can guide investigations into spiders and other invertebrate predators. Biotransformation of toxicants involves two phases of action. Phase I includes reactions that serve to make the toxicant slightly more hydrophilic, in that they may add or expose a functional group. The first step may be bypassed if there are compounds that already possess an exposed functional group. Other potential actions in phase I include oxidation, hydrolysis or reduction of the toxicant. Figure 1.5 displays some of the metabolic pathways of PCBs determined in reptiles, birds, and mammals. Phase II reactions add a hydrophilic group through a conjugation reaction. These groups can be sulfate groups, sugars, glutathione, or amino acids. Some reactions are meant to detoxify activated metabolites, so acetylation or methylation may be used to inactivate the products of phase I reactions. Overall, 25
hydrophilicity should increase by many orders of magnitude to facilitate excretion. Spider excrement is only via solids, so increasing hydrophilicity may allow for incorporation of PCB metabolites into the crystalline structure of the excreted solids. The biological half-­‐life of PCBs is determined by their susceptibility to metabolism, and is roughly determined from their bioconcentration rate, or rate of accumulation (ka) as compared to their rate of elimination (ke). If ka > ke, then compounds bioaccumulate. These rates are determined by the number and position of chlorine atoms. Metabolic resistance is great if there are five or more chlorines, especially if these are para positioned chlorines (Wiberg et al., 1998). For example, PCB congener 153 is highly resistant to metabolism because it lacks adjacent hydrogen atoms on the biphenyl system (Borlakoglu and Wilkins, 1993; Bühler et al., 1988). The result is a half life of approximately 338 days for PCB 153 in humans (Letcher et al. 2000). Congeners with meta-­‐ and para-­‐ positioned hydrogen are more susceptible to metabolism than those with meta-­‐ and para-­‐chlorine (James, 2001). All exogenous toxicants need to pass through cells to enter systemic circulation. All PCBs are lipophilic enough to be able to pass through the lipid bilayer of the cell membrane without becoming lodged in between the polar head groups, but from that point the fate of the chemicals is more variable. Equilibrium partitioning is not appropriate for use in this instance, because biota are not often at equilibrium with the surrounding environment. 26
Figure 1.5 Metabolism of PCBs. Adapted from Letcher et al. ( 2000). While PCBs are robust, legacy compounds, they can still be transformed in the environment. A better understanding of these processes may allow for their more reliable use as tracers to help understand ecological processes, much as we have used stable isotopes. 27
Hydroxylated PCBs are formed as a result of detoxifying action by cytochrome P450 (CYP) 1A and 2B. Detoxifying enzymes typically have very broad substrate specificity, but these isoforms are known to process mostly xenobiotics (Klassen, 2008). CYP enzymes are located on the endoplasmic reticulum of cells, and are found in high concentrations in the liver, as that organ typically encounters most of the xenobiotics introduced to an organism. CYP is found at low basal expression levels in all other tissues, too, because these enzymes are an important part of the defense against toxicants. As part of phase I of detoxification, these enzymes serve to make hydrophobic compounds slightly more hydrophilic through addition of polar functional groups. The most prevalent hydroxylated metabolites of PCBs are p-­‐OH substituted. The basic CYP reaction is a monooxygenase reaction, associated with NADPH-­‐CYP450 reductase. The general reaction is written as: RH + O2 + NADPH + H+ ↔ ROH + H2O + NADP+ (Equation 3) CYP epoxidation may occur as a result of this reaction, which is typically not problematic; however, when reorganization does not occur the ring can be distorted. OH-­‐PCBs can be a result of epoxidation or direct insertion of the OH-­‐ functionality by CYP1A/2B. Some arene intermediates may be stable for weeks if the three-­‐membered ring experiences the stabilizing effect of two chlorines on either side of the arene oxide (Reich et al., 1978). Phase II enzymes may make further modifications to the metabolites. Enzymes such as glutathione-­‐s-­‐transferase (GST) serve to make the compound of interest more hydrophilic by addition of additional polar groups, thus making them 28
likely to be excreted from the organism typically through urine or bile. As a result, OH-­‐PCBs are likely to be present at low, transient levels. GST modifies electrophilic xenobiotics such as PCBs, and may be able to bind and store these compounds. GST is abundant in mammals, constituting up to ten percent of liver protein (Klaasen, 2008). OH-­‐PCBs can be further processed to become quinines or catechols [(OH)2-­‐
PCB] (Amaro et al., 1996). In the laboratory, PCB metabolites may be oxidized if exposed to air for prolonged periods of time (Letcher et al., 2000). OH-­‐PCBs are hydrophilic enough that they may be excreted, but it is also possible that they may interact with other proteins (Kania-­‐Korwel et al., 2008a). OH-­‐PCBs are typically found in significant concentrations in the plasma; however they have also been found in the adipose, lung, kidney, and liver tissues in organisms such as mice, fish, albatross, grey seals, polar bears, and humans (Sinjari et al., 1998; Klasson-­‐Wehler et al., 1987, 1997, 1998; Bergman et al., 1994; Sandau and Norstrom, 1996; Buckman et al., 2006). Hydroxylated PCBs are known to affect the mitochondrial membrane, making it more permeable and affecting ATP production (Stadnicki and Allen, 1979; Ebner and Braselton, 1987; Yamamoto and Yoshimura, 1973; Narasimhan et al., 1991). When ATP is not generated at the appropriate levels cellular energy stores are affected, but also ionic and volume regulation, resulting in a more acidic pH of the cytoplasm. Endocrine disrupting compounds are a class of emerging toxicants currently of great interest. OH-­‐PCBs have been demonstrated to be weakly estrogenic (Kramer et al., 1997; Moore et al., 1997), though the studied congeners that 29
exhibited such behavior are not environmentally relevant. Ortho-­‐chlorinated biphenyl metabolites are associated with alterations in vitamin A function and thyroid hormone metabolism (Brouwer, 1991). The thyroid hormone thyroxine (T4) functions to increase the metabolic rate and oxygen consumption in an individual. In invertebrates, the molting hormone ecdysone has a similar structure as thyroxine, so the presence of OH-­‐PCBs in their systems may also negatively affect spiders. The iodine substitution pattern of T4 (3,3’,4’,5-­‐tetraiodo-­‐L-­‐thyroxine) resembles the structure of many OH-­‐PCB metabolites. Many of the structural elements of the metabolites can non-­‐covalently bind to transthyretin, (TTR) which functions as a transporter for thyroxine, causing the metabolites to mimic thyroxine and activate the hormone’s receptor (see Figure 1.6). The metabolites function as agonists, and induce UDP – glucuronosyl transferase enzymes to glucoronidate and remove the thyroxine more quickly. 30
Figure 1.6 A. T3 and T4 regulation by the Thyroid; B. Structure of thyroxine (T4). Source: <http://www.bio.miami.edu/oreilly/ganser/03-­‐01-­‐05Endo.htm>. While the concentrations in tissues other than the plasma are not significant, it is important to note that hormones in blood are typically present at very low concentrations. Free thyroxine is typically present at trace levels, so mimic compounds can be effective at very low concentrations. Typically mimic 31
compounds do not “fit” the receptor as well as the intended compound. However, the affinity of OH-­‐PCBs for TTR may be ten times greater than that of T4 (Brouwer et al., 1998;Lans et al., 1993). Reducing the amount of thyroxine present in the plasma leads to increased thyroid stimulating hormone (TSH), which creates hypertrophy (where the volume of an organ or tissue is increased due to the enlargement of its component cells) or neoplasia (abnormal growth of cells) of the thyroid. These effects are less problematic in vivo than in vitro (Schuur et al., 1996, 1998, 1999), though they may have an effect on fetal brain development (Brouwer et al., 1998). PCB Metabolism by Spiders The bulk of the research on hydroxylated PCB metabolites (OH-­‐PCBs) has been on mammals, fish, and birds. Since PCBs have been detected in spiders it is likely that both spiders and their predators will possess metabolites. Research may expand in time to include other arachnivorous predators such as reptiles and amphibians. While the number of PCB congeners detected in tissue is typically high, fewer congeners appear to undergo hydroxylation (Klasson-­‐Wehler et al., 1998). Digestion begins in spiders in the midgut (Figure 1.4), which is expected to have the same functionality as seen in the liver in vertebrates (Foelix, 1982). The muscles of this compartment will contract and expand such that a vacuum is created in the digestive tract, which effectively creates a suction pump that draws the prey-­‐
fluid into the tract. The midgut is the only section of the digestive physiology that 32
lacks the thin cuticular coating, thus making it the only area designed for digestion. First secretory cells empty enzymatic granules into the gut lumen, followed by resorptive cells that process the nutrients further in food vacuoles before the metabolites move into interstitial tissue or the hemolymph. The secretory cells may take up to eight hours to empty all granules, and at that time resorptive cells may begin to contain excretory products, as well as stores of glycogen and lipids. The hypothesized process of PCB metabolism described herein is inferred from the understanding of spider digestion and predation, as well as empirical evidence of OH-­‐PCB generation and activity. Summary Ecological and exposure studies tell us that riparian spiders are an important linkage between the aquatic and terrestrial habitats. Analyses of these compartments indicate that spiders can act as biovectors, transferring energy, carbon, and even contamination from aquatic to terrestrial food webs. In particular, Tetragnathidae spiders occupy a niche that makes them important biovectors. Their foraging tactics (Table 1.1) differentiate them from other spiders such as Araneidae and basilica spiders. Less is known about the physiology and detoxification of organic contaminants by spiders. Understanding these can inform our interpretation of the spider’s role in contaminant fate, transport, and transformation. 33
References Akamatzu, F., Toda, H., Okino, T., 2004. Food source of riparian spiders analyzed by using stable isotope ratios. Ecol. Res. 19, 655-­‐662. Amaro, A.R., Oakley, G.G., Bauer, U., Spielmann, H.P., Robertson, L.W., 1996. Metabolic activation of PCBs to quinones: Reactivity toward nitrogen and sulfur nucleophiles and influence of superoxide dismutase. Chem. Res. Toxicol. 9, 623-­‐629. Bartrons, M., Grimalt, J.O., Catalan, J., 2007. Concentration changes of organochlorine compounds and polybromodiphenyl ethers during metamorphosis of aquatic insects. Environ. Sci. Technol. 41, 6137-­‐6141. Baxter, C.V., Fausch, K.D., Saunders, W.C., 2005. Tangled webs: Reciprocal flows of invertebrate prey link streams and riparian zones. Freshwat. Biol. 50, 201-­‐
220. Bergman, Å, Klasson-­‐Wehler, E., Kuroki, H., 1994. Selective retention of hydroxylated PCB metabolites in blood. Environ. Health. Perspect. 102, 464-­‐
469. Blackledge, T.A., Scharff, N., Coddington, J.A., Szuts, T., Wenzel, J.W., Hayashi, C.Y., Agnarsson, I., 2009. Reconstructing web evolution and spider diversification in the molecular era. Proc. Natl Acad. Sci. USA 106, 5229-­‐5234. Borlakoglu, J.T., Wilkins, J.P.G., 1993. Correlations between the molecular structures of polyhalogenated biphenyls and their metabolism by hepatic-­‐microsomal monooxygenases. Comp. Biochem. Physiol. 105C, 113-­‐117. Brenner, R.C., Magar, V.S., Ickes, J.A., Foote, E.A., Abbot, J.E., Bingler, L.S., Crecelius, E.A., 2004. Long-­‐term recovery of PCB-­‐contaminated surface sediments at the Sangamo-­‐Weston/Twelvemile Creek/Lake Hartwell Superfund Site. Environ. Sci. Technol. , 2328-­‐2337. Bristowe, W.S., 1939, 1941. The Comity of Spiders. The Ray Society. Republished 1968 by Johnson Reprint Corp. New York, London. Brouwer, A., 1991. Role of biotansformation in PCB-­‐induced alterations in vitamin A and thyroid hormone metabolism in laboratory and wildlife species. Biochem Society Transact 19, 731-­‐737. 34
Brouwer, A., Morse, D., Lans, M., Schuur, A., Murk, A., Klasson-­‐Wehler, E., Bergman, A., Visser, T., 1998. Interactions of persistent environmental organohalogens with the thyroid hormone system: Mechanisms and possible consequences for animal and human health. Toxicol. Ind. Health 14, 59-­‐84. Buckman, A.H., Brown, S.B., Small, J., Muir, D.C.G., Parrott, J., Solomon, K.R., Fisk, A.T., 2007. Role of temperature and enzyme induction in the biotransformation of polychlorinated biphenyls and bioformation of hydroxylated polychlorinated biphenyls by rainbow trout (Oncorhynchus mykiss). Environ. Sci. Technol. 41, 3856-­‐3863. Buckman, A.H., Wong, C.S., Chow, E.A., Brown, S.B., Solomon, K.R., Fisk, A.T., 2006. Biotransformation of polychlorinated biphenyls (PCBs) and bioformation of hydroxylated PCBs in fish. Aquatic Toxicology 78, 176-­‐185. Bühler, F., Schmid, P., Schlatter, C., 1988. Kinetics of PCB elimination in man. Chemosphere 17, 1717-­‐1726. Calvo, R., Obregon, M.J., Ruiz de Ona, C., Escobar del Rey, F., Morreale de Escobar, G., 1990. Congenital hypothyroidism, as studied in rats: The crucial role of maternal thyroxine but not of 3,5,3'-­‐triiodothyronine in the protection of the fetal brain. J. Clin. Invest. 86, 889-­‐899. Campfens, J., Mackay, D., 1997. Fugacity-­‐based odel of PCB bioaccumulation in complex aquatic food webs. Environ. Sci. Technol. 31, 577-­‐583. Carlson, D.B., Williams, D.E., 2001. 4-­‐Hhydroxy-­‐2’,4’,6’-­‐trichlorobiphenyl and 4-­‐ hydroxy-­‐2’,3’,4’,5’-­‐tetrachlorobiphenyl are estrogenic in rainbow trout. Environmental Toxicology and Chemistry 20, 351-­‐358. Carrel, J.E., Heathcote, R.D., 1976. Heart rate in spiders: Influence of body size and foraging energetics. Science 193, 148-­‐150. Chacon, P., Eberhard, W.G., 1980. Factors affecting numbers and kinds of prey caught in artificial spider webs, with considerations of how orb webs trap prey. Bull. Br. Arachnol. Soc. 5, 28-­‐38. Chang, E.S., 1993. Comparitive endocrinology of molting and reproduction: Insects and crustaceans. Annu. Rev. Entomol. 38, 161-­‐180. 35
Cheek, A.O., Kow, K., Chen, J., McLachlan, J.A., 1999. Potential mechanisms of thyroid disruption in humans: Interaction of organochlorine compounds with thyroid receptor, transthyretin, and thyroid-­‐binding globulin. Environ. Health Perspect. 107, 273-­‐278. Danerud, P.O., Morse, D.C., Klasson-­‐Wehler, E., Brouwer, A., 1996. Binding of a 3,3',4,4'-­‐tetrachlorobibhenyl (CB-­‐77) metabolite to fetal transthyretin and effects of fetal thyroid hormone levels in mice. Toxicology 106, 105-­‐114. de Haan, L.H.J., van Dijk, L., Brouwer, A., 1994. Effects of hydroxylated PCB metabolites on the gap-­‐junctional intercellular communication between mouse Hepa 1c1c7 cells. Organohalogen Compounds 20, 491-­‐495. deVoogt, P., Brinkman, U.A.T., 1989. Production, properties, and usage of polychlorinated biphenyls, in: Kimbrough, R.D., Jensen, A.A. (Eds.), Halogenated Biphenyls, Terphenyls, Naphthalenes, Dibenzodioxin and Related Products, 2nd ed. Elsevier, Amsterdam, pp. 3-­‐46. Easson, L.H., Stedman, E., 1933. Studies on the relationship between chemical constitution and physiological action. V. Molecular dyssymmetry and physiological activity. Biochem J 27, 1257-­‐1266. Ebner, K.V., Braselton, W.E.J., 1987. Structural and chemical requirements for hydroxychlorobiphenyls to uncouple rat liver mitochonria and potentiation of uncoupling with Aroclor 1254. Chem-­‐Biol Interact 63, 139-­‐155. Eisoldt, L., Thamm, C., Scheibel, T., 2012. The role of terminal domains during storage and assembly of spider silk proteins. Biopolymers 97, 355-­‐361. Eisoldt, L., Smith, A., Scheibel, T., 2011. Decoding the secrets of spider silk. Materials Today 14, 80-­‐86. Foelix, R.F., 1982. Biology of Spiders, 1st ed. Harvard University Press, Cambridge, Massachusetts. Gillespie, R.G., Caraco, T., 1987. Risk-­‐sensitive foraging strategies of two spider populations. Ecology 68, 887-­‐899. Gillespie, R.G., 1987. The mechanism of habitat selection in the long-­‐jawed orb-­‐ weaving spider Tetragnatha elongata (Araneae, Tetragnathidae). J. Arachnol. 15, 81-­‐90. 36
Goldey, E.S., Kehn, L.S., Lau, C., Rehnberg, G.L., Crofton, K.M., 1995. Developmental exposure to polychlorinated biphenyls (Arochlor 1254) reduces circulating thyroid hormone concentrations and causes hearing deficits in rats. Toxicol. Appl. Pharmacol. 135, 77-­‐88. Greenstone, M.H., 1979. Spider feeding behaviour optimises dietary essential amino acid composition. Nature 282, 501-­‐503. Greim, H., Snyder, R., 2008. Toxicology and Risk Assessment: A Comprehensive Introduction. Wiley-­‐Interscience. 1, 387. Guvenius, D.M., Hassanzadeh, P., Bergman, A., Norent, K., 2002. Metabolites of polychlorinated biphenyls in human liver and adipose tissue. Environ. Toxicol. Chem. 21, 2264-­‐2269. Haglund, P., Wiberg, K., 1996. Determination of the gas chromatographic elution sequences of the (+) and (-­‐) enantiomers of stable atropisomeic PCBs on Chirasil-­‐Dex. J. High Resol. Chrom. 19, 373-­‐376. Harner, T., Wiberg, K., Norstrom, R., 2000. Enantiomer fractions are preferred to enantiomer ratios for describing chiral signatures in environmental analysis. Environ. Sci. Technol. 34, 218-­‐220. Heim, M., Romer, L., Scheibel, T., 2010. Hierarchical structures made of proteins. The complex architecture of spider webs and their constituent silk proteins. Chem. Soc. Rev. 39, 156-­‐164. Heim, M., Keerl, D., Scheibel, T., 2009. Spider silk: From soluble protein to extraordinary fiber. Angewandte Chemie 48, 3584-­‐3596. Helgason, L.B., Arukwe, A., Gabrielsen, G.W., Harju, M., Hegseth, M.N., Heimstad, E.S., Jørgensen, E.H., Mortensen, A.S., Wolkers, J., 2010. Biotransformation of PCBs in Arctic seabirds: Characterization of phase I and II pathways at transcriptional, translational and activity levels. Comp. Biochem. Physio. C 152, 34-­‐41. Henschel, J.R., Mahsberg, D., Stumpf, H., 2001. Allochthonous aquatic insects increase predation and decrease herbivory in river shore food webs. Oikos 93, 429-­‐438. Herberstein, M.E., Craig, C.L., Coddington, J.A., Elgar, M.A., 2000. The functional significance of silk decorations of orb-­‐web spiders: A critical review of the empirical evidence. Bio. Rev. 75, 649-­‐669. 37
Higgins, L., Rankin, M.A., 1999. Nutritional requirements for web synthesis in the tetragnathid spiderNephila clavipes. Physiol. Entomol. 24, 263-­‐270. Hoekstra, P.F., Letcher, R.J., O'Hara, T.M., Backus, S.M., Solomon, K.R., Muir, D.C.G., 2003. Hydroxylated and methylsulfone-­‐containing metabolites of polychlorinated biphenyls in the plasma and blubber of bowhead whales (Balaena mysticetus). Environ. Toxicol. Chem. 22, 2650-­‐2658. Holling, C.S., 1959. The components of predation as revealed by a study of small mammal predation of the European sawfly. Can. Entomol. 91, 293-­‐320. Hühnerfuss, H., Kallenborn, R., Konig, W.A., Rimkus, G., 1992. Preferential enrichment of the (+)-­‐α-­‐hexachlorocylcohexane in cerebral matter of harbour seals. Organohal. Compds. 10, 97-­‐100. Ishida, C., Koga, N., Hanioka, N., Saeki, H.K., Yoshimura, H., 1991. Metabolism in vitro of 3,4,3',4'-­‐ and 2,5,2',5'-­‐tetrachlorobiphenyl by rat liver microsomes and highly purified cytochrome P450. J. Pharmacobiodyn. 14, 276-­‐284. James, M.O., 2001. Polychlorinated biphenyls: Metabolism and metabolites, in: PCBs: Advances in Environmental Toxicology and Health Effects. Robertson, L., Hansen, L. (Eds.). The University Press of Kentucky, pp. 35-­‐46. Jaspers, V.L.B., Dirtu, A.C., Eens, M., Neels, H., Covaci, A., 2008. Predatory bird species show different patterns of hydroxylated polychlorinated biphenyls (OH-­‐
PCBs) and polychlorinated biphenyls (PCBs). Environ. Sci. Technol. 42, 3465-­‐3471. Johnson, D.M., Akre, B.G., Crowley, P.H., 1975. Modeling arthropod predation: Wasteful killing by damselfly naiads. Ecology 56, 1081-­‐1093. Kallenborn, R., Hühnerfuss, H., 2001. Chiral Environmental Pollutants: Trace Analysis and Ecotoxicology. Springer. Kania-­‐Korwel, I., Xie, W., Hornbuckle, K.C., Robertson, L.W., Lehmler, H., 2008a. Enantiomeric enrichment of 2,2′,3,3′,6,6′-­‐hexachlorobiphenyl (PCB 136) in ice after induction of CYP enzymes. Arch. Environ. Contam. Toxicol. 55, 510-­‐
517. Kania-­‐Korwel, I., Zhao, H., Norstrom, K., Li, X., Hornbuckle, K.C., Lehmler, H., 2008b. Simultaneous extraction and clean-­‐up of polychlorinated biphenyls and their metabolites from small tissue samples using pressurized liquid extraction. Journal of Chromatography A 1214, 37-­‐46. 38
Kaston, B.J., 1970. Comparative biology of the American Black Widow spiders. Tansact. San Diego Soc. Nat. Hist. 16, 33. Klassen, C.D., 2008. Toxicology: The Basic Science of Poisons. , 7th ed. McGraw Hill,1309. Klasson-­‐Wehler, E., Hovander, L., Bergman, Å, 1997. New organohalogens in human plasma – identification and quantification. In: 17th Symposium on Halogenated Environmental Organic Pollutants, Dioxin’97. Organohalogen Compounds 33, 420-­‐425. Klasson-­‐Wehler, E., Bergman, Å, Kowalski, B., Brandt, I., 1987. Metabolism of 2,3,4',6-­‐tetrachlorobiphenyl -­‐ formation and tissue localization of mercapturic acid pathway metabolites in mice. Xenobiotica 17, 477-­‐486. Klasson-­‐Wehler, E., Bergman, Å, Athanasiadou, M., Ludwig, J.P., Auman, H.J., Kannan, K., Van Den Berg, M., Murk, A.J., Feyk, L.A., Giesy, J.P., 1998. Hydroxylated and methylsulfonyl polychlorinated biphenyl metabolites in albatrosses from Midway Atoll, North Pacific Ocean. Environ. Toxicol. Chem. 17, 1620-­‐1625. Kodavanti, P.R.S., Ward, T.R., Mckinney, J.D., Tilson, H.A., 1995. Increased (3H) phorbol ester binding in rat cerebellar granule cells by polychlorinated biphenyl mixtures and congeners: Structure-­‐activity relationships. Toxicol. Appl. Pharmacol. 130, 140-­‐148. Koopman-­‐Esseboom, C., Morse, D.C., Weisglas-­‐Kuperus, N., Lutkeschipholt, I.J., VanDerPaauw, C.G., Tuinstra, L.G.M.T., Brouwer, A., Sauer, P.J.J., 1994. Effects of dioxin and polychlorinated biphenyls on thyroid hormone status of pregnant women and their infants. Pediatr. Res. 36, 468-­‐473. Kramer, V.J., Helferich, W.G., Bergman, Å, Klasson-­‐Wehler, E., Giesy, J.P., 1997. Hydroxylated polychlorinated biphenyl metabolites are anti-­‐estrogenic in a stably transfected human breast adenocarcinoma (MCF7) cell line. Toxicol. Appl. Pharmacol. 144, 363-­‐376. Lans, M.C., Brouwer, A., Koppe, J.G., van den Berg, M., 1990. Enzyme-­‐Induction and alterations in thyroid-­‐hormone, vitamin-­‐A and vitamin-­‐K levels by TCDD in neonatal and maternal rats. Chemosphere 20, 1129-­‐1134. Lans, M.C., 1995. Thyroid hormone binding proteins as novel targets for hydroxylated polyhalogenated aromatic hydrocarbons (PHAHs): Possible implications for toxicity. Agricultural University Wageningen . 39
Lans, M.C., Klasson-­‐Wehler, E., Willemsen, M., Meussen, E., Safe, S., Brouwer, A., 1993. Structure-­‐dependent, competitive interaction of hydroxy-­‐
polychlorobiphenyls, -­‐dibenzo-­‐p-­‐dioxins and -­‐dibenzofurans with human transthyretin. Chem. Biol. Interact. 88, 7-­‐21. Laudet, V., 1997. Evolution of the nuclear receptor superfamily: Early diversificiation from an ancestral orphan receptor. J. Mol. Endocrinol. 19, 207-­‐226. Lehmler, H.J., Harrad, S.J., Huhnerfuss, H., Kania-­‐Korwel, I., Lee, C.M., Lu, Z., Wong, C.S., 2009. Chiral polychlorinated biphenyl transport, metabolism, and distribution: A review. Environ. Sci. Technol. 44, 2757-­‐2766. Letcher, R.J., Klasson-­‐Wehler, E., Bergman, Å, 2000. Methyl sulfone and hydroxylated metabolites of polychlorinated biphenyls, in: New Types of Persistent Halogenated Compounds. Hutzinger, O. (Ed.). Springer, pp. 315. Letcher, R.J., Bustnes, J.O., Dietz, R., Jenssen, B.M., Jørgensen, E.H., Sonne, C., Verreault, J., Vijayan, M.M., Gabrielsen, G.W., 2010. Exposure and effects assessment of persistent organohalogen contaminants in arctic wildlife and fish. Sci. Total Environ. 408, 2995-­‐3043. Łuczak, J., 1963. Differences in the structure of communities of web spiders in one type of environment (young pine forest). Ekologia Polska 11, 160-­‐221. Mackay, D., Fraser, A., 2000. Bioaccumulation of persistent organic chemicals: mechanisms and models. Environ. Pollut. 110, 375-­‐391. Mackay, D., Shiu, W.Y., Ma, K.C., 1992. Illustrated Handbook of Physical-­‐Chemical Properties and Environmental Fate for Organic Chemicals, Vol. I: Monoaromatice Hydrocarbons, Chlorobenzenes, and PCBs. CRC Press, Boca Raton, FL. I. McCune, B., Grace, J.B., Urban, D.L., 2002. Analysis of Ecological Communities. MjM Software Design, Gleneden Beach, OR. McCune, B., Mefford, M.J., 1999. PC-­‐ORD. Multivariate Analysis of Ecological Data, Version 4. MjM Software Design, Gleneden Beach, OR. McKinney, M.A., Guise, S.D., Martineau, D., Béland, P., Lebeuf, M., Letcher, R.J., 2006. Organohalogen contaminants and metabolites in beluga whale (Delphinapterus leucas) liver from two Canadian populations. Environ. Toxicol. Chem. 25, 1246-­‐1257. 40
Moore, M., Mustain, M., Daniel, K., Chen, I., Safe, S., Zacharewski, T., Gillesby, B., Joyeux, A., Balaguer, P., 1997. Antiestrogenic activity of hydroxylated polychlorinated biphenyl congeners identified in human serum. Toxicol. Appl. Pharmacol. 142, 160-­‐168. Morse, D.C., Klasson-­‐Wehler, E., Wesseling, W., Koeman, J.H., Brouwer, A., 1996. Alterations in rat brain thyroid hormone status following pre-­‐ and post-­‐natal exposure to polychlorinated biphenyls (Aroclor 1254). Toxicol. App. Pharmacol 136, 269-­‐279. Morse, D.C., van Bladeren, P.J., Klasson-­‐Wehler, E., Brouwer, A., 1995. β-­‐ Naphthoflavone-­‐ and self-­‐induced metabolism of 3,3',4,4'-­‐
tetrachlorobuiphenyl in hepatic microsomes of the male, pregnant female, and foetal rat. Xenobiotica 25, 245-­‐260. Murdoch, W.W., 1977. Stabilizing effects of spatial heterogeneity in predator-­‐prey systems. Theor. Popul. Biol. 11, 252-­‐273. Narasimhan, T.R., Kim, H.L., Safe, S., 1991. Effects of hydroxylated polychlorinated biphenyls on mouse liver mitochondrial oxidative phosphorylation. J Biochem Toxicol 6, 229-­‐236. Nomiyama, K., Nomura, Y., Takahashi, T., Uchiyama, Y., Arizono, K., Shinohara, R., 2010. Hydroxylated polychlorinated biphenyls (OH-­‐PCBs) induce vitellogenin through estrogenic activity in primary-­‐cultured hepatocytes of the Xenopus laevis. Chemosphere 78, 800-­‐806. Oberdörster, E., Cottam, D.M., Wilmot, F.A., Milner, M.J., McLachlan, J.A., 1999. Interaction of PAHs and PCBs with ecdysone-­‐dependent gene expression and cell proliferation. Toxicol. Appl. Pharmacol. 160, 101-­‐108. Opell, B.D., Markley, B.J., Hannum, C.D., Hendricks, M.L., 2008. The contribution of axial fiber extensibility to the adhesion of viscous capture threads spun by orb-­‐weaving spiders. J. Exp. Biol. 211, 2243-­‐2251. Opell, B.D., Bond, J.E., 2000. Capture thread extensibility of orb-­‐weaving spiders: testing punctuated and associative explanations of character evolution. Biol. J. Linn. Soc. 70, 120-­‐120. Opell, B.D., 1998. Economics of spider orb-­‐webs: The benefits of producing adhesive capture thread and of recycling silk. Funct. Ecol. 12, 613-­‐624. 41
Pakdeesusuk, U., Lee, C.M., Coates, J.T., Freedman, D.L., 2005. Assessment of natural attenuation via in situ reductive dechlorination of polychlorinated biphenyls in sediments of the Twelve Mile Creek Arm of Lake Hartwell, SC. Environ. Sci. Technol. 39, 945-­‐952. Pakdeesusuk, U., Freedman, D.L., Lee, C.M., Coates, J.T., 2003a. Reductive dechlorination of polychlorinated biphenyls in sediment from the Twelve Mile Creek arm of Lake Hartwell, South Carolina, USA. Environ. Toxicol. Chem. 22, 1214-­‐1220. Pakdeesusuk, U., Jones, W.J., Lee, C.M., Garrison, A.W., O'Niell, W.L., Freedman, D.L., Coates, J.T., Wong, C.S., 2003b. Changes in enantiomeric fractions during microbial reductive dechlorination of PCB132, PCB149, and Aroclor 1254 in Lake Hartwell sediment microcosms. Environ. Sci. Technol. 37, 1100-­‐1107. Park, J., Kalantzi, O.I., Kopec, D., Petreas, M., 2009. Polychlorinated biphenyls (PCBs) and their hydroxylated metabolites (OH-­‐PCBs) in livers of harbor seals (Phoca vitulina) from San Francisco Bay, California and Gulf of Maine. Mar. Environ. Res. 67, 129-­‐135. Peakall, D.B., 1971. Conservation of web proteins in the spiderAraneus diadematus. J. Exp. Zool. 176, 257. Prouvost, O., Trabalon, M., Papke, M., Schulz, S., 1999. Contact sex signals on web and cuticle of Tegenaria atrica (Araneae, Agelenidae). Arch. Insect Biochem. Physiol. 40, 194-­‐202. Purkey, H.E., Palaninathan, S.K., Kent, K.C., Smith, C., Safe, S.H., Sacchettini, J.C., Kelly, J.W., 2004. Hydroxylated polychlorinated biphenyls selectively bind transthyretin in blood and inhibit amyloidogenesis: Rationalizing rodent PCB toxicity. Chem. Biol. 11, 1719-­‐1728. Raikow, D., Walters, D., Fritz, K., Mills, M., 2011. The distance that contaminated aquatic subsidies extend into lake riparian zones. Ecol. Appl. 21, 983-­‐990. Reich, H.J., Reich, I.L., Wollowitz, S., 1978.Conversion of allyl alcohols to dienes by sulfoxide and selenoxide syn-­‐elimination—Synthesis of PCB arene oxides. J Am Chem Soc 100, 5981-­‐5982. Riechert, S.E., Lockley, T., 1984. Spiders as biological control agents. Annu. Rev. Entomol. 29, 299-­‐320. 42
Riechert, S.E., 1981. The consequences of being territorial: Spiders, a case study. Am. Nat. 117, 871-­‐892. Safe, S., 1993. Toxicology, structure–function relationship, and human and environmental health impacts of polychlorinated biphenyls: Progress and problems. Environ. Health. Perspect. 100, 259-­‐268. Safe, S.H., 1990. Polychlorinated aromatics: Uptake, disposition and metabolism, in: Halogenated biphenyls, terphenyls, naphthalenes, dibenzodioxins, and related products. Kimbrough, R.D., Jensen, A.A. (Eds.), 2nd ed. Nature, London, pp. 131. Safe, S.H., 1994. Polychlorinated-­‐biphenyls (PCBs) -­‐ Environmental-­‐impact, biochemical and toxic responses, and implications for risk assessment. Crit. Rev. Toxicol. 24, 87-­‐149. Sandau, C.D., Ayotte, P., Dewailly, E., Duffe, J., Norstrom, R.J., 2000. Analysis of hydroxylated metabolites of PCBs (OH-­‐PCBs) and other chlorinated phenolic compounds in whole blood from Canadian Inuit. Environ. Health. Perspect. 108, 611-­‐616. Sandau, C.D., Norstrom, R.J., 1996. Comparison of methylation methods for the determination of hydroxy-­‐PCBs and preliminary results for polar bear blood plasma. Organohal. Compds. 29, 412-­‐417. SCDHEC, 2006. Lake Hartwell Restoration and Compensation Determination Plan. Schulz, S., Toft, S., 1993. Branched long chain alkyl methyl ethers: A new class of lipids from spider silk. Tetrahedron 49, 6805-­‐6820. Schurig, V., Glausch, A., Fluck, M., 1995. On the enantiomerization barrier of atropisomeric 2,2’,3,3’,4,6’-­‐hexachlorobiphenyl (PCB 132). Tetrahed. Asym. 6, 2161-­‐2164. Schuur, A.G., van Meeteren, M.E., Jong, W.M.C., Bergman, Å, Visser, T.J., Brouwer, A., 1996. Organohal. Compds. 28, 509. Schuur, A.G., Bergman, Å, Brouwer, A., Visser, T.J., 1999. Effects of pentachlorophenol and hydroxylated polychlorinated biphenyls on thyroid hormone conjugation in a rat and a human hepatoma cell line. Toxicology in Vitro 13, 417-­‐425. 43
Schuur, A.G., Legger, F.F., van Meeteren, M.E., Moonen, M.J.H., van Leeuwen-­‐Bol, I., Bergman, A., Visser, T.J., Brouwer, A., 1998. In vitro inhibition of thyroid hormone sulfation by hydroxylated metabolites of halogenated aromatic hydrocarbons. Chem. Res. Toxicol. 11, 1075-­‐1081. Schwarzenbach, R.P., Gschwend, P.M., Imboden, D.M., 2003. Environmental Organic Chemistry. John Wiley & Sons. Seegal, R.F., Bush, B., Shain, W., 1990. Lightly chlorinated ortho-­‐substituted PCB congeners decrease dopamine in nonhuman primate brain and in tissue culture. Toxicol. Appl. Pharmacol. 106, 134-­‐136. Selden, P.A., Shear, W.A., Sutton, M.D., 2008. Fossil evidence for the origin of spider spinnerets, and a proposed arachnid order. Proc. Natl Acad. Sci. USA 105, 20781-­‐20785. Society of Environmental Toxicology and Chemistry (SETAC). Biotransformation in Environmental Risk Assessment, 2004. 2 ed. Society of Environmental Toxicology and Chemistry (SETAC), Pensacola FL, USA. Shain, W., Bush, B., Seegal, R.F., 1991. Neurotoxicity of polychlorinated biphenyls: Structure-­‐activitye relationships of individual congeners. Toxicol. Appl. Pharmacol. 111, 33-­‐42. Sinjari, T., Klasson-­‐Wehler, E., Hovander, L., Darnerud, P.O., 1998. Hydroxylated polychlorinated biphenyls: distribution in the pregnant mouse. Xenobiotica 1, 31-­‐40. Southwell, B.R., Duan, W., Alcorn, D., Brack, C., Richardson, S.J., Kohrle, J., Schreiber, G., 1993. Thyroxine transport to the brain: Role of protein synthesis by the choroid plexus. Endocrinology 133, 2116-­‐2126. Stadnicki, S.S., Allen, J.R., 1979. Toxicity of 2,2’,5,5’-­‐tetrachlorobiphenyl and its metabolites, 2,2’,5,5’-­‐tetrachlorobiphenyl-­‐3,4-­‐oxide and 2,2’,5,5’-­‐
tetrachlorobiphenyl-­‐4-­‐ol to cultured cells in vitro. Bull. Environ. Contam. Toxicol. 23, 788-­‐796. Tanabe, S., Hidaka, H., Tatsukawa, R., 1983. PCBs and chlorinated-­‐hydrocarbon pesticides in Antarctic atmosphere and hydrosphere. Chemosphere 12, 277-­‐
288. Tretzel, E., 1955. Intragenerische Isolation und interspezifische Konkurrenz bei Spinnen. Zeitschrift fur Morphologie und Okologie Tierre 44, 43-­‐162. 44
Truman, P., van den Hurk, P., 2010. Xenoestrogen exposure and effects in bluegill from the Reedy River, South Carolina, USA. Arch. Environ. Contam.Toxicol. 58, 165-­‐175. U.S. EPA, 2005. Method 3545: Standard Operating Procedure for Accelerated Solvent Extraction (ASE-­‐200) of Organics. EPS, Washington, DC. U.S. EPA, 2004. The Incidence and Severity of Sediment Contamination in Surface Waters of the United States, National Sediment Quality Survey. U.S. EPA, 1994. Final Record of Decision for the Sangamo Weston/Twelve Mile Creek/Lake Hartwell PCB Contamination Superfund Site-­‐Operable Unit Two, Pickens, Pickens County, South Carolina. Atlanta, GA: U. S. Environmental Protection Agency, Region IV. U.S. EPA, 1980. Ambient Water Quality Criteria for PCBs. U. S. EPA, Criteria and Standards Division, Washington, D. C. U.S. EPA, 1976. Toxic Substances Control Act (TSCA). Public Law 94-­‐469, 90 Stat. 2003. Verreault, J., Maisonneuve, F., Dietz, R., Sonne, C., Letcher, R.J., 2009. Comparative hepatic activity of xenobiotic-­‐metabolizing enzymes and concentrations of organohalogens and their hydroxylated analogues in captive Greenland sledge dogs (Canis familiaris). Environ. Toxicol. Chem. 28, 162-­‐172. Verreault, J., Letcher, R.J., Ropstad, E., Dahl, E., Gabrielsen, G.W., 2006. Organohalogen contaminants and reproductive hormones in incubating glaucous gulls (Larus hyperboreus) from the Norwegian Arctic. Environ. Toxicol. Chem. 25, 2990-­‐2996. Vollrath, F., Selden, P.A., 2007. The role of behavior in the evolution of spiders, silks, and webs. Ann. Rev. Ecol. Svol. Syst. 38, 819-­‐846. Walker, C.H., Hopkin, S.P., Sibly, R.M., and Peakall, D.B. 2006. Principles of Ecotoxicology. Taylor and Francis Group. Walters, D.M., Fritz, K.M., Otter, R.R., 2008. The dark side of subsidies: Adult stream insects export organic contaminants to riparian predators. Ecol. Appl. 18, 1835-­‐1841. 45
Walters, D.M., Mills, M.A., Fritz, K.M., Raikow, D.F., 2010. Spider-­‐mediated flux of PCBs from contaminated sediments to terrestrial ecosystems and potential risks to arachnivorous birds. Environ. Sci. Technol. 44, 2849-­‐2856. Wiberg, K., Letcher, R., Sandau, C., Duffe, J., Norstrom, R., Haglund, P., Bidleman, T., 1998. Enantioselective gas chromatography mass spectrometry of methylsulfonyl PCBs with application to arctic marine mammals. Anal. Chem. 70, 3845-­‐3852. Wong, C.S., Garrison, A.W., Smith, P.D., Foreman, W.T., 2001. Enantiomeric composition of chiral polychlorinated biphenyl atropisomers in aquatic and riparian biota. Environ. Sci. Technol. 35, 2448-­‐2454. Yamamoto, H., Yoshimura, H., 1973. Metabolic studies on polychlorinated biphenyls—3. Complete structure and acute toxicity metabolites of 2,4,3’,4’-­‐
tetrachlorbiphenyl. Chem Pharma Bull 21, 2237-­‐2242. 46
CHAPTER TWO CHIRAL SIGNATURES OF PCBS IN RIPARIAN SPIDERS Introduction Spiders along Twelvemile Creek and at Lake Hartwell have a PCB body burden, and they redirect transport of these contaminants from the aquatic to the terrestrial ecosystems (Walters et al., 2008; Walters et al., 2010; Raikow et al., 2011). Their body burden is sufficient to pose significant threat to predators (Walters et al., 2010); however, biotransformation of PCBs within spiders has not been investigated. Using enantioselective chromatography as a tool, it may be possible to discern if the changes between trophic levels (in our study, between emergent insects and spiders) is reflected only by changes in concentration, or if there are also enantioselective changes in specific congeners that are transferred. Spiders may be able to regulate their exposure via enantioselective uptake or retention of PCB congeners. Seventy-­‐eight of the 209 PCB congeners are chiral (Lehmler et al., 2009). That is, their asymmetric substitution patterns of chlorines about the biphenyl rings creates non-­‐planar conformations, and generates two molecules with the same elemental composition that cannot be superimposed upon each other. These atropisomers, called enantiomers, were released into the environment in equal parts, known as racemic mixtures. Enantiomers possess the same physical and chemical properties, but differ in their interactions with biological molecules such 47
as enzymes and the direction they orient polarized light. Biological interactions such as enzymatic degradation, enantioselective uptake, and enantiospecific metabolism can change the ratios and provide insight as to the mechanisms of PCB fate and transport. Such interactions can yield different adverse and toxicological effects for individual enantiomers. Enantiomers have different activities, affinities and efficacies for interacting with receptors and enzymes, resulting in different effects and even varying toxicities (Kallenborn and Hühnerfuss, 2001). The Three-­‐
Point-­‐Attachment model finds that enantiomeric specificity on the part of enzymes or receptors occurs only when there are three nonequivalent binding sites to discriminate between the enantiomers (Easson and Stedman, 1933), causing enantioselective toxicities. Diastereomers may have agonist or antagonist functionality, or participate in enantioselective active transport, such as is seen with α-­‐hexachlorocyclohexane (α-­‐HCH) and the blood-­‐brain barrier (Hühnerfuss et al., 1992). Nineteen chiral PCBs are environmentally stable and feature three or four ortho-­‐chlorine atoms that create the non-­‐planar conformation (Lehmler et al., 2009). Twelve of those were commonly used in commercial mixtures and are frequently detected in various environmental media: PCB 45, 84, 88, 91, 95, 132, 136, 144, 149, 171, 174, and 183 (see Table B.1 for substitution patterns) (Frame et al., 1996; Dang, 2007; Dang et al., 2010). A few of these twelve chiral congeners have been investigated in various aquatic food webs, where fish predators such as bass and salmon were found to have non-­‐racemic EF values (EF≠0.5) although the 48
compounds were released as racemic mixtures (EF=0.5) (Dang 2007; Dang et al. 2010; Wong et al. 2001; Buckman et al. 2006). Ratios of chiral diastereomers are quantified by the enantiomeric fraction (EF), defined by Harner at al. (2000) as: !
𝐸𝐹 = !!! = (!)
! ! (!)
(Equation 4) where A represents the first eluting enantiomer and B the second eluting enantiomer, or where (+) represents the clockwise optically directed enantiomer and (-­‐) the counterclockwise orientation. An equal concentration of both enantiomers yields an EF value of 0.5, and is called racemic. Nonracemic EF values are those that differ from 0.5 in either the positive or negative direction. PCB chirality has yet to be investigated in terrestrial predators linked to aquatic food webs. The Twelvemile Creek / Lake Hartwell Superfund site has been the location of many monitoring and research efforts. Walters et al. (2010) were able to use isotopic analysis of carbon and nitrogen to determine the trophic placement of Tetragnathidae, Araneidae, and basilica spiders in the Twelvemile Creek / Lake Hartwell riparian ecosystem. Tetragnathid and shoreline spiders consumed mainly aquatic insects, and upland araneid spiders consume primarily terrestrial prey (Table 1.1). Assessment of how distance from the shoreline affected spider PCB body burdens indicated that with increasing distance the concentration decreases, falling below detection limits at 30m (Raikow et al., 2011). The resultant annals of data cover sediment, benthic organisms, aquatic insects, and fish PCB 49
concentrations and chirality; however, only the most recent investigations have begun to describe how terrestrial predators are influenced by aquatic contamination (Walters et al., 2008; Walters et al., 2010). The goal of this study was to investigate patterns in chirality among three spider taxa and their prey. The hypothesis tested was that the three spider species show distinct EF values based on their consumption of aquatic insects. The three spider taxa have distinct ecological roles that may influence their prey, and as a result, their EF values. Different values may also be influenced by differences among taxa in metabolism of chiral congeners. Additionally, it was hypothesized that comparison with prey species EF values would demonstrate that Tetragnathidae samples would be distinct from their prey sources. Biotransforming processes may differ between prey and predator, and may manifest as changes in EF value. Additionally, changes in bacterial colonies along the exposure gradient may affect the concentrations of each enantiomer available to be transported through the food web. Materials and Methods Sample Collection Spiders were obtained from along the Twelvemile Creek arm of Lake Hartwell between June and August 2007 as described by Walters et al. (2010). Samples were composites of individuals collected at night from either vegetation suspended over the water column or within two meters of shoreline. Riparian spider taxa Tetragnathidae, Araneidae, and basilica were collected, and extracted by the EPA lab 50
in Cinncinati, OH. Adult chironomids were also collected at night from boats trolled across the width of the lake using florescent lights and sweep nets. Sample sizes are listed in Appendix B, Table B.2. Sampling locations are depicted in Figure 2.1, and their GPS coordinates are listed in Table 2.1. Extractions were performed by the EPA. Samples were mixed with 25g of baked sodium sulfate, dried for one hour, and transferred into Accelerated Solvent Extractor (ASE) cells (Dionex, Sunnyvale, CA). The extraction was done using methylene chloride and hexanes. Figure 2.1 EPA Sampling Locations. Gradient map from Walters et al. (2010). Color gradient indicates contamination of sediment (center) and Tetragnathidae spiders (ribbons at either side, indicating banks of lake). 51
Table 2.1 GPS coordinates for EPA sampling sites Site Label Coordinates Site Label Coordinates T12 N 34.734°, W 82.811° I N 34.709°, W 82.835° O N 34.728°, W 82.813° H N 34.703°, W 82.841° N N 34.725°, W 82.819° T6 N 34.697°, W 82.842° K N 34.714°, W 82.829° G N 34.688°, W 82.851° Hexanes and isooctane were obtained as pesticide grade from Fisher Scientific, as well as sulfuric acid. Standards to confirm the identity of field samples for chiral analysis were obtained from AccuStandards (New Haven, CT): PCB 91 (2,2’,3,4’,6-­‐pentaCB), PCB 95 (2,2’,3,5’,6-­‐pentaCB), PCB 136 (2,2’,3,3’,6,6’-­‐hexaCB), PCB 149 (2,2’,3,4’,5’,6-­‐hexaCB), and PCB 174 (2,2’,3,3’,4,5,6’-­‐heptaCB). Additional standards were obtained for use in achiral analysis: Aroclor mixtures 1016, 1242, and 1260 for instrument calibration; PCBs 14 (3,5-­‐diCB) and PCB 169 (3,3’,4,4’,5,5’-­‐
hexaCB) for use as recovery standards; and aldrin and PCB 204 (2,2’,3,4,4’,5,6,6’-­‐
octaCB) for use as internal standards. Lipid content in extracts was measured in our lab by a gravimetric method. Extracts were treated with 4mL of an ethanolic KOH solution to extract any metabolites present in the extracts (Kania-­‐Korwel et al., 2008b). The KOH solution was removed and used for metabolite analysis, which is described in Chapter 3. 52
Sample Clean-­‐up Organic materials that can cause interference in the chromatograms were removed through use of cleanup columns. Alumina – sodium sulfate columns (2.5g and 1.5g respectively) were conditioned with pesticide grade hexanes to remove any interfering polar compounds. Alumina was 6% deactivated. Extracts received from the EPA were added to the columns, and eluted with hexanes. Extracts were further cleaned by centrifugation with sulfuric acid to remove any remaining lipids in the sample, then solvent exchanged for isooctane (U.S. EPA, 2005). Chiral Analysis Enantioselective analysis of five chiral congeners was conducted with an Agilent 6850-­‐GC equipped with a Chirasil-­‐dex DB column (25m x 0.25mm) and a 63Ni electron detector. The temperature program began at 100˚C for 2min, then ramped at 10˚C/min to 170 and held for 10min, then ramped at 8˚C/min to 180˚C and held for 20min, and finally ramped at 1˚C/min to 190˚ and held for 5min. Quality of assessment was maintained through use of procedural blanks, laboratory control samples in duplicate, matrix spikes in duplicate (aldrin and PCB 204, Accustandard), blank tissue samples, and check standards with each batch of 10 samples. Concentrations were not corrected for recovery. The detection limit did vary somewhat by congener, but was approximately 25ng/L for chiral PCB congeners. The EFs for the racemic standards (PCBs 91, 95, 136, 149, and 174) were between 0.492 (± 0.003 SD) and 0.504 (±0.006 SD). 53
Statistical Analysis Differences in EF values among different species were assessed by one-­‐way analysis of variance (ANOVA) using Tukey’s test to determine significance (p<0.05). This assessed the significance of evidence gathered for the hypothesis that EF values would differ among spider taxa. Analysis of co-­‐variance (ANCOVA) was enlisted to determine if there were changes in spider EF among sampling locations. EF values were considered to be significantly non-­‐racemic if their 95% confidence intervals did not encapsulate a racemic value of 0.5. Nonmetric-­‐multidimensional scaling statistical analysis (NMS) was completed using PC-­‐ORD 4.0 software (MjM software). NMS is best utilized when data are not normally distributed, or have discontinuous and unequal scales. The autopilot mode was utilized such that the program would choose the best solution at each dimensionality (McCune and Mefford, 1999). The chi squared coefficient was used as the distance measure in the analysis, as it is the most appropriate measure enlisted when the domain of data inputted is x ≥ 0 and the range of distance measures is d ≥ 0 when d = f(x) (McCune et al., 2002).
Results EF Differences among Spider Taxa Four chiral congeners (PCBs 91, 95, 149, and 174) were above the detection limit for entantioselective analysis in the three spider taxa (Figure 2.2). Chiral signatures for congener 91 were significantly non-­‐racemic in favor of the first eluting congener. 54
The average EF (± 1 SD) for Tetragnathidae spiders was 1.0 (± 0.0); no Tetragnathidae samples contained the second-­‐eluting enantiomer above the detection limit. Araneidae spiders were also significantly non-­‐racemic with an average EF value of 0.90 (± 0.22 SD), and basilica samples also had an average EF of 0.90 (±0.17 SD). These EF values were significantly non-­‐racemic across all taxa (confidence interval, α=0.05). Enantiomeric fractions for PCB 91 did not vary significantly among taxa (ANOVA, p = 0.16) when data were averaged across all sample sites. An ANCOVA analysis for PCB 91 enantiomeric fractions along the exposure gradient among sites for each taxa indicated that the EF did not change significantly (α=0.05, F=0.1, p-­‐value = 0.905; Figure 2.2A). 55
Figure 2.2 EF measurements averaged across sites. Error bars represent 95% confidence interval. Data for spider taxa represent congeners (A) PCB 91, (B) PCB 95, (C) PCB 149, and (D) PCB 174. Panel (E) displays Midge EF data averaged across all sampling locations. Samples containing PCB 95 had detectable concentrations of both enantiomers (Figure 2.2B), in contrast to PCB 91 which was skewed in favor of one enantiomer. Mean (± 1 SD) EFs were 0.23 (± 0.12), 0.20 (± 0.08), and 0.31 (± 0.14) for tetragnathid, areaneid, and basilica spiders respectively. These values were 56
clearly non-­‐racemic, but were not different from one another. (ANOVA, p-­‐value = 0.06) Likewise, EFs did not vary spatially along the exposure gradient (ANCOVA, α=0.05, F=1.24, p = 0.301; Figure 2.2B). Samples containing congener 149 had approximately equal concentrations of both congeners. Tetragnathidae spiders had an EF value of 0.55 (± 0.14 ); whereas, Araneidae spiders had a value of 0.49 (± 0.18 ), and basilica spiders had an EF value of 0.44 (± 0.05). Only for basilica spiders was this value significantly different from racemic values (95% confidence interval 0.46 ± 0.03, Figure 2.2C). It is important to note that the concentration of PCB 149 in basilica spiders fell below detection limits for enantioselective analysis for all but two sampling sites. EF values did not differ among taxa (ANOVA, p-­‐value = 0.33), or among sampling locations (ANCOVA, α=0.05, α=0.05, F=0.58, p-­‐value=.567; Figure 2.2C). Congener 174 had an average EF value of 0.92 (± 0.05), 0.85 (± 0.09), and 0.83 (±0.24) for Tetragnathidae, Araneidae, and basilica spiders, respectively. Confidence intervals indicate that all three species were significantly non-­‐racemic in favor of the first eluting enantiomer. No significant differences existed among the species (ANOVA, p-­‐value = 0.14). ANCOVA analysis did not reveal any significant differences in EF along the exposure gradient for PCB 174 (α=0.05, F=0.56, p-­‐
value=.577; Figure 2.2D). 57
EF Differences between Prey and Spiders Differences between the EFs of the spider taxa and aquatic insects that serve as prey for the spiders were measured by considering Chironomidae (midges). Midges were collected from the same sites, and represent potential prey for the spiders; however, only three of the congeners investigated were above the detection limit. Midge samples had fairly consistent EF values across all sampling locations (Figure 2.2E) that were distinct from those for spiders, except for PCB 95 which is enriched in the same enantiomer (second-­‐eluting) as the midge sample. EF values had a wide range for different congeners. PCB 91 had an EF of 0.27 (± 0.05 SD), whereas PCB 95 was close to racemic with an EF of 0.43 (±0.04 SD). PCB 149 has the greatest EF value, 0.74 (± 0.07 SD), and was the only congener with higher concentrations of the second eluting enantiomer. A Student’s T test comparing midges to each spider taxa determined that midge EF values for PCB 95 were significantly different from Tetragnathidae, Araneidae, and basilica spiders (α=0.05, p-­‐valuetetragnathid = 4.0e-­‐6, p-­‐valuearaneid = 2.0e-­‐8, p-­‐valuebasilica = 0.02). Use of NMS to discern trends in the data revealed that for PCB 95 and 149 spider EF values are distinct from those of Midges. For the plot shown, spider and midge samples from all species and locations were incorporated (Figure 2.3). Within the groups the relationships were less clear. Midge samples, while grouped together, did not indicate any spatial trends. Points that fall outside of the circles all represent spider samples. Many of the spider samples were difficult to assess using 58
this analysis because of their retention of only one enantiomer, thus they fell outside of the general area where the majority of spider samples clustered. Figure 2.3 Nonmetric Multidimensional Scaling (NMS) plots of EFs for PCBs 95 and 149 for both spiders and midges. Spiders and midges are clearly grouped for these two congeners. Points that fall outside of the circles represent spiders. 59
Discussion The EF results did not support the hypothesis that the three spider taxa would differ due to differences in their consumption of aquatic insects. ANOVA analysis showed that average EF values for the three spider taxa were not statistically distinct from each other. Isotopic analysis by Walters et al. (2010) indicated that riparian spiders consumed aquatic insects. In particular, the tetragnathid and araneid spiders consumed mostly aquatic prey; whereas, basilica spiders consumed approximately half aquatic prey and half terrestrial prey. All of the samples analyzed were obtained within 2m from the shoreline, and so all would opportunistically prey upon emergent aquatic insects. Terrestrial insects were observed to have low PCB concentrations and are not likely to contribute to EF. Spider enantiomeric fractions are likely very similar because their prey source that contains PCBs is very similar. Using this theory as a guide to interpret the data presented here, these spiders may have similar enzymatic capabilities and different habitat spaces; therefore, the greatest influence on contaminant profile is prey consumption, not biotransformation. Chironomidae were selected for comparison with spiders because they commonly dominate the aquatic insect assemblages (Fairchild et al., 1992; Menzie,
198; Maul et al., 2006). They also are capable of transporting significant quantities of carbon, nitrogen,and PCBs in the Twelvemile Creek watershed (Walters et al., 2010). They are a good representative of prey because spiders are opportunistic predators that feed upon what their webs can capture. Additional enantioselective 60
analyses could be performed on other aquatic insect prey to corroborate the comparisons discussed here. Midges are compared to Tetragnathidae spiders because those spiders are obligate aquatic predators, so their EF values should most closely reflect the influence of aquatic emergent insects. Prey (midge) EFs and predator (spider) EFs were proven significantly different for all three PCB congeners – 91, 95, and 149 – via Student’s T test. Midge samples were nonracemic favoring the second eluting enantiomer for PCBs 91 and 95. Congener 149 was non-­‐racemic favoring the first eluting enantiomer. Tetragnathidae spiders were nonracemic favoring the first eluting enantiomer for PCB 91. The change for PCB 91 between chironomids and tetragnathids, shifting towards favoring the first eluting enantiomer, could be due in part to selective uptake or excretion by spiders. If the spiders do not uptake a great amount of the second eluting enantiomer, or if they can excrete it efficiently, the EF value would subsequently be greater than 0.5. Metabolism could also yield these results. If spiders are able to more easily metabolize, and therefore, preferentially excrete the second eluting enantiomer the EF value would shift in favor of the first eluting enantiomer. PCB 95 was nonracemic in both organisms, with prey and predator both retaining the second eluting enantiomer in greater concentration than the first enantiomer. The EF became more enriched in the second-­‐eluting enantiomer with the change in trophic level from prey to predator. For PCB 149, midge samples were nonracemic favoring the first eluting enantiomer and Tetragnathidae samples were racemic. These data for PCB 95 and 149 indicate that enantioselective metabolism 61
by spiders is likely not occurring as it may for PCB 91. PCB 149 may also be experiencing enantioselective metabolism as the value changes from prey to predator. Walters et al. (2008) determined that midges and other emergent insects were not good indicator species when compared to local sediment contamination concentrations. It was proposed that due to the insects’ low mass prevailing winds could force the populations upstream or downstream at any given time, which allows for mixing of the populations. Even if the midge populations do vary in enantiomeric body burden, it would be difficult to discern from field sampling due to population mixing. The lack of statistical significance in EF values for Midge samples along the contamination gradient supports the ideas that mixing of emergent insect populations is occurring. Confirmation that the system has not changed such that sediment EF is identical along what was previously considered to be a gradient would need to be obtained in order to assert that population mixing is occurring. Assuming this to be true, all spider species consuming those mixed populations would receive the same average enantiomeric mixtures. It can be concluded, therefore, that because the enantiomeric fractions do not vary among the spider species that there is no significant difference in contaminant processing among these three riparian species for PCB congeners 91, 95, 149, and 174. Dang et al. (2010) investigated the riparian food web in the Lake Hartwell tributary of Twelvemile Creek. Analysis of the mayfly did not show any statistical differences among site, providing further evidence for the “mixing” theory proposed 62
by Walters et al. (2010) and supported by midge data herein. EF values were reported for PCB 91 (0.51 ± 0.03), PCB 95 (0.24 ± 0.03), and PCB 149 (0.35 ± 0.04). These differ from the data obtained for midges in Lake Hartwell. Such differences could be a result of the locations having a somewhat different contamination profile. The EFs for the spider taxa examined in this study do differ from data reported for aqueous predators in the same watershed. Yellowfin shiner were analyzed for EF in Twelvemile Creek by Dang et al. (2010) assuming that they were predators of mayflies. Yellowfin were found to have EF values for PCB 91 and 149 that were also nonracemic in favor of the second eluting enantiomer; whereas, PCB 95 was not above the detection limit. Spider concentrations can be as many as three times higher than those reported for fish (Walters et al. 2010), so they may have a sufficient concentration of PCB 95 for enantioselective analysis when yellowfin do not. Spider and fish EF values were skewed in different directions for PCB 91, but values for PCB 149 were much more similar. This again may indicate that spiders experience enantioselection for the first eluting enantiomer of PCB 91. A more detailed comparison of diet for spiders and yellowfin would need to be completed to determine the difference in values is not a result of different dietary composition. It was concluded by Dang et al. (2010) that the nonracemic values for Yellowfin were due to consumption of nonracemic residues. This same conclusion holds for Tetragnathidae. The difference in the EF values between the terrestrial and aquatic predators could also be attributed to diet. Spiders can only consume insects that emerge from the water column, whereas fish can consume the larval 63
instars, too. Insects that emerge will not acquire more PCB mass, but growth may dilute the concentration, or metabolism affect the EF. If metabolism of the residues on the part of the insects changes over the course of their life, the spiders would be a better representation of the EF for adult insects after they emerge, and shiner values would reflect an averaging of EF over the life of the insects prior to emersion from the water column. The lack of spatial trends for Midge EFs is in accordance with previously published work (Walters et al., 2010). It has been theorized that their low mass causes these insects to be easily blown up and downstream, causing them to be poor indicators of local PCB sediment and food web concentrations. The NMS data supports this claim, as well as indicates clear differences in the values for spiders and midges (Figure 2.3). Enantioselective processes that change the EF as PCBs move up the food chain would yield such a plot. Conclusions The hypothesis that each spider taxa would be distinct from the others was not supported by the data collection. Tetragnathids, araneids, and basilica spiders had essentially the same EF for each of the congeners; therefore, difference in niche does not appear to influence the EF profile in spiders. Additionally, we can conclude that if the spiders are capable of metabolism, the three taxa use the same mechanism. 64
Comparison of EF values between spider samples and midge samples also supported the hypothesis that predator EF values would be distinct from prey EF values. Spider values shifted in the direction favoring retention of the second eluting enantiomer for PCB 91. The EF data for PCBs 95 and 149 are not as clear. The change in EF could be a product of metabolism on the part of the spider. If one enantiomer is preferentially detoxified the EF value would shift to favor the more recalcitrant enantiomer. References Buckman, A.H., Wong, C.S., Chow, E.A., Brown, S.B., Solomon, K.R., Fisk, A.T., 2006. Biotransformation of polychlorinated biphenyls (PCBs) and bioformation of hydroxylated PCBs in fish. Aquatic Toxicology 78, 176-­‐185. Dang, V.D., Walters, D.M., Lee, C.M., 2010. Transformation of chiral polychlorinated biphenyls (PCBs) in a stream food web. Environ. Sci. Technol. 44, 2836-­‐2841. Dang, V.D., 2007. Achiral and chiral analysis of polychlorinated biphenyls (PCBs) in the aquatic and riparian food webs in Twevle Mile Creek, South Carolina. MS Thesis. Clemson University. Easson, L.H., Stedman, E., 1933. Studies on the relationship between chemical constitution and physiological action. V. Molecular dyssymmetry and physiological activity. Biochem J 27, 1257-­‐1266. Fairchild, W.L.; Muir, D.C.G.; Curries, R.S.; Yarechewski, A.L. 1992. Emerging
insects as a biotic pathway for movement of 2,3,7,8-tetrachlorodibenzofuran
from lake sediments. Environ. Toxicol. Chem. 11, 867-872
Frame, G.M., Cochran, J.W., Boewadt, S.S., 1996. Complete PCB congener distributions for 17 Aroclor mixtures determined by 3 HRGC systems optimized for comprehensive, quantitative, congener-­‐specific analysis. J. High Resol. Chrom. 19, 657-­‐668. 65
Harner, T., Wiberg, K., Norstrom, R., 2000. Enantiomer fractions are preferred to enantiomer ratios for describing chiral signatures in environmental analysis. Environ. Sci. Technol. 34, 218-­‐220. Hühnerfuss, H., Kallenborn, R., Konig, W.A., Rimkus, G., 1992. Preferential enrichment of the (+)-­‐α-­‐hexachlorocylcohexane in cerebral matter of harbour seals. Organohal. Compds. 10, 97-­‐100. Kallenborn, R., Hühnerfuss, H., 2001. Chiral Environmental Pollutants: Trace Analysis and Ecotoxicology. Springer. Kania-­‐Korwel, I., Hrycay, E.G., Bandiera, S.M., Lehmler, H., 2008a. 2,2 ',3,3 ',6,6 '-­‐ hexachlorobiphenyl (PCB 136) atropisomers interact enantioselectively with hepatic microsomal cytochrome P450 enzymes. Chem. Res. Toxicol. 21, 1295-­‐1303. Kania-­‐Korwel, I., Zhao, H., Norstrom, K., Li, X., Hornbuckle, K.C., Lehmler, H., 2008b. Simultaneous extraction and clean-­‐up of polychlorinated biphenyls and their metabolites from small tissue samples using pressurized liquid extraction. Journal of Chromatography A 1214, 37-­‐46. Lehmler, H.J., Harrad, S.J., Huhnerfuss, H., Kania-­‐Korwel, I., Lee, C.M., Lu, Z., Wong, C.S., 2009. Chiral polychlorinated biphenyl transport, metabolism, and distribution: A review. Environ. Sci. Technol. 44, 2757-­‐2766. Letcher, R.J., Klasson-­‐Wehler, E., Bergman, Å, 2000. Methyl Sulfone and Hydroxylated Metabolites of Polychlorinated Biphenyls, in: The Handbook of Environmental Chemistry Vol. 3 Part K New Types of Persistent Halogenated Compounds. Paasivirta, J. (Ed.). Springer-­‐Verlag, pp. 315 -­‐ 359. Maul, J.D.; Belden, J.B.; Schwab, B.A.; Whiles, M.R.; Spears, B.; Farries, J.L.;
Lydy, M.J. 2006. Bioaccumulation and trophic transfer of polychlorinated
biphenyls by aquatic and terrestrial insects to tree swallows (Tachycineta
bicolor). Environ. Toxicol. Chem. 25, 1017-1025.
McCune, B., Grace, J.B., Urban, D.L., 2002. Analysis of Ecological Communities. MjM Software Design, Gleneden Beach, OR. McCune, B., Mefford, M.J., 1999. PC-­‐ORD. Multivariate Analysis of Ecological Data, Version 4. MjM Software Design, Gleneden Beach, OR. Menzie, C.A. 1980. Potential significance of insects in the removal of contaminants
from aquatic systms. Water, Air Soil Pollut. 13, 473-479.
66
Raikow, D., Walters, D., Fritz, K., Mills, M., 2011. The distance that contaminated aquatic subsidies extend into lake riparian zones. Ecol. Appl. 21, 983-­‐990. Riechert, S.E., 1981. The consequences of being territorial: Spiders, a case study. Am. Nat. 117, 871-­‐892. Safe, S.H., 1994. Polychlorinated-­‐biphenyls (PCBs) -­‐ Environmental-­‐impact, biochemical and toxic responses, and implications for risk assessment. Crit. Rev. Toxicol. 24, 87-­‐149. U.S. EPA, 2005. Method 3545: Standard Operating Procedure for Accelerated Solvent Extraction (ASE-­‐200) of Organics. EPS, Washington, DC. Walters, D.M., Fritz, K.M., Otter, R.R., 2008. The dark side of subsidies: Adult stream insects export organic contaminants to riparian predators. Ecol. Appl. 18, 1835-­‐1841. Walters, D.M., Mills, M.A., Fritz, K.M., Raikow, D.F., 2010. Spider-­‐mediated flux of PCBs from contaminated sediments to terrestrial ecosystems and potential risks to arachnivorous birds. Environ. Sci. Technol. 44, 2849-­‐2856. Warner, N.A., Martin, J.W., Wong, C.S., 2009. Chiral polychlorinated biphenyls are biotransformed enantioselectively by mammalian cytochrome P-­‐450 isozymes to form hydroxylated metabolites. Environ. Sci. Technol. 43, 114-­‐
121. Wong, C.S., Hoekstra, P.F., Karlsson, H., Backus, S.M., Mabury, S.A., Muir, D.C.G., 2002. Enantiomer fractions of chiral organochlorine pesticides and polychlorinated biphenyls in standard and certified reference materials. Chemosphere 49, 1339-­‐1347. Wong, C.S., Garrison, A.W., Smith, P.D., Foreman, W.T., 2001. Enantiomeric composition of chiral polychlorinated biphenyl atropisomers in aquatic and riparian biota. Environ. Sci. Technol. 35, 2448-­‐2454. Wu, X., Pramanik, A., Duffel, M.W., Hrycay, E.G., Bandiera, S.M., Lehmler, H.J., Kania-­‐ Korwel, I., 2011. 2,2′,3,3′,6,6′-­‐Hexachlorobiphenyl (PCB 136) is enantioselectively oxidized to hydroxylated metabolites by rat liver microsomes. Chem. Res. Toxicol. 24, 2249-­‐2257. 67
CHAPTER 3 SPIDER METABOLISM OF PCBS: IN VIVO AND IN VITRO INVESTIGATIVE APPROACHES Introduction Toxicants, or parent compounds, are metabolized in an effort to make them easier to excrete; however, even at low concentrations metabolites have a marked effect. Hydroxylated PCBs (OH-­‐PCBs) in particular are more toxic than their parent compounds, especially with respect to endocrine disruption (Purkey et al., 2004). OH-­‐PCBs have a structure similar to that of thyroxine (T4), a thyroid hormone that is transported by transthyretin (TTR). The metabolite has a higher affinity for TTR than does T4, which alters response by the endocrine system function (Brouwer et al., 1998). T4 is effective at concentrations of approximately 10pM in humans, so even at low concentrations OH-­‐PCBs can be potent toxicants (Lans et al., 1990). Different chlorination patterns of PCBs lead to different volatilities (Mackay et al., 1992a), potentials for biomagnification (Campfens and Mackay, 1997;Mackay and Fraser, 2000;Letcher et al., 2010), and potentials for metabolism (Schurig et al., 1995). Congeners that feature chlorine with adjacent hydrogens are easier to metabolize, and chlorines in meta-­‐ or para-­‐positions are easier to remove or replace with hydroxyl groups than are ortho-­‐Cl (Letcher et al., 2000). Some congeners are vulnerable to biotransformation to lower chlorinated congeners via reductive dechlorination by bacteria (Pakdeesusuk et al., 2003b;Pakdeesusuk et al., 68
2003a;Pakdeesusuk et al., 2005), or can be metabolized into compounds such as methlysulfonyl-­‐PCBS (MeSO-­‐PCBs), methoxy-­‐PCBs (MeO-­‐PCBs), and hydroxylated PCBS (OH-­‐PCBs) (Letcher et al., 2000;Lehmler et al., 2009). While accumulation of OH-­‐PCBs has been demonstrated in fish (Buckman et al., 2006;Carlson and Williams, 2001), birds (Klasson-­‐Wehler et al., 1998;Verreault et al., 2006;Helgason et al., 2010;Jaspers et al., 2008), frogs (Nomiyama et al., 2010),and mammals (Park et al., 2009;Verreault et al., 2009;McKinney et al., 2006;Hoekstra et al., 2003;Sandau et al., 2000;Guvenius et al., 2002), there has been little research on invertebrate predators. For example, the Twelvemile Creek / Lake Hartwell ecosystem has been the focus of many sediment (Pakdeesusuk et al., 2003b;Pakdeesusuk et al., 2003a;Pakdeesusuk et al., 2005) and fish investigations (Wong et al., 2001); however, recent analyses have identified that non-­‐fish predators in the riparian zone also participate in contaminant transport (Walters et al., 2008;Walters et al., 2010). Spiders that consume aquatic emergent insects have PCB concentrations greater than 5.5ppm (Walters et al. 2010). If other predators in the food web accumulate and metabolize their PCB body burden it is possible that spiders can do the same. It is valuable to know if spiders have the metabolic capacity to metabolize their PCB body burden because they can transfer those metabolites alongside the carbon, nitrogen, and PCBs to their predators, namely frogs, lizards, and birds (Walters et al., 2010). If this is so, then ecological risk assessments may need to be expanded to incorporate OH-­‐PCBs and invertebrates. 69
The goal of this work is to test two hypotheses. First, spiders have the capacity to metabolize PCBs to produce OH-­‐PCBs. Second, metabolism of co-­‐planar PCB 61 will differ from metabolism of the non-­‐planar congeners. The metabolism study had four objectives to explore the hypotheses: •
Screening for OH-­‐PCB metabolites generated in vivo by spiders •
An in vitro exposure of spider enzymes to non-­‐planar PCBs 88 and 149 •
An in vitro exposure of spider enzymes to co-­‐planar PCB 61 •
Use of biomarker assays to determine the metabolic pathway for PCB detoxification Materials and Methods Sample Collection Extracts from samples collected along the Twelvemile Creek arm of Lake Hartwell were provided to the lab by the US EPA and used for the in vivo study. These samples have previously been analyzed for PCB concentration (see Walters et al. 2010) and for enantiomeric fraction trends (see Chapter 2 of this dissertation for EF data and more detail about collection). The extracts were treated to remove OH-­‐
PCBs prior to the chiral analysis in Chapter 2, and the metabolite fraction used to screen for presence of metabolites in spiders from the field. The in vitro study utilized an enzyme fraction isolated from Tetragnathidae spiders obtained from two sites along Twelvemile Creek (Figure 3.1). The collection sites are contaminated by PCBs from the Sangamo-­‐Weston point source, and are 70
locations where enzymatic activities are hypothesized to be elevated because of exposure to contamination. Spiders were collected in jars and frozen with liquid nitrogen before transport back to the lab. Each site had one pooled sample of approximately 15 spiders, and each sample had a total organism mass of 1.2g. Additionally, one pooled sample was collected from the Reese Mill site without liquid nitrogen, which was analyzed separately than the sample obtained with liquid nitrogen. Samples were stored in a -­‐80˚F freezer until preparation for analysis. 71
Figure 3.1 Sampling locations. RM = Reese Mill, RB = Robinson Bridge. Note that two types of samples were collected at Reese Mill: spiders collected with liquid nitrogen, and spiders collected without nitrogen. Sample preparation for in vivo study Extractions were performed by the US EPA. Samples were mixed with 25g of baked sodium sulfate, dried for one hour, and transferred into Accelerated Solvent Extractor (ASE) cells (Dionex, Sunnyvale, CA). The extraction was done using methylene chloride and hexanes. Analysis of these extracts for enantiomeric fraction values is discussed in Chapter 2. Hydroxylated recovery standards were not used for the in vivo experiments due to uncertainty about elution patterns of metabolites. These extracts were treated with an ethanol:KOH solution (1:1, v/v) to deprotonate hydroxylated metabolites such that they would partition from the neutral organic phase. The aqueous layer was drawn off, and treated with HCl to pH 72
4 in order to reprotonate the metabolites that partitioned into the aqueous layer. A hexanes:methyl-­‐tert-­‐butyl-­‐ether solution (9:1, v/v) was used to extract metabolites from the aqueous phase. The organic phase was derivitized with diazomethane that was synthesized in our lab. Sample clean-­‐up columns are described below. S9 fraction preparation The in vitro studies used post-­‐mitochondrial, or S9, fractions. An S9 fraction is the supernatant obtained after centrifuging organ homogenant, and contains both microsomes and cytosol (Greim and Snyder, 2008). The microsomal portion contains the cytochrome P-­‐450 enzymes (CYP) responsible for phase I metabolism, and the cytosolic portion contains various other enzymes involved in phase II metabolism. For all components the samples were derivitized and analyzed in identical manners, though sample preparation differed. Homogenization buffer was prepared to maintain pH, to prevent ice crystal damage, to inhibit proteases that would destroy target enzymes, and to prevent oxidation (0.25M sucrose, 0.05M Tris-­‐base at pH 7.4, 1mM ethylenediaminetetraacetic acid [EDTA, to chelate metal proteases], 1mM dithiothreitol [DTT, to prevent oxidation], 0.2mM α-­‐toluene sulfonyl fluoride [PMSF, to inhibit proteases]). Whole body samples of Tetragnathidae spiders were homogenized with 5mL of buffer. A manual tissue tearer was used to homogenate the sample, taking care to avoid any foaming or bubbles, which could oxidize the sample. Fifteen spiders were combined to create a sample weight of approximately 73
1.2g. Samples were adjusted to identical volume and spun down at 9000 rpm (10,000 g) at 4˚C for 20min on a Beckman J2021M/E centrifuge with JA-­‐20.1 rotor. The fat layer was aspirated from the top of the extract, and the supernatant stored at -­‐80˚C until analysis. This protocol was adapted from biomarker assays performed by Truman and van den Hurk (2010). Enzymatic incubations S9 fractions from each site were treated with 200µL of a 2ppm concentration of an individual PCB congener in acetone – PCB 61 (2,3,4,5-­‐tetraCB), 88 (2,2;,3,4,6-­‐
pentaCB), or 149 (2,2;,3,4;,5;,6-­‐hexaCB). Individual neat standards of PCB congeners were obtained from Accustandard (New Haven, CT). The S9 fractions were incubated for 24hrs in a 34˚C water bath. NADPH was added to provide the energy necessary to fuel metabolism. Samples were then removed from the bath and extracted. The enzyme incubations used a negative control, created by boiling a composite S9 fraction, consisting of equal parts from the two sampling sites. Boiling effectively quenches enzymatic activity, and combining enzymes from both locations removes site bias. Rat microsomes enriched in CYP2B was used as a positive control, and a reagent blank was prepared to confirm that positive results were due to compound transformation and were not artifacts from the reagents. Reagent blanks consisted of PCBs with magnesium chloride, bovine serum albumin, Tris buffer at pH 7.4, and NADPH. Enzyme incubations were completed in triplicate. 74
Incubation extraction The post-­‐incubation samples were extracted via accelerated solvent extraction on a Dionex 200 Accelerated Solvent Extractor (ASE). Cells were packed with diatomaceous earth and sand. The in vitro experiment used a 2-­‐OH-­‐tetraCB recovery standard obtained from Accustandard (New Haven, CT) with recoveries of 34% (± 4 SD). These recoveries are similar to those obtained in other laboratories (Kania-­‐Korwell et al., 2008). Results were not corrected based upon percent recovery. A 9:1 hexanes:acetone solution (v/v) was used for the extraction. Heating time was 5min followed by two 2min static cycles. Cells were heated to 100˚C temperature at a pressure of 1500 psi, followed by a 60% volume flush. Extraction methods were adopted from Kania-­‐Korwell et al. (2008b) and Garcia et al. (2008). Derivitization of in vivo and in vitro samples To detect metabolites in a GC-­‐MS system, the hydroxyl group needs to be derivitized to a methoxylated compound. All in vivo samples, S9 extracts, and hydroxylated standards were derivitized in the laboratory with diazomethane that was synthesized immediately before use, and concurrent with extraction procedures. Diazomethane was collected in methyl-­‐tert-­‐butyl ether, from carbitol (diethylene glycol ethyl ether) and diazald (N-­‐methyl-­‐N-­‐nitroso-­‐p-­‐toluene sulfonamide) under alkaline conditions in a glass diazomethane generator. After one hour of synthesis on ice, the ether from the outer compartment was collected, and 75
400μL added to each sample. All samples were then stored overnight at -­‐20˚C before clean-­‐up. Sample clean-­‐up for in vivo and in vitro studies A clean-­‐up column was prepared with 3.5g of alumina (6% deactivated) and 2.5g anhydrous sodium sulfate. Both in vivo and in vitro samples were cleaned up on these columns. The samples were eluted with hexanes to a final volume of 10mL. The solvent was exchanged for isooctane, and samples were condensed to a final volume of 2mL. Sample Analysis by GC-­‐MS-­‐MS Tandem mass spectroscopy was used for analysis of in vivo and in vitro samples to determine the structures of the compounds detected. Use of standards to identify and quantify these compounds would have been ideal; however, obtaining standards for each potential metabolite was not feasible due to both unavailability and cost. Instead, mass spectra were used to determine presence and degree of chlorination for the compounds detected. Different SIM settings were used to screen for metabolites by homolog series (Appendix C). Aliquots of the samples with aldrin and PCB 209 as internal standards (Accustandard) were run on a Varian 4000 GC-­‐MS-­‐MS with ion-­‐trap detection using electron impact ionization (70 eV) with selected ion monitoring (m/z 306, 308, 310 76
for PCB 61; m/z 324, 326, 328 for PCB 88; m/z 358, 360, 362 for 149). An RTX-­‐5 column was used for separation (30m×0.25mm×0.25µm, Restek). Samples were injected in splitless mode at 215˚C with helium as a carrier gas at a constant flow rate of 50.6 cm/s. Injected volumes were high at 7µL because sample concentration fell below detection limits with 1µL injections. Initial oven temperature was 90˚C and held for 2min, then ramped to 180˚C at an 8˚C/min rate, held for 10min, and then ramped at 5˚C/min to 200˚C and held for 10min. Standards for PCB 61, OH-­‐PCB 61, and MeO-­‐PCB 61 were obtained from Accustandard (New Haven, CT). PCB and MeO-­‐PCB 61 were used to develop the quantification method and to externally calibrate enzyme samples. OH-­‐PCB 61 was used to optimize the derivitization of metabolites. 2-­‐MeO-­‐3,4-­‐diCB was used as a recovery standard for these incubations, and recovery averaged 72.0% (±2.3) among samples. Qualitative data about metabolite structure were obtained for PCBs 88 and 149, as no metabolite standards for these congeners are available from commercial sources at this time. Biomarker assays S9 fractions were used in biomarker assays enlisted to help detail the metabolic pathway for PCB detoxification. The EROD and PROD assays work identically, and differ only in their target enzyme: the EROD assay targets the CYP1A isoform; whereas, the PROD assay targets CYP2B. 77
For the EROD assay, a starting solution of Tris buffer (to maintain pH at 7.4), bovine serum albumin (to keep exthoxyresorufin in solution), MgCl2 (to maintain ionic strength), water, and ethoxyresorufin (the substrate) is prepared. Each well of the plate has 100uL of S9 fraction added to 100uL of starting solution. Addition of 50uL of a 2.5mM NADPH solution starts the reaction. A plate reader set at 530nm excitation and 585nm emission measures the fluorescence generated when CYP1A deethylates ethoxyresorufin. The PROD assay uses the same protocol, replacing the exthoxyresorufin with pentoxyresorufin, thus requiring the use of CYP2B to complete depentylation of the substrate. The same plate reader settings are used because the same end product, resorufin, is measured in both arrays. Results Field study for in vivo metabolism The chromatograms for the field samples indicate that many different compounds are present. The extraction and derivitization procedures ensure that only OH-­‐PCBs are being detected (Kania-­‐Korwell et al. 2004); however, many different congeners could be metabolized to yield multiple metabolites for each congener. Incidence of chlorine containing compounds was confirmed by the presence of isotopic ratios for chlorine (Figure 3.2). Such compounds are not naturally occurring. Screening the same samples under different SIM settings allows for structure specificity. There is evidence to suggest that more than one 78
homolog class is transformed (Figure 3.3). SIM settings for hepta, octa, and nona-­‐
chlorinated compounds did not detect any compounds, suggesting that highly chlorinated congeners are not metabolized. Figure 3.2 Tetragnathidae sample for in vivo analysis. The spectra corresponds to the peak indicated in the chromatogram. The scan using SIM settings for m/z=324 for penta-­‐chlorinated metoxylated compounds indicates presence of metabolites. Structure could not be discerned, but presence of isotopic ratios for chlorine (100:30 isotopic peaks) indicate that these compounds are not naturally occurring. 79
Figure 3.3 The same Tetragnathidae sample indicated in the previous figure under
different SIM settings, shown here for m/z=353 to screen for hexachlorinated
congeners. Spectral results for the peak indicated in the chromatogram
confirm that compounds with six chlorines can be metabolized.
Qualitative description of PCB 88 and 149 in vitro metabolism The incubation results indicated enzyme activity. An identical experimental design was used for PCBs 88 and 149. The reagent blanks were identical to the negative controls, and both were significantly different from the samples (see Figure 3.4). Panel A is the chromatogram for the negative control, and panels B, and C are chromatograms for the two sample sites where spiders were obtained with liquid nitrogen. All panels reflect a treatment with PCB 88. In the negative control (Figure 3.4A) there is a peak that elutes at 38min, whereas the samples (Figures 3.4b and 80
3.4C) yielded a peak at 26min. The negative control samples showed peaks that were dissimilar from both the reagent blanks and the samples obtained with liquid nitrogen. The samples obtained with liquid nitrogen have chromatograms with peaks that were not present in the negative controls or the samples obtained without liquid nitrogen. These peaks in the samples collected with liquid nitrogen are well defined, and feature unique mass spectra patterns. 81
Figure 3.4 Chromatograms for PCB 88 enzyme incubations. (A) negative control, (B) Reese Mill, and (C) Robinson Bridge. 82
Additionally, there was good repeatability of results between samples. Figure 3.5 shows chromatograms for samples incubated with PCB 88. Panels A through E indicate that each sample had the peak appear at identical retention times. It is important to note that where the magnitude of the transformation appears to vary between replicate samples obtained at Reese Mill (Panels C, D, and E) the experiment had to be modified for the last two replicates to account for low S9 fraction supplies. For the replicates pictured in panels D and E all reagent volumes were halved so as to allow for triplicate analysis. Panel F features the mass spectra that was common to all samples and replicates for incubations with PCB 88, providing additional evidence that the transformation occurring in the samples is not only different from the peak seen for PCB 88, featuring a retention time nearly two minutes after PCB 88, but also that there is a common metabolite generated by the tetragnathid samples at both sites. 83
Figure 3.5 Chromatograms for enzymes obtained with liquid nitrogen and incubated with PCB 88 at Robinson Bridge (A & B) and Reese Mill (C, D, & E); and (F) a representative mass spectra. All samples were run in triplicate; however, for aesthetic clarity only five are shown here. 84
The chromatographic signal for PCB 149 was distinct from that of PCB 88. In Figure 3.6, Panel A1 shows a chromatogram with a peak eluting at 26min for PCB 88, and Panel B1 shows a peak at a different time (29min) for PCB 149. The samples from the same site were treated in the same manner. These peaks also possess two different spectra, which is not surprising due to the differences in structure of the parent compounds. 85
Figure 3.6 Chromatograms(1) and mass spectra(2) for enzyme incubations at Robinson Bridge location for PCB 88(A) and PCB 149(B). 86
Quantification of PCB 61 metabolism Isooctane blanks displayed no peaks, and both acetone control and negative controls were identical to those runs (data not shown). Positive controls with rat microsomes enriched with CYP2B showed multiple peaks with metabolites at high concentrations(Figure 3.7). Figure 3.7 Chromatogram for positive control. Note that no solvent peak is present because filament was off for first 10min. Chromatograms indicated significant metabolism of PCB 61 (Figure 3.8) with multiple peaks exhibited in addition to those expected at 29min for the recovery standard and at 31min for 2-­‐MeO-­‐PCB 61. The standard used was for the ortho-­‐
87
hydroxylated metabolite, which was the only standard available at the time. It is likely that the other peaks represent the 3-­‐MeO-­‐ and 4-­‐MeO-­‐2’,3’,4’,5’-­‐tetraCB metabolites or cofactors. Figure 3.8 Chromatogram for Reese Mill S9 fraction incubation with PCB 61, SIM m/z = 310. Peak at 31min is for MeO-­‐PCB 61 (indicated by thick arrow). The recovery standard elutes at 29min (indicated by thin arrow). The quantitative data from incubations with PCB 61 indicate that spider enzymes do have the capacity to generate OH-­‐PCB metabolites, and potentially at significant concentrations. The sample from the Robinson Bridge site did not differ from the negative control, whereas the enzyme fraction obtained without liquid 88
nitrogen from Reece Mill did show low activity (Figure 3.9). An OH-­‐PCB 61 concentration of 0.03ppm (±0.01 SD) was detected for enzymes obtained without liquid nitrogen from Reese Mill (labeled as “no N2”), whereas enzymes obtained with liquid nitrogen (labeled as “Reese Mill”) yielded a concentration of 1.63ppm (±0.35). Robinson Bridge samples did not generate a detectable quantity of 2-­‐MeO-­‐
PCB61; however, this could be a result of halving the reagent volumes in order to obtain results in triplicate. Concentration (ppm) 2.5 2 1.5 1 0.5 0 Negative Control RM -­‐ no N2 RM -­‐ N2 Robinson Bridge Figure 3.9 Measured 2-­‐MeO-­‐PCB 61 after incubation with spider S9 fractions. Metabolic Pathway Biomarker analyses for CYP1A and CYP2B shed little light on metabolic processes, as they were unable to detect any enzymatic activity in the spider samples. It appeared that even for pooled samples, the enzyme concentrations 89
might not have been high enough to yield a measurable signal, as the only measureable signals came from the positive control (rat microsomes enriched in CYP2B) for the PROD assay. Discussion In vivo anaylsis
In vivo results indicate that metabolites for PCBs with six or fewer chlorines are present in spiders at the Twelvemile Creek arm of Lake Hartwell. Expectations were confirmed when metabolites for congeners with seven or more chlorines were not detected, as such congeners are highly persistent in the environment due to their recalcitrance to metabolism (Lehmler et al. 2000). If spiders are able to metabolize their body burden, it is not surprising that the higher chlorinated and thus more bioaccumulative congeners fail to be detoxified. These analyses provided data that confirmed the presence of chlorine containing compounds, and while they could not all be identified, it did indicate that in vitro studies could yield results that could confirm the enzymatic capabilities of spiders. Qualitative evidence of metabolism is an important contribution at this time given the lack of commercially available standards for the selected PCB congeners (88 and 149). While it was not possible to identify specific compounds with this analysis, these qualitative results provide compelling evidence that chlorine-­‐
containing metabolites are present in spiders – the first evidence of this kind. 90
Selected ion monitoring indicate what PCBs can be metabolized and which structures are present. In vitro analyses
The incubation results only had one sample for each site, however the
composite of 15 individuals and the same results from each site indicate a certain
robustness to the findings. Obtaining good recoveries because of the numerous
extractions and derivatization lends validity to the process. Use of diazomethane to
derivitize samples was 80% effective, but recoveries for individual extraction steps
are unknown. Initial samples collected were obtained without liquid nitrogen, and
did not indicate any metabolism. Use of liquid nitrogen allows the enzymes to remain in their active conformations, whereas samples obtained without it lose enzymatic activity as the enzymes may denature or change upon death. The sample collection method was revised, and samples collected with liquid nitrogen were used to obtained the results presented here. Time constraints prevented the collection of additional samples; however, all samples had three separate incubations to assess reproducibility of results, which was good (see Figure 3.5).
Results for PCB 88 incubations indicate that there is enzymatic capacity for spiders to metabolize PCB 88. In vitro studies investigating PCB metabolism have been completed, but mainly in mammalian systems (McKinney et al., 2006; Verrault et al., 2009). The metabolism measured here represents the capacity of spider enzymes for metabolism, but may not be representative of what occurs in the living 91
organisms. Enzymes that were able to transform the PCBs may not be the primary enzymes used in detoxification, or may generate different metabolites than would be generated in the living organism. The metabolites most likely generated for PCB 88 are pictured in Figure 3.10. These compounds are formed when adjacent hydrogenated positions are present, and a hydroxyl group can replace a hydrogen. These are the most favorable conditions for hydroxylation (James, 2001). While other metabolites are possible, steric hindrance may bar the generation of 3’-­‐OH-­‐ and 6’-­‐OH-­‐2,2’,3,4,6-­‐CB metabolites. Figure 3.10 PCB 88 and potential metabolite structures. Results were expected to differ between PCBs 88 and 149 due to their different chlorination substitution. PCB 88 is penta-­‐chlorinated; whereas, PCB 149 92
is hexa-­‐chlorinated. Both PCBs 88 and 149 are non-­‐planar congeners. Incubations with PCB 149 yielded a unique chromatographic signal and spectrum from those obtained for PCB 88. The potential hydroxylated metabolites for PCB 149 are shown in Figure 3.11. Compared to PCB 88, PCB 149 has an additional chlorine and just one location where there are two adjacent hydrogenated positions on a ring. The high level of chlorination for this PCB (six chlorines) makes it a more recalcitrant compound, so it is possible that only one metabolite for PCB 149 can be generated. The 4-­‐OH-­‐ and 5-­‐OH-­‐2,2’,3,4’,5’,6’-­‐CB metabolites are the more likely compounds to be generated because of the adjacent hydrogens (James, 2001). The 3’-­‐hydroxy structure is not likely to be generated due to steric hindrance. Similar to the previous incubation, there was just one peak generated for the transformed PCB 149. This could also be due to just one metabolite being generated, or co-­‐elution of the 4-­‐hydroxy and 5-­‐hydroxy compounds. 93
Figure 3.11 PCB 149 and potential metabolite structures PCB 61 is a planar congener, which are known to have higher metabolism rates than reported for non-­‐planar congeners such as PCBs 88 and 149 (James, 2001). The chlorination pattern for PCB 61 allows for easier hydroxylation because one ring is completely devoid of chlorines; which allows for many potential pathways for hydroxylation since there are several adjacent ring positions with hydrogens, not chlorines (Figure 3.12). Hydroxylated PCB 61 is also noted to be an endocrine disrupting compound, causing estrogenic effects in fish and wildlife (Carlson and Williams, 2001). Most importantly, standards were available for hydroxylated and methoxylated compounds, so quantification could be completed. 94
Figure 3.12 Structure of PCB 61 and potential metabolites. Quantification of metabolism indicated higher concentrations of MeO-­‐PCB 61 generated with enzymes from Reese Mill than for those obtained at Robinson Bridge, which were not different from the negative control. Detoxifying enzymes may occur at a higher concentration in spiders at Reese Mill, resulting in higher activity and greater concentrations of metabolites, as compared to spiders at Robinson Bridge due to their history of exposure. Spiders at Reese Mill are located very close to the point source of contamination, and have possibly up-­‐regulated their detoxifying enzymes to handle their PCB body burden. These evolutionary stresses were not as severe down river at Robinson Bridge, so this same activity is not seen in that population. The concentration data for spiders at both locations are 95
reported in Chapter 4 (see Table 4.2 and Figure 4.2), which confirms that Reese Mill spiders have a higher body burden than spiders at Robinson Bridge. Many peaks other than those identified as 2-­‐MeO-­‐PCB61 were present in the chromatogram for the Reese Mill incubation with PCB 61. For some peaks, the mass spectra did not indicate presence of chlorine containing compounds, so it is likely that the other compounds were cofactors generated by the S9 fraction in response to exposure to PCB 61, not metabolites. The S9 fraction contains not just the CYP isoforms likely responsible for the biotransformation of PCBs, but the entire enzymatic library available to spiders. Enzymes that have the materials and energy available to them in the S9 fraction can complete their functions, and subsequently result in additional peaks in the chromatogram. Co-­‐elution proved to be a challenge to successful identification of metabolite structure. A single peak yields two possible interpretations: 1) one metabolite was generated, or 2) multiple metabolites are present but they co-­‐elute. Use of ortho-­‐ and para-­‐substituted metabolite standards confirmed that the method utilized is capable of separating methoxylated isomers. Commercially obtained standards for isomers with identical chlorination and different methoxylation patterns were shown to elute at different times. This check is representative of just two metabolites for a single PCB congener, as standards for all isomers for PCBs 88 and 149 were not available. The lack of available metabolite standards for these congeners precluded both quantification and resolution of this question. 96
Spiders are likely to be the organism responsible for metabolizing the parent compounds. The focus of the investigations to date have been fish and mammals (Buckman et al., 2006; McKinney et al. 2006; Verrault et al. 2009); however, these results indicate that spiders may have evolved the capacity to metabolize their contamination body burden. It is unlikely that spiders are receiving these compounds from prey. When spiders liquefy and consume insects the OH-­‐PCBs are exposed to the oxygenated environment. Extended exposure to oxygen in the laboratory can result in OH-­‐PCBs transforming to catechols (Lehmler et al., 2009). Lag time between expelling digestive enzymes onto the prey and consumption of liquefied tissue may be long enough for catechols and other compounds to form. Experiments to measure the catechol load in spiders and the OH-­‐PCB content in prey species would be good complementary studies. Comparison of in vivo and in vitro results The retention times and mass spectra obtained for PCBs 88 and 149 were cross referenced with the data obtained for the in vivo study. There were no incidences where peaks at the retention times of the in vitro study featured the same mass spectra, which could be due to co-­‐elution in the in vivo samples. Alternatively, these two highly chlorinated PCBs may not be metabolized in the field. If there are other PCBs that are more amenable to metabolism and less recalcitrant they could be preferentially metabolized (James, 2001). Continuing input of such PCBs to the metabolic system of the spider would result in lower concentrations for more easily 97
metabolized compounds, and greater concentrations of the more recalcitrant congeners. Likewise, the results obtained for PCB 61 were not present in the field samples. This was expected, as PCB 61 is not present in the Aroclor mixtures known to have contaminated the Twelvemile Creek / Lake Hartwell system. The procedure used in Chapter 4 to analyze spider PCB concentrations did not detect PCB 61. Pathway description
Biomarker assays were attempted to indicate the metabolic pathway for PCBs in spiders. EROD and PROD assays would have been able to describe the enzymatic activity of CYP1A and 2B, respectively, as well as indicate the PCB detoxification pathway (Yang et al., 2008). There was no detection of enzymatic activity for spider samples obtained with liquid nitrogen. It is likely that there was not enough protein in the S9 fractions to generate a detectable signal. To try to elucidate the pathway, rat microsomes with enriched CYP2B concentrations were used as positive controls for the incubation experiments. They were exposed to PCBs 61, 88, and 149 at the same time and in the same way as the spider samples. The results from this positive control were identical to the results from Reese Mill and Robinson Bridge samples obtained with liquid nitrogen, which supports the hypothesis that spiders have the ability to metabolize. These results support the idea that the same metabolites yielded by CYP2B in rats are generated by spider S9 fractions. The enzymes in spiders have yet to be determined, and may 98
be deserving of future research; however, the results of these experiments would direct investigations towards the cytochrome P450 suite of detoxifying enzymes, in particular CYP2B. Conclusions The first hypothesis that spiders can metabolize PCBs was supported by both the in vivo and in vitro studies. The screening of field samples for presence of PCB metabolites indicated that PCBs with six or fewer chlorines can be metabolized. PCB metabolites with seven or more chlorines were not present in the samples analyzed. Their recalcitrance to metabolism is a key driver behind their potential for bioaccumulation, in conjunction with their increased chlorine character. Metabolites generated in vitro for PCBs 61, 88, and 149 were not detected in the field extracts, which could be due to their absence in the environment (PCB 61) or to a lack of metabolism by spiders in the field (PCBs 88 and 149). The laboratory studies provide evidence that spiders have the potential to metabolize PCB congeners. PCBs 88 and 149 were demonstrated to be metabolized by spider enzymes. Screening with unique SIM parameters for PCB 88 allowed for detection of a single peak with a retention time and mass spectra different from that of the parent material. Results for incubations with PCB 149 also had retention times and mass spectra different from those for the parent material. These results indicate that spider enzymes do have the potential to metabolize PCBs with as many as six chlorines. 99
Transformation of PCB 61 to 2-­‐OH-­‐PCB 61 was quantified, and also supports the hypothesis that spiders have the potential to metabolize their PCB body burdens. Chromatograms for PCB 61 incubations differed from those with PCBs 88 and 149 in that there were many detected peaks, which supported the second hypothesis that metabolism of PCB 61 would produce multiple OH-­‐PCBs. This is likely due to the multiple pathways that are available for OH substitution onto the second ring of PCB 61; whereas a single likely pathway exists for PCBs 88 and 149. The peaks that did not indicate presence of chlorine containing compounds may represent cofactors and other compounds associated with metabolism that were generated by the S9 fraction. References Bergman, Å, Klasson-­‐Wehler, E., Kuroki, H., 1994. Selective retention of hydroxylated PCB metabolites in blood. Environ. Health. Perspect. 102, 464-­‐
469. Brouwer, A., Morse, D., Lans, M., Schuur, A., Murk, A., Klasson-­‐Wehler, E., Bergman, A., Visser, T., 1998. Interactions of persistent environmental organohalogens with the thyroid hormone system: Mechanisms and possible consequences for animal and human health. Toxicol. Ind. Health 14, 59-­‐84. Buckman, A.H., Brown, S.B., Small, J., Muir, D.C.G., Parrott, J., Solomon, K.R., Fisk, A.T., 2007. Role of temperature and enzyme induction in the biotransformation of polychlorinated biphenyls and bioformation of hydroxylated polychlorinated biphenyls by Rainbow Trout (Oncorhynchus mykiss). Environ. Sci. Technol. 41, 3856-­‐3863. 100
Buckman, A.H., Wong, C.S., Chow, E.A., Brown, S.B., Solomon, K.R., Fisk, A.T., 2006. Biotransformation of polychlorinated biphenyls (PCBs) and bioformation of hydroxylated PCBs in fish. Aquat. Tox. 78, 176-­‐185. Calvo, R., Obregon, M.J., Ruiz de Ona, C., Escobar del Rey, F., Morreale de Escobar, G., 1990. Congenital hypothyroidism, as studied in rats: the crucial role of maternal thyroxine but not of 3,5,3'-­‐triiodothyronine in the protection of the fetal brain. J. Clin. Invest. 86, 889-­‐899. Campfens, J., Mackay, D., 1997. Fugacity-­‐based model of PCB bioaccumulation in complex aquatic food webs. Environ. Sci. Technol. 31, 577-­‐583 d. Carlson, D.B., Williams, D.E., 2001. 4-­‐hydroxy-­‐2',4',6'-­‐trichlorobiphenyl and 4-­‐ hydroxy-­‐2',3',4',5'-­‐tetrachlorobiphenyl are estrogenic in rainbow trout. Environ. Tox. Chem. 20, 351-­‐358. Danerud, P.O., Morse, D.C., Klasson-­‐Wehler, E., Brouwer, A., 1996. Binding of a 3,3',4,4'-­‐tetrachlorobibhenyl (CB-­‐77) metabolite to fetal transthyretin and effects of fetal thyroid hormone levels in mice. Tox. 106, 105-­‐114. Dang, V.D., Walters, D.M., Lee, C.M., 2010. Transformation of chiral polychlorinated biphenyls (PCBs) in a stream food web. Environ. Sci. Technol. 44, 2836-­‐2841 Dang, V.D., 2007. Achiral and chiral analysis of polychlorinated biphenyls (PCBs) in the aquatic and riparian food webs in Twevle Mile Creek, South Carolina. de Haan, L.H.J., van Dijk, L., Brouwer, A., 1994. Effects of hydroxylated PCB metabolites on the Gap-­‐Junctional Intercellular Communication between mouse Hepa 1c1c7 cells. Organohal. Cmpds. 20, 491-­‐495. Denomme, M.A., Bandiera, S., Lambert, I., Copp, L., Safe, S., 1983. Polychlorinated biphenyls as phenobarbitone-­‐type inducers of microsomal enzymes. Structure-­‐activity relationships for a series of 2,4-­‐dichloro-­‐substituted congeners. Biochem. Pharmacol. 32, 2955-­‐2963. García, I., Ignacio, M., Mouteira, A., Cobas, J., Carro, N., 2008. Assisted solvent extraction and ion-­‐trap tandem mass spectrometry for the determination of ploychlorinated biphenyls in mussels. Comparison with other extraction techniques. Anal. Bioanal. Chem. 390, 729-­‐737. Greim, H., Snyder, R., 2008. Toxicology and Risk Assessment: A Comprehensive Introduction. Wiley-­‐Interscience. 1, 387. 101
Guvenius, D.M., Hassanzadeh, P., Bergman, A., Norent, K., 2002. Metabolites of polychlorinated biphenyls in human liver and adipose tissue. Environ. Tox. Chem. 21, 2264-­‐2269. Helgason, L.B., Arukwe, A., Gabrielsen, G.W., Harju, M., Hegseth, M.N., Heimstad, E.S., Jørgensen, E.H., Mortensen, A.S., Wolkers, J., 2010. Biotransformation of PCBs in arctic seabirds: characterization of phase I and II pathways at transcriptional, translational and activity levels. Comp. Biochem. Physiol. C 152, 34-­‐41. Hoekstra, P.F., Letcher, R.J., O'Hara, T.M., Backus, S.M., Solomon, K.R., Muir, D.C.G., 2003. Hydroxylated and methylsulfone-­‐containing metabolites of polychlorinated biphenyls in the plasma and blubber of bowhead whales (Balaena mysticetus). Environ. Tox. Chem. 22, 2650-­‐2658. Ishida, C., Koga, N., Hanioka, N., Saeki, H.K., Yoshimura, H., 1991. Metabolism in vitro of 3,4,3',4'-­‐ and 2,5,2',5'-­‐tetrachlorobiphenyl by rat liver microsomes and highly purified cytochrome P450. J. Pharmacobiodyn. 14, 276-­‐284. James, M.O., 2001. Polychlorinated biphenyls: Metabolism and metabolites, in: PCBs: Advances in Environmental Toxicology and Health Effects. Robertson, L., Hansen, L. (Eds.). The University Press of Kentucky, pp. 35-­‐46. Jaspers, V.L.B., Dirtu, A.C., Eens, M., Neels, H., Covaci, A., 2008. Predatory bird species show different patterns of ydroxylated polychlorinated biphenyls (OH-­‐PCBs) and polychlorinated biphenyls (PCBs). Environ. Sci. Technol. 42, 3465-­‐3471. Kania-­‐Korwel, I., Zhao, H., Norstrom, K., Li, X., Hornbuckle, K.C., Lehmler, H., 2008. Simultaneous extraction and clean-­‐up of polychlorinated biphenyls and their metabolites from small tissue samples using pressurized liquid extraction. J. Chroma. A 1214, 37-­‐46. Klasson-­‐Wehler, E., Bergman, Å, Athanasiadou, M., Ludwig, J.P., Auman, H.J., Kannan, K., Van Den Berg, M., Murk, A.J., Feyk, L.A., Giesy, J.P., 1998. Hydroxylated and methylsulfonyl polychlorinated biphenyl metabolites in albatrosses from Midway Atoll, North Pacific Ocean. Environ. Tox. Chem. 17, 1620-­‐1625 Lans, M.C., Brouwer, A., Koppe, J.G., van den Berg, M., 1990. Enzyme-­‐Induction and Alterations in Thyroid-­‐Hormone, Vitamin-­‐a and Vitamin-­‐K Levels by Tcdd in Neonatal and Maternal Rats. Chemosph. 20, 1129-­‐1134. 102
Lans, M.C., 1995. Thyroid hormone binding proteins as novel targets for hydroxylated polyhalogenated aromatic hydrocarbons (PHAHs): Possible implications for toxicity. Agricultural University Wageningen. Lehmler, H.J., Harrad, S.J., Huhnerfuss, H., Kania-­‐Korwel, I., Lee, C.M., Lu, Z., Wong, C.S., 2009. Chiral polychlorinated biphenyl transport, metabolism, and distribution: a review. Environ. Sci. Technol. 44, 2757-­‐2766. Letcher, R.J., Klasson-­‐Wehler, E., Bergman, Å, 2000. Methyl sulfone and hydroxylated metabolites of polychlorinated biphenyls, in: The Handbook of Environmental Chemistry Vol. 3 Part K New Types of Persistent Halogenated Compounds. Paasivirta, J. (Ed.). Springer-­‐Verlag, pp. 315 -­‐ 359. Letcher, R.J., Bustnes, J.O., Dietz, R., Jenssen, B.M., Jørgensen, E.H., Sonne, C., Verreault, J., Vijayan, M.M., Gabrielsen, G.W., 2010. Exposure and effects assessment of persistent organohalogen contaminants in arctic wildlife and fish. Sci. Tot. Environ. 408, 2995-­‐3043. Mackay, D., Fraser, A., 2000. Bioaccumulation of persistent organic chemicals: mechanisms and models. Environ. Poll. 110, 375-­‐391 Mackay, D., Shiu, W.Y., Ma, K.C., 1992a. Illustrated handbook of Physical-­‐chemical properties and environmental fate for organic chemicals, vol. I: monoaromatice hydrocarbons, chlorobenzenes, and PCBs. CRC Press, Boca Raton, FL. I. McKinney, M.A., Guise, S.D., Martineau, D., Béland, P., Lebeuf, M., Letcher, R.J., 2006. Organohalogen contaminants and metabolites in beluga whale (Delphinapterus leucas) liver from two Canadian populations. Environ. Tox. Chem. 25, 1246-­‐1257. Morse, D.C., Klasson-­‐Wehler, E., Wesseling, W., Koeman, J.H., Brouwer, A., 1996. Alterations in rat brain thyroid hormone status following pre-­‐ and post-­‐natal exposure to polychlorinated biphenyls (Aroclor 1254). Toxicol. App. Pharmacol 136, 269-­‐279. Morse, D.C., van Bladeren, P.J., Klasson-­‐Wehler, E., Brouwer, A., 1995. β-­‐ Naphthoflavone-­‐ and self-­‐induced metabolism of 3,3',4,4'-­‐
tetrachlorobuiphenyl in hepatic microsomes of the male, pregnant female, and foetal rat. Xenobiot. 25, 245-­‐260. 103
Narasimhan, T.R., Kim, H.L., Safe, S., 1991. Effects of hydroxylated polychlorinated biphenyls on mouse liver mitochondrial oxidative phosphorylation. J. Biochem. Toxicol. 6, 229-­‐236. Nomiyama, K., Nomura, Y., Takahashi, T., Uchiyama, Y., Arizono, K., Shinohara, R., 2010. Hydroxylated polychlorinated biphenyls (OH-­‐PCBs) induce vitellogenin through estrogenic activity in primary-­‐cultured hepatocytes of the Xenopus laevis. Chemosph. 78, 800-­‐806. Pakdeesusuk, U., Lee, C.M., Coates, J.T., Freedman, D.L., 2005. Assessment of Natural Attenuation via in Situ Reductive Dechlorination of Polychlorinated Biphenyls in Sediments of the Twelve Mile Creek Arm of Lake Hartwell, SC. Environ. Sci. Technol. 39, 945-­‐952 Pakdeesusuk, U., Freedman, D.L., Lee, C.M., Coates, J.T., 2003a. Reductive dechlorination of polychlorinated biphenyls in sediment from the Twelve Mile Creek arm of Lake Hartwell, South Carolina, USA. Environ. Tox. Chem. 22, 1214-­‐1220. Pakdeesusuk, U., Jones, W.J., Lee, C.M., Garrison, A.W., O'Niell, W.L., Freedman, D.L., Coates, J.T., Wong, C.S., 2003b. Changes in enantiomeric fractions during microbial reductive dechlorination of PCB132, PCB149, and Aroclor 1254 in Lake Hartwell sediment microcosms. Environ. Sci. Technol. 37, 1100-­‐1107. Park, J., Kalantzi, O.I., Kopec, D., Petreas, M., 2009. Polychlorinated biphenyls (PCBs) and their hydroxylated metabolites (OH-­‐PCBs) in livers of harbor seals (Phoca vitulina) from San Francisco Bay, California and Gulf of Maine. Mar. Environ. Res. 67, 129-­‐135. Purkey, H.E., Palaninathan, S.K., Kent, K.C., Smith, C., Safe, S.H., Sacchettini, J.C., Kelly, J.W., 2004. Hydroxylated polychlorinated biphenyls selectively bind transthyretin in blood and inhibit amyloidogenesis: rationalizing rodent PCB toxicity. Chem. Biol. 11, 1719-­‐1728. Safe, S., 1993. Toxicology, structure–function relationship, and human and environmental health impacts of polychlorinated biphenyls: progress and problems. Environ. Health. Perspect. 100, 259-­‐268. Safe, S.H., 1994. Polychlorinated-­‐biphenyls (PCBs) -­‐ environmental-­‐impact, biochemical and toxic responses, and implications for risk assessment. Crit. Rev. Toxicol. 24, 87-­‐149. 104
Sandau, C.D., Ayotte, P., Dewailly, E., Duffe, J., Norstrom, R.J., 2000. Analysis of hydroxylated metabolites of PCBs (OH-­‐PCBs) and other chlorinated phenolic compounds in whole blood from Canadian inuit. Environ. Health. Perspect. 108, 611-­‐616. Schurig, V., Glausch, A., Fluck, M., 1995. On the enantiomerization barrier of atropisomeric 2,2’,3,3’,4,6’-­‐hexachlorobiphenyl (PCB 132). Tetrahed. Asym. 6, 2161-­‐2164. Southwell, B.R., Duan, W., Alcorn, D., Brack, C., Richardson, S.J., Kohrle, J., Schreiber, G., 1993. Thyroxine transport to the brain: role of protein synthesis by the choroid plexus. Endocrin. 133, 2116-­‐2126. Truman, P., van den Hurk, P., 2010. Xenoestrogen exposure and effects in bluegill from the Reedy River, South Carolina, USA. Archiv. Environ. Contam. Tox. 58, 165-­‐175. Verreault, J., Maisonneuve, F., Dietz, R., Sonne, C., Letcher, R.J., 2009. Comparative hepatic activity of xenobiotic-­‐metabolizing enzymes and concentrations of organohalogens and their hydroxylated analogues in captive greenland sledge dogs (Canis familiaris). Environ. Tox. Chem. 28, 162-­‐172. Verreault, J., Letcher, R.J., Ropstad, E., Dahl, E., Gabrielsen, G.W., 2006. Organohalogen contaminants and reproductive hormones in incubating glaucous gulls (Larus hyperboreus) from the Norwegian Arctic. Environ. Tox. Chem. 25, 2990-­‐2996. Walters, D.M., Fritz, K.M., Otter, R.R., 2008. The dark side of subsidies: Adult stream insects export organic contaminants to riparian predators. Ecol. Apps. 18, 1835-­‐1841. Walters, D.M., Mills, M.A., Fritz, K.M., Raikow, D.F., 2010. Spider-­‐mediated flux of PCBs from contaminated sediments to terrestrial ecosystems and potential risks to arachnivorous birds. Environ. Sci. Technol. 44, 2849-­‐2856. Wong, C.S., Garrison, A.W., Smith, P.D., Foreman, W.T., 2001. Enantiomeric composition of chiral polychlorinated biphenyl atropisomers in aquatic and riparian biota. Environ. Sci. Technol. 35, 2448-­‐2454. Yang, F.; Xu, Y.; Pan, H.; Wu, D. 2008. Induction of hepatic cytochrome P450 1A1/2B activity and disruption of thyroglobulin synthesis/secretion by mono-­‐ortho-­‐
polychlorinated biphenyl and its hydroxylated metabolites in cell lines. Environ. Toxicol. Chem. 27, 220-­‐225. 105
CHAPTER 4 WEBS AS SUBSTRATE FOR PCB PARTITIONING Introduction Spider webs are one of the most ubiquitous examples of predation strategy in our environments. The primary function of the spider web is to capture prey. To do so, the web must support the spider, withstand the force of prey striking and struggling in the web, and retain prey long enough for the spider to consume the creatures (Chacon and Eberhard, 1980). Research into the form, chemistry, materials, and ecology of the webs has been investigated over the past thirty years.; however, a research gap exists where the characteristics and function of webs are applied to toxicology and contaminant exposure pathways for spiders. Based upon information about the composition, construction, and ecological role of webs, it is hypothesized here that a key element of detoxification of PCBs by spiders may involve exporting body burdens of contaminants to webs to alleviate stress on the organism. Araneidae spiders, which are a spider family that is closely related to Tetragnathidae spiders, are also orb-­‐weaving spiders and use a sticky, spiral thread to constitute the bulk of the web. It is important to note that the amino acids that comprise the silk proteins are all chiral, and all oriented in the D-­‐amino acid form. The majority of the protein is in the β-­‐sheet formation (Eisoldt et al., 2011;Eisoldt et al., 2012). The core of the silk proteins is an amphiphilic repetitive sequence of the 106
same amino acids, for multiple thread types and across species types (Heim et al., 2009;Heim et al., 2010). Web extracts of orb-­‐weavers include fatty acids, alcohols, esters, ketones, and low molecular weight hydrocarbons (Prouvost et al., 1999). At contaminated sites, it is suspected that webs also contain contaminants. The most important behavior from a toxicological standpoint is that Tetragnathidae spiders consume their webs and rebuild them daily. Up to 90% of the web material is recycled without being digested or reworked (Peakall, 1971), which saves energy and creates a storage location for lipid-­‐soluble contaminants. The goal of the study was to test the hypotheses that spider webs contain PCBs, and that Tetragnathidae spiders use their webs to store PCBs. Data from analyses of the total PCBs, chiral PCBs, and homologs were examined for support of these hypotheses. Materials and Methods Sample collection Webs were collected at locations along Twelvemile Creek (see Figure 4.1 for sampling locations, Table 4.1 for distances between sites and point source). Web density along the shoreline varies. Our observations corroborate the claim that Tetragnathidae spiders require approximately one meter of cubic space to construct their webs (Gillespie, 1987); however, this varies with available habitat space. Narrow sections of stream with good tree cover and many branches overlaying the 107
water column were host to many more webs than sections that were wider with poor tree coverage, and hence more accessible to avian predators. Stainless steel spatulas were cleaned with isooctane prior to collection at each site, and used to spool webs. Care was taken to avoid collecting extraneous matter including leaf debris and prey remnants. Non-­‐web material was removed by tweezers during collection, and again before processing in the lab. This was done to prevent misappropriation of PCB content to webs from the materials captured in the silk material. Webs were stored at -­‐20˚F until processed. Each site had three composite web samples. Approximately 100 webs were pooled at each site to allow for sufficient mass for analysis. One representative sample of 15 spiders was collected along with the webs at each site. All spiders analyzed in this study were removed from the webs that were collected. Table 4.1 Sampling locations and their distances from the point source. Site Distance from Distance from Point Source Previous Site Reese Mill 0.3 km 0.3 km Belle Shoals 8.85 km 8.85 km Stewart Gin 16.31 km 7.46 km Robinson Bridge 22.39 km 6.08 km 108
Figure 4.1 Map of Web Sampling Locations. (R) = Reference site upstream of Sangamo-­‐Weston point source; (RM) = Reese Mill site; (BS) = Belle Shoals site; (SG)= Stewart Gin site; (RB) = Robinson Bridge site. Sample extraction and clean-­‐up Web samples were weighed and recovery standards PCBs 14 and 169 (Accustandard, Wellington, CT) were added to the cells prior to extraction on Dionex 200 ASE. PCBs 14 and 169 were used as recovery standards, at 55% (±24 SD) and 30% (±23 SD) respectively across both sample types. Concentrations were not corrected for recovery. 109
A 9:1 hexanes:acetone solution (v/v) was used for the extraction (García et al., 2008). Heating time was 5min followed by two 2min static cycle. Cells were heated to 100˚C at 1500psi, using two cycles of 5min heating and 5min static. A flush volume of 60% was used, and solvent lines were purged for 60sec (Kania-­‐
Korwel et al., 2008b). After each sample the system was rinsed with the solvent mixture to prevent carryover between cells. The extracts were condensed to 6mL under a steady stream of high purity nitrogen, and lipid content was measured by a gravimetrical method. Briefly, samples were cleaned on an alumina-­‐sodium sulfate column (3.5g and 1.5g, respectively). Columns were eluted with hexanes to a final volume of 10mL, then solvent exchanged for isooctane. Samples were condensed under nitrogen to a final volume of 2mL for spider samples, and 1mL for web samples. Internal standards of aldrin and PCB 204 were added before analysis. Quality of assessment was maintained through use of procedural blanks, laboratory control samples in duplicate, matrix spikes in duplicate (aldrin and PCB 204, Accustandard), blank tissue samples, and check standards with each batch of 10 samples. GC-­‐ECD analysis of samples An HP 6890 gas chromatograph with an RTX-­‐5 column (Restek, 60m x 0.25mm x 0.25μm) was used to quantify the PCB content of spider webs. The temperature program began at 115˚C, and was held steady for 2min. It was then ramped at a rate of 8°C/min to a temperature of 185°C and held for 8min. Lastly the 110
temperature was ramped at a rate of 2°C/min to a final temperature of 265°C, and held for 20min. An HP 6850 gas chromatograph equipped with a Chirasil-­‐Dex column (25m x 0.25mm) and a 63Ni electron detector was used for enantioselective analysis. The temperature program began at 100˚C for 2min, then ramped at 10˚C/min to 170 and held for 10min, then ramped at 8˚C/min to 180˚C. The detection limit did vary somewhat by congener, but was approximately 25ng/L for chiral PCB congeners. The EFs for the racemic standards (PCBs 91, 95, 136, 149, and 174) were between 0.492 (± 0.003 SD) and 0.504 (±0.006 SD). Chiral standards were run prior to samples in every sequence to maintain confidence in the results. Precision was maintained by repeat injections every fifth sample. PCB concentrations and EF values among different sampling locations were assessed by one-­‐way analysis of variance (ANOVA) using Tukey’s test to determine significance (p<0.05). Nonmetric-­‐multidimensional scaling statistical analysis (NMS) was completed using PC-­‐ORD 4.0 software (MjM software). The autopilot mode was utilized, and the chi squared coefficient used as the distance measure (McCune and Mefford, 1999; McCune et al., 2002). Results Spider concentrations were greater than those observed in their respective webs, though the extent to which they were greater varied (Figure 4.2). Concentration 111
ratios between webs and spiders fall sharply to become much more even with distance from the point source (Table 4.2). Σ PCB (ng/g lipid) 3500 3000 2500 2000 spider 1500 web 1000 500 0 Reese Mills Belle Shoals Stewart Robinson Gin Bridge Figure 4.2 ΣPCB concentration (ng/g lipid) for spiders (n=1) and webs (n=3). Error bars indicate one standard deviation in either direction. Table 4.2 ΣPCB content in spiders and webs at each sampling location. a No PCBs were detected at the Belle Shoal site, which is likely a function of low sample weight. Sampling Location Reese Mill Belle Shoals Stewart Gin Robinson Bridge ΣPCBspiders (ng/g lipids) 2902.2 899.7 284.0 1608.3 ΣPCBwebs (ng/g lipids) 154.4 ± 40.2 n.d.a 203.8 ± 42.3 356.1 ± 107.9 [spiders] / [webs] 18.8 -­‐-­‐ 1.4 4.5 This trend is described for the Reese Mill sample site in greater detail in Table 4.3, which shows the homolog distributions, and is discussed below.. While the table lists values obtained for the Reese Mill site only, it is representative of the 112
patterns observed at all sampling locations where web concentrations were sufficient for detection. Spiders had greater concentrations for all homolog groups analyzed except for di-­‐chlorinated and nona-­‐chlorinated congeners (Table 4.2). Webs had a concentration of 40.3ng/g lipids of di-­‐chlorinated congeners, whereas spiders did not have detectable quantities of such congeners, and webs also had nona-­‐chlorinated compounds present, whereas spiders did not. Table 4.3 Homolog concentrations in spiders and webs at the Reese Mill sampling site. Homolog Class [spiders] [webs] Dominant (ng/g lipids) (ng/g lipids) compartment Di 0 40.3 Webs Tri 135.2 3.9 Spiders Tetra 903.5 75.1 Spiders Penta 1357.5 26.3 Spiders Hexa 396.0 2.7 Spiders Hepta 108.4 1.6 Spiders Octa 1.6 3.2 Webs Nona 0 1.4 Webs Homolog Patterns When averaged across the sampling sites, spiders had consistent homolog patterns among sampling sites (Figure 4.3). The penta-­‐chlorinated congeners had the highest percentages, followed by tetra-­‐chlorinated and hexa-­‐chlorinated congeners. The lowest percentages were seen in the di-­‐ and octa-­‐chlorinated homolog classes, and no nona-­‐chlorinated congeners was measured in spiders at any site other than 113
at Belle Shoals. Spiders from the Reese Mill site differed from the other three locations in that tri-­‐chlorinated congeners contributed to their overall PCB concentration at a higher percentage, comparable to that of hexa-­‐chlorinated congeners at that site (Figure 4.3A). Figure 4.3 Homolog patterns of PCB concentrations in spiders (A) and webs (B). 114
The web samples had a somewhat different homolog pattern (Figure 4.3B). There was no consistent concentration pattern among sites; however, nona-­‐
chlorinated compounds were observed to be the lowest contributing homolog class, just as they were in both the spider samples. Results at the individual sites varied. Reese Mill was the site closest to the point source, and also had the greatest contribution by di-­‐chlorinated PCBs among the sites analyzed. Webs at Reese Mill had a significant contribution by di-­‐
chlorinated PCBs at 26%. The source of these congeners was likely volatilization from the water column. Webs at the Stewart Gin site did not have an abundance of di-­‐ and tri-­‐chlorinated congeners, with less than 1% of the total PCB concentration being attributed to the lower chlorinated congeners. This differed from Robinson Bridge, which while further downstream from the point source, featured almost 5% each of di-­‐ and tri-­‐chlorinated congeners. The Robinson Bridge site had a significantly higher contribution of tetra-­‐chlorinated congeners to webs than spiders (48.5% in webs, 10.6% in spiders), which indicates that there is likely input of these congeners to webs from both spiders and volatilization from the water column. Chirality Data Nearly all web samples contained adequate concentrations of PCBs 91, 95, and 149 to quantify and assess for chirality. Some of the Stewart Gin replicate samples did not have high enough concentrations of the chiral PCBs to assess their 115
chirality on the Chiralsil-­‐Dex column; as a result, the EF graphs for this site reflect values for only one sample and do not feature error bars. Additionally, the Stewart Gin site did not have a high abundance of spiders, and a pooled sample of sufficient mass was not able to be obtained for use in chiral analysis. In Figure 4.4 the EF values for webs from all four sampling locations is shown. Both PCB 91 and 95 were approximately racemic. PCB 95 concentrations fell below detection limits in webs from Belle Shoals and Steward Gin, but at Reese Mill and Robinson Bridge the EF was found to be approximately racemic. PCB 149 is significantly nonracemic at all locations, with the second eluting congener being preferentially retained in the organisms. Haglund and Wiberg (1996) determined that on the Chiralsil-­‐Dex column the (+) enantiomer elutes first, and the (-­‐) enantiomer elutes second for PCB 149. The data for PCB 149 are dissimilar from that observed for PCBs 91 and 95 in this study, noted in Chapter 2, and as well as what was observed in leaves by Dang et al. (2010). 116
Figure 4.4 Enantiomeric fraction values for chiral PCBs detected in spiders (A) and webs (B). Error bars indicate 95% confidence interval. 117
The EF data for PCB 149 also fails to support the hypothesis that webs exhibit similar EF patterns as spiders (Figure 4.4 The EFs in the spider samples were racemic (Reese Mill) or less enriched in the second enantiomer than the web samples (Belle Shoals and Robinson Bridge). Figure 4.5 NMS plot for spider and web samples. Samples were included for all sites. The points that fall outside of the circles represent webs. The NMS plot of web and spider EF values for PCBs 95 and 149 indicated a different distribution for the two sample types. For the plot shown, web and spider samples from all species and locations were incorporated (Figure 4.5). Within the groups the relationships were less clear. Although grouped together, the data for 118
webs did not indicate any spatial trends. For spiders, samples closest to the point source fall near the top of the plot, and the most distant spider sample falls at the bottom of the oval; however, this trend cannot be deemed significant as only one pooled spider sample was supplied for each site. Points that fall outside of the circles all represent web samples. Discussion Total PCBs The change in ratios likely tied to the concentration in spiders decreasing with distance from the point source. Spider concentration varied by up to an order of magnitude, whereas web concentration only changed by an approximate factor of two. It is proposed here that differences in evolution at sampling sites may affect the concentrations observed. Spiders at Reese Mill have been shown to exhibit higher metabolic activity (see Chapter 3), which is likely due to their need to detoxify a more significant body burden than observed at other sampling locations. If the spiders expend energy metabolizing their body burden they may not have as great a need to sequester contaminants. Spiders have a lesser body burden at the more distant sites (see data for Robinson Bridge spiders), which would not place as great a demand on the spiders to actively detoxify and decrease their PCB concentrations. This would allow more energy to be directed towards growth and reproduction (Walker et al., 2006), which could also contribute to a lesser body burden. 119
It is important to note that the transfer of PCBs could move in either direction – from the spiders to the webs, or the webs to the spider. One anomaly in the data was that the Belle Shoals site did not have any detectable PCBs in the webs, which is likely a function of low sample weight. It was not possible to obtain samples from Belle Shoals in excess of 1g due to low web and spider density at that site. The river was wider at this location, and there was less tree cover, allowing for greater access by birds to spiders. The PCB concentration in webs from all sampling locations were similar, varying by only a factor of about two, despite variation in spider concentration among sites. This could be linked to longevity of the material as it relates to the lifetime of the spider. Spiders accumulate lipid tissue that may remain in their system over the course of their life; whereas, web material may not last as long as the lipids within the spider regardless of material recycling (Peakall, 1971). If a web is destroyed, the contaminated material cannot be consumed by the spider and a new web must be constructed from new, less contaminated material. The ratio reflects not so much a chemical limit for the mass of PCBs that can associate with the material, but rather points towards physical reduction of PCBs exposure when webs are destroyed. Homologs The observed homolog patterns may be explained by contributions from both volatilization from the stream and spiders. Spiders receive their entire body 120
burden of PCBs via trophic transfer. Webs function as a tissue external to the body, and are suspended directly above the water column. Analysis of homolog fugacity indicates that lower chlorinated congeners are much more volatile than highly chlorinated congeners (Schwarzenbach et al., 2003). The distribution for tetra-­‐, penta-­‐, and hexa-­‐chlorinated congeners in webs supports the hypothesis that spiders offload their PCB body burden to their webs; however, the higher percentage of lower chlorinated congeners in webs as compared to spiders fails to support the spider off-­‐loading hypothesis. It is possible that PCBs volatilize from the water column due to the contaminated sediment of Twelvemile Creek and sorb onto the webs. If true, then it follows that the PCBs present would more closely mirror the profile seen in the water column. Dang (2012), using polyethylene passive samplers, determined that the distribution of PCBs in Twelvemile Creek was skewed towards the lower chlorinated congeners, so an abundance of lower chlorinated congeners could be due to volatilization. Additionally, Dang (2012, unpublished data) determined that about 10% of the PCBs sorbed onto Rhododendron leaves at a site just down river at Shady Grove in Pickens, SC, were di-­‐chlorinated. It was concluded that because the water column also had detectable concentrations of di-­‐
chlorinated congeners that volatilization was the source of the lower-­‐chlorinated congeners. Volatilization as a source of PCBs to webs at Stewart Gin is not supported; rather, the abundance of hepta-­‐ and octa-­‐chlorinated compounds indicated that PCBs from spiders are the predominant source to webs. The homolog distributions 121
of spiders and webs most closely mirror each other at Stewart Gin, again providing evidence that spiders off-­‐load their body burden to webs. The homolog distribution at the Robinson Bridge site (Figure 4.3), like the data from Reese Mill, provides evidence that volatilization is the likely supplier of di-­‐ and tri-­‐chlorinated congeners. While the contribution of the lower chlorinated congeners is less than what was observed at Reese Mill, their presence is evidence that the water column is supplying a portion of the PCBs observed in webs. A spatial pattern in homolog enrichment appears in the data. Sites closest to the point source (Reese Mill) have broader PCB distributions in spiders that capture lower chlorinated congeners, whereas sites more distant (Robinson Bridge) are enriched in the tetra-­‐, penta-­‐, and hexa-­‐chlorinated congeners. There may be a change in the source to the spiders. Dang (2012) also determined that there are ongoing sources of PCBs to Twelvemile Creek. These sources are closer to the Reese Mill site. The lower chlorinated congeners are the most easily removed from the ecosystem via volatilization and bacterial degradation (Farley et al., 1996). Such congeners may not be present at high concentrations at the more downstream sampling locations; hence, the smaller contribution of lower chlorinated compounds in webs at those downstream sites as compared to the upstream web distributions. Turbulence of the water in the stream could also influence volatility of the PCBs in the water column. High turbulence can reduce the energy necessary to escape the water phase, thus encouraging higher weight PCBs to volatilize. The 122
process is likely more complicated than web PCB content being derived from a single source. While the greater concentrations of di-­‐chlorinated congeners in webs can be explained through fugacity, the presence of nona-­‐chlorinated congeners is more difficult to explain. Nona-­‐chlorinated congeners are more bioaccumulative than lower chlorinated congeners due to their higher chlorine content, and Kow value (Schwarzenbach et al., 2003). These congeners should preferentially be retained in the organism. Their presence in the web may be attributed to spiders offloading their body burden into lipid tissues outside of the body. Spider webs are recycled by the organisms, with nearly 90% of web material being reused in new webs spun nightly without being metabolized by the organism (Peakall, 1971). Spiders may be able to store these contaminants outside of their bodies. The high Kow value, homolog recalcitrance to metabolism, and recycling of web lipids may all contribute to the accumulation of highly chlorinated congeners in the web material. Chirality The chirality data fail to support the hypothesis that webs are mirrors of the contamination seen in spiders. Rather, it provides evidence that suggests that PCB sorption to webs is an enantioselective process. The data from the Twelvemile Creek arm of Lake Hartwell discussed previously (see Figure 2.2) indicated that the EF value for spiders at the lake was racemic. The data obtained for the river spiders indicate that over the course of the gradient down the river the EF values for PCBs 123
95 and 149 both reflect an increasing prevalence of the first eluting enantiomer. This behavior is not likely attributed to a change in the spider uptake process, but rather a change in the source of PCBs. The results for PCB 95 are similar to those obtained for lakeshore spiders (Chapter 2), as are the EF results for PCB 91 and 149 at Reese Mill, but it is possible that with an increased sample size the trend would be found nonsignificant. Alternatively, the process of off-­‐loading the PCB body burden to the webs could be a non-­‐selective process, whereas PCBs sourced to the webs from the water column via volatilization may have a non-­‐racemic signature. Fugacity should be identical for both enantiomers, but if the starting material in the water column does not have a racemic signature, then the webs would have an enriched EF value as well. Previous investigations at this site have indicated that PCB 95 is racemic in the water column, and PCB 91 is enriched in the first eluting enantiomer (Dang, 2012). PCB 95 was found to be racemic, which is similar to the water column but differs from that measured in spiders. PCB 91 in webs was also racemic, but differed from the values seen for both spiders and the water column. Comparison between the spider and the web EF data would indicate that webs are not good indicators for the contamination profile in spiders; however a larger sample size of spiders would allow for a better assessment of the contamination profile for that population. The NMS plot of web and spider samples failed to indicate any spatial patterns (Figure 4.5), but suggested that webs have a distinct EF profile compared with spiders at the same location. Web EF values were 124
very similar among sites, and resulted in a clear grouping of those samples in the plot. Conclusions The hypothesis that webs have PCB content was confirmed by these investigations. Homolog patterns support the hypothesis that spiders off-­‐load their body burden to their webs. The distribution pattern for webs roughly mirrors that of spiders, though with a higher contribution of di-­‐ and tri-­‐chlorinated congeners, as well as nona-­‐chlorinated congeners. It is likely that while the majority of the highly chlorinated congeners are sourced from the spiders, most of the mass of lower chlorinated congeners is due to volatilization from the water column. Chirality data provide additional support for the spider off-­‐loading hypothesis. The non-­‐racemic values for PCB 149 indicate that a biological process is likely affecting the transport from spiders to webs. NMS plots where web and spider samples are clearly grouped and do not overlap also suggest enantioselection. 125
References Blackledge, T.A., Scharff, N., Coddington, J.A., Szuts, T., Wenzel, J.W., Hayashi, C.Y., Agnarsson, I., 2009. Reconstructing web evolution and spider diversification in the molecular era. Proc. Natl Acad. Sci. USA 106, 5229-­‐5234. Chacon, P., Eberhard, W.G., 1980. Factors affecting numbers and kinds of prey caught in artifical spider webs, with considerations of how orb webs trap prey. Bull. Br. Arachnol. Soc. 5, 28-­‐38. Dang, V.D. 2012. Assessing ongoing sources and fate of dissolved polychlorinated biphenyls (PCBs) in a stream. Ph.D. dissertation. Dang, V.D., Walters, D.M., Lee, C.M., 2010. Transformation of Chiral Polychlorinated Biphenyls (PCBs) in a Stream Food Webs. Environ. Sci. Technol. 44, 2836-­‐
2841. Eisoldt, L., Thamm, C., Scheibel, T., 2012. The role of terminal domains during storage and assembly of spider silk proteins. Biopolymers 97, 355-­‐361. Eisoldt, L., Smith, A., Scheibel, T., 2011. Decoding the secrets of spider silk. Materials Today 14, 80-­‐86. Farley, .J.; Germann, G.G.; and Elzerman, A.W. 1994. Differential weathering of PCB congeners in Lake Hartwell, South Carolina. In: Environmental Chemistry of Lakes and Reservoirs. Baker, L.A., Ed. Advances in Chemistry; American Chemical Society. Foelix, R.F., 1982. Biology of Spiders, 1st ed. Harvard University Press, Cambridge, Massachusetts. García, I., Ignacio, M., Mouteira, A., Cobas, J., Carro, N., 2008. Assisted solvent extraction and ion-­‐trap tandem mass spectrometry for the determination of polychlorinated biphenyls in mussels. Comparison with other extraction techniques. Anal. Bioanal. Chem. 390, 729-­‐737. Gillespie, R.G., 1987. The Mechanism of Habitat Selection in the Long-­‐Jawed Orb-­‐ Weaving Spider Tetragnatha elongata (Araneae, Tetragnathidae). J. Arachnol. 15, 81-­‐90. 126
Hawthorn, A.C., Opell, B.D., 2003. van der Waals and hygroscopic forces of adhesion generated by spider capture threads. J. Exp. Biol. 206, 3905-­‐3911. Hawthorn, A.C., Opell, B.D., 2002. Evolution of adhesive mechanisms in cribellar spider prey capture thread: evidence for van der Waals and hygroscopic forces. Biol. J. Linn. Soc. 77, 1-­‐8. Heim, M., Romer, L., Scheibel, T., 2010. Hierarchical structures made of proteins. The complex architecture of spider webs and their constituent silk proteins. Chem. Soc. Rev. 39, 156-­‐164. Heim, M., Keerl, D., Scheibel, T., 2009. Spider Silk: From Soluble Protein to Extraordinary Fiber. Angewandte Chemie International Edition 48, 3584-­‐
3596. Kania-­‐Korwel, I., Zhao, H., Norstrom, K., Li, X., Hornbuckle, K.C., Lehmler, H., 2008. Simultaneous extraction and clean-­‐up of polychlorinated biphenyls and their metabolites from small tissue samples using pressurized liquid extraction. Journal of Chromatography A 1214, 37-­‐46. Kennedy, C.H., 1950. The Relation of American Dragonfly-­‐eating Birds to Their Prey. Ecol. Monogr. 20, 103-­‐142. McCune, B., Grace, J.B., Urban, D.L., 2002. Analysis of Ecological Communities. MjM Software Design, Gleneden Beach, OR. McCune, B., Mefford, M.J., 1999. PC-­‐ORD. Multivariate Analysis of Ecological Data, Version 4. MjM Software Design, Gleneden Beach, OR. Opell, B.D., Markley, B.J., Hannum, C.D., Hendricks, M.L., 2008. The contribution of axial fiber extensibility to the adhesion of viscous capture threads spun by orb-­‐weaving spiders. J. Exp. Biol. 211, 2243-­‐2251. Opell, B.D., Bond, J.E., 2000. Capture thread extensibility of orb-­‐weaving spiders: testing punctuated and associative explanations of character evolution. Biol. J. Linn. Soc. 70, 120-­‐120. Opell, B.D., 1999. Changes in spinning anatomy and thread stickiness associated with the origin of orb-­‐weaving spiders. Biol. J. Linn. Soc. 68, 593-­‐612. Opell, B.D., 1998. Economics of spider orb-­‐webs: the benefits of producing adhesive capture thread and of recycling silk. Funct. Ecol. 12, 613-­‐624. 127
Peakall, D.B., 1971. Conservation of web proteins in the spider Araneus diadematus. J. Exp. Zool. 176, 257. Prouvost, O., Trabalon, M., Papke, M., Schulz, S., 1999. Contact sex signals on web and cuticle of Tegenaria atrica (Araneae, Agelenidae). Arch. Insect Biochem. Physiol. 40, 194-­‐202 Schulz, S., Toft, S., 1993. Branched long chain alkyl methyl ethers: a new class of lipids from spider silk. Tetrahedron 49, 6805-­‐6820. Selden, P.A., Shear, W.A., Sutton, M.D., 2008. Fossil evidence for the origin of spider spinnerets, and a proposed arachnid order. Proc. Natl Acad. Sci. USA 105, 20781-­‐20785. Walker, C.H., Hopkin, S.P., Sibly, R.M., and Peakall, D.B. 2006. Principles of Ecotoxicology. Taylor and Francis Group. Vollrath, F., Selden, P.A., 2007. The role of behavior in the evolution of spiders, silks, and webs. Ann. Rev. Ecol. Svol. Syst. 38, 819-­‐846. 128
CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS Conclusions Chirality investigations failed to support the hypotheses that each spider species would be distinct from the others. The second hypothesis, that spider EF values would be distinct from prey EF values, was supported by the chirality data. Spider values shifted in the direction favoring retention of the first eluting enantiomer for PCB 91 compared to the midges; EF values for PCBs 95 and 149 did not have a clear shift in direction. This change in EF for PCB 91 could be a product of metabolism on the part of the spider, or spider off-­‐loading of their body burden to web material. Evidence from in vivo and in vitro studies support the hypothesis that spiders can metabolize their PCB body burden. Analysis of field extracts provided evidence that PCB metabolites are present in spiders located along the Twelvemile Creek arm of Lake Hartwell. There was no detection of metabolites with seven or more chlorines. Although these results supported the hypothesis that spider can transform their PCB body burden, the actual products of metabolism were less clear due to the number of unidentified peaks in the metabolic fraction. Laboratory incubations of enzymes with PCBs 88 and 149 provided evidence that these congeners can be transformed by spider enzymes to a single compound distinct from the parent material. PCB 61 was partially metabolized to 2-­‐MeO-­‐PCB 61, and potentially to other metabolites. 129
Web investigations affirmed the hypothesis that webs have measurable concentrations of PCBs. The concentrations are not significant enough to warrant their inclusion in ecological risk assessments, but may help spiders manage their PCB body burden. The hypothesis that spiders are the source of these PCBs was partially supported due to similarities in the homolog distribution patterns between spiders and webs; however, the greater presence of di-­‐ and tri-­‐chlorinated congeners in webs indicates that volatilization may also contribute to the total PCB concentration in webs. Changes in chirality from spiders to webs indicate that biological processes may be affecting the transport of chiral PCBs from spiders to webs, providing additional evidence that spiders contribute to web PCB content. Recommendations for Future Work An exposure study could be completed where prey insects are reared with exposure to a limited number of PCB congeners, and are in turn fed to spiders. This would allow for targeted analysis of the transfer between trophic levels, at what concentrations, and how the chirality is affected at each level of the food chain. Such a study could also investigate the metabolism of the PCB congeners at each level of the food chain, or even investigate if trophic transfer of PCB metabolites occurs. Such investigations would answer lingering questions about the source of OH-­‐PCBs in spider, and whether or not OH-­‐PCBs in prey are transformed to catechols during the feeding process. 130
Such a study could allow for a mass balance to be conducted. This would be the optimal way to describe fate and transport. An enclosed food web would also allow for collection of spider excrement, which has yet to be investigated for PCB content. The material is challenging to collect, as the spiders are suspended above the water when they release it to the environment, and it drops into the water column where it dissolves. If spiders are removed from the habitat after controlled feeding exposures and sequestered in a fish tank for a few days enough excrement might be collected to begin analyzing what exits the organism via excretion. A study of this nature was attempted for this dissertation; however, it was unsuccessful due to low population yield of exposed midges. Additional complications arose when spiders were unable to spin webs successfully in the provided enclosures. It is hypothesized that a better prey yield coupled with a cooler, more humid synthetic habitat for spiders could provide a better study system. Additional metabolite studies can also be conducted. The wider array of metabolite standards now available would allow for consideration of different parent congeners. Additionally, obtaining standard for 3-­‐MeO-­‐ and 4-­‐MeO-­‐PCB61 could allow for a more comprehensive analysis of the data already obtained. The attempts to determine the metabolic pathway for spiders were inconclusive. Use of antibodies to identify which enzymes are active could be an alternative way to determine the pathway. Antibodies for CYP1A and CYP2B can be developed with fluorescent tags such that when activity is detected there is a 131
measureable signal from the antibodies. This could be challenging, too, because many of these antibodies have been developed in mammalian systems, and may not be optimal for use in arthropod systems. A volatility exposure study could be conducted to better determine the source of PCBs in webs. A controlled laboratory experiment that uses web material, water, and the Aroclor mixtures present at this Superfund site could provide information about which PCBs are volatilizing from the water column, and which of those sorb on to the webs if at all. 132
APPENDICES 133
Appendix A Supplemental Figures for Chapter 2 1
PCB 91
PCB 95
0.5
PCB 149
PCB174
0
T6
G
H
I
K
N
O
T12
Figure A.1 EF values for chiral congeners at sampling locations in Tetragnathidae. 1
PCB 91
PCB 95
0.5
PCB 149
PCB174
0
T6
G
H
I
K
N
O
T12
Figure A.2 EF values for chiral congeners at sampling locations in Araneidae. 134
1
EF
PCB 91
PCB 95
0.5
PCB 149
PCB174
0
T6
G
H
I
K
N
O
T12
Figure A.3 EF values for chiral congeners at sampling locations in basilica. 1
EF
PCB 91
0.5
PCB 95
PCB 149
0
T6
G
H
I
K
N
O
T12
Figure A.4 EF values for chiral congeners at sampling locations for Chironomidae. 135
Appendix B Supplemental Tables for Chapter 2 Table B.1 Environmentally stable chiral congener substitution patterns PCB Congener Chlorine Substitution Pattern 45 2, 2’, 3, 6 84 2, 2’, 3, 3’, 6 88 2, 2’, 3, 4, 6 91 2, 2’, 3, 4’, 6 95 2, 2’, 3, 5’, 6 132 2, 2’, 3, 3’, 4, 6’ 136 2, 2’, 3, 3’, 6, 6’ 144 2, 2’, 3, 4, 5’, 6 149 2, 2’, 3, 4’, 5’, 6 171 2, 2’, 3, 3’, 4, 4’, 6 174 2, 2’, 3, 3’, 4, 5, 6’ 183 2, 2’, 3, 4, 4’, 5’, 6 Table B.2 Sample sizes (n) for each EPA sampling site. Individuals were pooled to create samples with approximately 1.5g mass. Sampling site G T6 H I K N O T12 Tetragnathidae 4 Araneidae 0 Basilica 1 3 3 4 3 3 3 3 3 1 3 4 3 0 3 4 3 3 3 1 1 1 136
Appendix C Supplemental Tables for Chapter 3 Table C.1 SIM parameters used to screen for PCB metabolites. Three settings were used for each homolog class. # Cl 1 2 3 4 5 6 7 8 9 m/z SIM settings 218.57 253.02 287.47 321.92 356.37 390.82 425.27 459.72 494.17 216 / 218 / 220 251 / 253 / 255 286 / 288 / 290 320 / 322 / 324 354 / 356 / 358 388 / 390 / 392 423 / 425 / 427 458 / 460 / 462 492 / 494 / 496 137