UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Interdisciplinary approaches to endocrine disruption : effects of dietary exposure of 4-nonylphenol on… Keen, Patricia Lynn 2002

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2002-0128.pdf [ 23.21MB ]
JSON: 831-1.0090212.json
JSON-LD: 831-1.0090212-ld.json
RDF/XML (Pretty): 831-1.0090212-rdf.xml
RDF/JSON: 831-1.0090212-rdf.json
Turtle: 831-1.0090212-turtle.txt
N-Triples: 831-1.0090212-rdf-ntriples.txt
Original Record: 831-1.0090212-source.json
Full Text

Full Text

INTERDISCIPLINARY APPROACHES TO ENDOCRINE DISRUPTION: EFFECTS OF DIETARY EXPOSURE OF 4-NONYLPHENOL ON SMOLTIFICATION OF JUVENILE COHO SALMON (Oncorhynchus kisutch) AND HOW RISK EFFECTS ARE PERCEIVED By PATRICIA LYNN KEEN B.Sc , University of British Columbia, 1986 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R OF ARTS In THE F A C U L T Y OF G R A D U A T E STUDIES (RESOURCE M A N A G E M E N T A N D E N V I R O N M E N T A L STUDIES) We accept this thesis as conforming To the required standard THE UNIVERSITY OF BRITISH COLUMBIA March 2002 © Patricia Lynn Keen, 2002 •i A B S T R A C T Recently, the scientific community has been concerned about the potential of exposing organisms to natural estrogens and xenobiotic compounds that can interfere with normal endocrine functions in these organisms. Indeed, over the past decade, it has been recognized that there is a need to conduct more research that involves studying the effects of exposure of organisms to endocrine disrupting compounds in order to understand their implications for human and ecosystem health. Accordingly, the objective of this thesis was to contribute to our knowledge of this societal concern. This thesis approaches the endocrine disruption hypothesis from two streams of knowledge. The first half of the thesis describes the results of a controlled experiment to investigate the effect of dietary exposure of juvenile coho salmon (Oncorhynchus kisutch) to 4-nonylphenol with respect to their ability to complete smoltification and grow well during early marine residency. I hypothesized that dietary exposure of juvenile coho salmon to 4-nonylphenol during the freshwater phase immediately prior to parr-smolt transformation would adversely affect their subsequent growth, survival and growth performance upon transfer to seawater. A randomized block experiment using duplicate groups of coho salmon exposed to four broad range dietary concentrations of 4-nonylphenol, one positive estrogen control and one negative control was conducted to test this hypothesis. The results of the experiment did not support the hypothesis under these particular experimental conditions. However, important information regarding the degree of ubiquitous contamination of 4-nonylphenol was gained. Chemical analyses determined the presence of 4-nonylphenol in significant concentrations in the marine oil that was incorporated in the diets and in the commercially prepared krill coating that was used to make the diets more palatable. Therefore, relative responses of the fish to the diet treatments are only possible since there was no negative control diet without 4-nonylphenol. Subtle differences in retention of parr marks after seawater transfer were observed between fish given the treatments although no chemical analyses were performed to substantiate these observations. i i The second half of the thesis was concerned with the examination of the nature of the perceived risk posed by endocrine disrupting compounds from a social scientific perspective. Using a conceptual framework built around the condition of "risk society", some key issues involving risk communication and the perception of the risk are described. To apply qualitative research methods to investigate the perception of endocrine disrupting compounds as a risk to human and ecosystem health, a questionnaire was designed to examine public awareness of the EDC issue. I hypothesized that physicians would represent "intermediary experts" whose scientific background and familiarity with the concerns of the public at large would give an indication of public perception of risk that exposure to EDCs would pose to human or ecosystem health. The response to the survey that was provided to the physicians was extremely poor within the six-week period of study and the questionnaire provided no significant new knowledge upon which informed decisions could be made for constructing a future comprehensive survey. The thesis concludes with a discussion of the integration of the two pathways to examine the endocrine disruption hypothesis. Some specific recommendations for further research are proposed in the context of integrated approaches to assessing the effects and perception of risk of exposure to endocrine disrupting compounds. i i i TABLE OF CONTENTS ABSTRACT i i TABLE OF CONTENTS iv LIST OF TABLES viii LIST OF FIGURES ix LIST OF ABBREVIATIONS xi PREFACE xiv ACKNOWLEDGEMENTS xvi DEDICATION xvii 1. INTRODUCTION 1 2. BACKGROUND 6 2.1 Endocrine Disrupting Compounds in the Environment 6 2.2 Implications of EDCs on Human Health 7 2.3 Environmental Effects of EDCs on Other Organisms 9 2.4 Assessing Estrogenicity of Contaminants . 10 2.5 Contaminants that Demonstrate Estrogenicity 12 2.6 Endocrine Disruption in Fish 13 2.7 Summary of Coho Salmon Life Cycle 16 2.8 Basic Nutrition of Coho Salmon 22 2.9 Risk Assessment of Endocrine Disrupting Compounds 25 2.10 Environmental Relevance of 4-Nonylphenol in British Columbia Watershed 29 iv 2.11 Estrogenic Effects of 4-Nonylphenol 31 3. METHODOLOGY 34 3.1 E X P E R I M E N T A L D E S I G N . 34 3.1.1 Animals 34 3.1.2 Test Conditions 34 3.1.3 Diet Composition and Administration of 4-NP 36 3.1.4 Treatment 38 3.2 A N A L Y S E S 39 3.2.1 Fish Weighing and Sampling 39 3.2.2 Hematocrit Measurement 40 3.2.3 Seawater Challenge Test 40 3.2.4 T 3 and T4Hormone Measurement 41 Assay Procedure 42 3.2.5 Vitellogenin Assay : 43 Reagent Vitellogenin 43 Antibodies 44 ELISA Procedure 44 3.2.6 Proximate Analyses 45 Moisture and Ash Determination 46 Protein Analyses '. 46 Lipid Analyses 47 3.2.6A Gross Energy Determinations 47 v 3.2.7 Determination of Concentrations of 4-Nonylphenol 47 Chemical Analyses of 4-Nonylphenol 48 Collection and Analyses of Water Samples 48 Sample Preparation 49 Glassware Preparation 49 Steam Distillation Extraction 50 Sample Clean-up 50 Normal Phase HPLC 50 Liquid Chromatography Ionization Mass Spectrometry Determinations 51 3.3 D A T A C A L C U L A T I O N A N D S T A T I S T I C A L A N A L Y S E S 51 4. RESULTS 54 4.1 Fish Growth - Lengths and Weight Measurements 54 4.2 Dietary Performance Parameters 64 4.3 Hematocrit Measurement 70 4.4 Seawater Challenge Test and Fish Appearance 70 4.5 Proximate Analyses 73 4.6 T 3 and T 4 Hormone Measurement 78 4.7 Vitellogenin Assay Results 80 4.8 Chemical Analyses of 4-NP 80 5. DISCUSSION OF RESULTS 92 5.1 General Observations of Fish Feeding and Behaviour 92 5.2 Observations of Fish Growth 93 vi 5.3 Effects of Estrogen (E2) on Growth 96 5.4 T 3 and T 4 Determinations '. 97 5.5 Determination of 4-NP in Samples 98 5.6 Concluding Remarks for Experimental Section 99 6. EXAMINING RISK - WITHIN AN ECOSYSTEM CONTEXT 101 7. THE EDC CONTROVERSY 106 Risk Society 112 Perceived Risk 115 Stigma 118 Risk Communication 120 8. EXAMINATION OF THE PERCEPTION OF RISK 127 The E D C Controversy Revisited 131 9. CONCLUSIONS 136 10. FURTHER RESEARCH 139 11. REFERENCES CITED 141 12. ADDITIONAL BIBLIOGRAPHY 151 APPENDIX A - Analytical Data - Proximate composition 158 APPENDIX B - Statistical Summaries 179 APPENDIX C - Questionnaire 181 APPENDIX D - 10th Annual SET A C Europe 4-NP Poster 184 vii List of Tables Table 2.1: Toxicity Endpoints for Nonylphenol to Fish, Invertebrates and Algae. (UK assessment report, 1997) 28 Table 3.1 - Ingredient composition of basal diet 36 Table 4.1 - Summary of fish performance and survival - fresh water phase 63 Table 4.2 - Summary of fish performance and survival - sea water phase 63 Table 4.3 - Summary of proximate composition of treatment diets 74 Table 4.4 - Proximate composition of fish - fresh water exposure phase 74 Table 4.5 - Proximate composition of fish - sea water exposure phase 75 Table 4.6 - Fresh Water PhaseT3 and T 4 values 78 Table 4.7 - Sea Water PhaseT3 and T 4 values 79 Table 4.8 - Nonylphenol and related alkylphenols found in fish food pellets and diet components 84 Table 4.9 - Comparison between tissue concentrations relative to tissue depth 85 Table 4.10 - Summary of 4-NP analyses by ESI L C - M S - fresh water 87 Table 4.11 - Summary of 4-NP analyses by ESI LC-MS - sea water 88 Table 4.12 - Determination of 4-NP in water samples , 90 Table 8.1 - Profile of questionnaire respondents 128 Table 8.2 - Summary of questionnaire responses 129 viii List of Figures Figure 2.1 - The general framework for ecological risk assessment ..26 Figure 2.2 - Structures and abbreviations of nonylphenol and estrogenic metabolites of nonylphenol polyethoxylates 27 Figure 3.1 - Task versus time matrix of experimental phase of research 35 Figure 4.1 - Weight and length distribution histograms - May Sampling 55 Figure 4.2 - Weight and length distribution histograms - June Sampling 57 Figure 4.3 - Weight and length distribution histograms - August Sampling 59 Figure 4.4 - Weight and length distribution histograms - September Sampling 61 Figure 4.5 - Comparison of percent survival between treatments 64 Figure 4.6 - Comparison of DFI and FE between treatments 65 Figure 4.7 - Comparison of SGR & G E U between treatments 66 Figure 4.8 - Comparison of PER & PPD between treatments 67 Figure 4.9 - Condition factor and hepatosomatic index - fresh and sea water phases 69 Figure 4.10 - Hematocrit Index at end of fresh water phase 70 Figure 4.11- Plasma sodium concentration - June 21, 2000 71 Figure 4.12- Plasma potassium concentration - June 21, 2000 72 Figure 4.13 - Retention of parr marks after four weeks growth in sea water 73 Figure 4.14- Summary of proximate composition of fish - protein & lipid 76 Figure 4.15 - Summary of proximate composition of fish - moisture & ash .77 ix Figure 4.16 - Selected ion chromatogram for recovery, internal and calibration standards 81 Figure 4.17 - Example of selected ion chromatogram for one water sample with internal and calibration standards 82 Figure 5.1 - Hormone changes during parr-smolt transformation (Hoar & Randall, 1988) 97 Figure 6.1 - Ecosystem approach to considering risk 102 Figure 7.1- Influence diagram of key interest groups in the endocrine disruption controversy 107 4 X List of Abbreviations 4-NP 4-nonylphenol APnEO alkylphenol ethoxylates A N O V A analysis of variance B A T E A best available technologies economically achievable BBC British Broadcasting Corporation B K D bacterial kidney disease CBC Canadian Broadcasting Corporation CEPA Canadian Environmental Protection Act DES diethylstilbestrol DFI dry food intake D M dry matter DMQ double Mil l i -Q filtered (Millipore) water DP decylphenol E2 17p-estradiol EDC endocrine disrupting compound ELISA enzyme-linked immunosorbent assay ESI electrospray ionization FE feed efficiency FW fresh water GEU gross energy utilization xi HPLC high performance liquid chromatography HP Heptylphenol ID inside diameter 11 -k-T 11 -ketotestosterone Kow water-octanol partition coefficient LC-MS liquid chromatography - mass spectrometry LC50 50% lethal concentration LOEC lowest observed effect concentration MS222 tricaine methanesulfonate NOEC no observed effect concentration NPE nonylphenol ethoxylates NPEC nonylphenol polyethoxycarboxylates NPEO nonylphenol polyethoxylates OP Octylphenol 4-tert-0 4-ter/-octylphenol P A H polycyclic aromatic hydrocarbons PCB polychlorinated biphenyls PBS phosphate buffered saline PER protein efficiency ratio PPD percent protein deposited PSL priority substances list QA/QC Quality assurance/quality control xii RIA radioimmunoassay Std Dev standard deviation SGR specific growth rate SIM selected ion monitoring SW sea water T testosterone T 3 triiodothyronine T 4 thyroxine TBT tributyl tin T D M toluene: dichloromethane: methanol TSMP Toxic Substances Management Plan TPP tert-pentylphenol Vg vitellogenin YES yeast estrogen system xiii 1 P R E F A C E The closest approximation to solutions to complex problems is realized through the synthesis of knowledge originating from a vast array of disciplines and actors. A l l of these contributors view the world through different lenses and thus orchestrating this synthesis is perhaps the greatest challenge to addressing contemporary social, economic and environmental problems. Multidisciplinary approaches for solving problems that connect scientific assessment to socially driven action are key to successful integration of knowledge needed to make wise decisions when the number of worldview permutations could equal the number of individual actors that are involved. I accept a worldview that describes the world as a series of interconnected systems, viewed in more holistic terms as an ecosystem. Modernity in the developed world has been greatly influenced by technology and human ingenuity and has evolved into a risk society, as first proposed by Ulrich Beck and Anthony Giddens. Within the framework of their model, the process of recognizing and understanding the link between social definitions within the scientific examination of risk, in the context of environmental problems, is critical to addressing the consequences of an ever-growing number of hazards that people face in contemporary risk society. This process is best illustrated through the use of a case study. Over the past two decades, a significant body of scientific literature has grown that suggests that the endocrine system of many organisms is particularly sensitive to very low levels of a wide variety of contaminants. It is increasingly recognized that endocrine disrupting compounds do not follow the normal pattern on which standard toxicological testing is based. This is because their effects may be seen only during critical windows within the life history of the organism. This thesis is designed to be a synthesis of a multidisciplinary approach to examining some aspects of the endocrine disruption issue. Current scientific understanding of the mechanism of biochemical, physiological and whole organism responses to exposure to the class of compounds that can moderate normal hormone function is limited. At the same time, there is a myriad of complex social considerations peripherally connected to the endocrine disrupting issue that inspire examination from a sociological perspective. In this thesis, I will describe the results of xiv one experiment that investigated the effect of dietary exposure of of juvenile coho salmon to 4-nonylphenol on their ability to successfully transfer into seawater. Also, I describe some of the social considerations that alter perceptions of endocrine disruption risk in the context of risk society. xv Acknowledgments There are many individuals who have been instrumental in the research and preparation of this thesis and have assisted me in many ways. My sincerest thanks go to my thesis committee Dr. Ken Hall, Dr. Dave Higgs, Dr. Michael Ikonomou, Dr. Les Lavkulich and Dr. Ralph Matthews. I thank Janice Oakes, John Blackburn, Scott Brown, Nahid Roshandeli, Joanna Wieruszewski, Karen Kinnee and Trevor Davies for their help sampling and analyzing my fish. I owe very many thanks to Natasha Hoover for analyzing the 4-NP concentrations in the fish and to Mahmoud Rowshandeli for determining the gross energy content of the fish diets and assisting in all aspects of my experiment. I am very grateful to Terre Satterfield, Hadi Dowlatabadi and May du Monceau for teaching me new ways to think. Shannon Balfry and Tony Kozak assisted with the statistical analyses. Mike Healey, Raul Pacheco, Glenys Webster, Fred Koch, Hans Schreier, Sandra Brown, Gina Bestbier have helped tremendously with guidance, editorial review and poster preparation. M y special thanks go to Steve Clark for his help throughout this whole process. Experiments conducted in this research were supported by funds provided by Steve Samis and John Pringle of the Canada Department of Fisheries and Oceans. I, as well, sincerely appreciate the scholarship award from the West Vancouver branch of the Canadian Federation of University Women. xvi This one is for you, Mom. You see, I really do finish everything, Eventually. In Memory of: Margaret Ada Prescott Keen (1929-1978) xvii INTRODUCTION " ...the argument for the indirect role of pesticides in cancer is based on their proven ability to damage the liver and to reduce the supply of B vitamins, thus leading to an increase in the endogenous estrogens, or those produced by the body itself. Added to these are the wide variety of synthetic estrogens to which we are increasingly exposed - those in cosmetics, drugs, foods and occupational exposures. The combined effect is a matter that warrants most serious concern. " Rachel Carson Silent Spring, 1962 Rachel Carson brought the issue of chemicals that could potentially disrupt normal endocrine system function in organisms to the attention of the world in her landmark treatise on environmental and human health hazards of pesticides in 1962. For about forty years, scientists have suspected that environmental exposure of organisms to natural and/or synthetic estrogens have deleterious effects on their normal endocrine functions. The scientific community and the lay public answered Rachel Carson's call for concern about the deleterious effects, including the potential endocrine disruption capability of these compounds, of contaminants in the environment. There was, however, roughly, twenty years of delay before considerable action was universally apparent to address impacts of environmental pollution. In 1996, another book was published that drew the world's attention to the potential of many environmental contaminants that could mimic the hormonal behaviour of natural estrogen in organisms including humans. Theo Colburn collaborated with a professional science journalist, Dianne Dumanoski and Pete Myers of the W. Alton Jones Foundation to publish Our Stolen Future. This became one of the most prominent popular books that promoted awareness and concern for an environmental risk since Silent Spring. This time, worldwide reaction was apparent within one year. Recent years have seen growing interest in the scientific community to conduct studies of reproductive effects of exposure to environmental contaminants. Toxicologists, endocrinologists, epidemiologists and ecologists have all brought considerable attention to the 1 potential hazardous effects of endocrine disrupters on human and ecosystem health. Several xenobiotic substances belonging to this class of environmental pollutants have been demonstrated to adversely affect the normal functions of the endocrine and reproductive systems in fish, reptiles and animals. These substances have been shown to inhibit the action of hormones, impair reproduction and adversely affect immunocompetance and nervous system function in organisms. In some cases, contaminants that have sex steroid activities but nevertheless are structurally very different from the hormone are hypothesized to cause profound disruptions in the endocrine systems in laboratory fish and rodents by: • mimicking naturally occurring endogenous hormones such as estrogens and androgens in higher organisms, • antagonizing normal, endogenous hormones, • altering the pattern of synthesis and metabolism of natural hormones, and • modifying hormone receptor levels. It has been hypothesized that such substances could elicit similar responses in humans. The group of environmental pollutants that include industrial chemicals, naturally derived chemicals and estrogenic pharmaceuticals has now become the focus of considerable research in environmental toxicology and medicine in the last few years. One suggested reason for estrogenic activity caused by substances other than naturally occurring 17p-estriadiol is the apparent low specificity of the human estrogen receptor possibly due to the binding pocket being nearly twice as large as the molecular volume of 17p-estriadiol (Brzozowski et al., 1997). It must be stressed that although there have been adverse trends found with respect to some aspects of reproductive performance of various organisms; the causes of these trends are largely unknown. A substantial body of evidence supports the hypothesis that contaminants that are present in industrial discharges such as pulp and paper effluents, municipal wastewater and agricultural runoff have deleterious effects on the reproductive health of aquatic organisms. Particular attention has focused on xenobiotic chemicals which demonstrate estrogenic and/or androgenic biochemical behaviour as well as the synthetic estrogens derived from the production of oral 2 contraceptives. Compounds such as alkylphenol ethoxylates and their derivatives, phthalates, pesticides, polychlorinated biphenyls, chlorinated dibenzodioxins and difurans, tributyl tin and bisphenol A have been demonstrated to modulate the endocrine system and affect reproduction of organisms (US National Research Council, 1999). Increasing concerns over persistent bioaccumulative chemicals and the appropriate means of assessing toxicology and ecology support the need for further study of endocrine disrupting compounds. The endocrine system plays a crucial role, together with the immune and nervous systems, in regulating an organism's normal health and physiological functions. Through the production of hormones by the pituitary gland, thyroid, pancreas, adrenal and male and female gonads, the endocrine system influences regulatory, developmental, reproductive, growth and homeostasis mechanisms. There is considerable homology in the structure of hormones across classes of vertebrates with some hormones, such as thyroid hormones and catecholamines appearing the same from fish to humans (Lister & Van Der Kraak, 2001). Although steroid hormones are biosynthesized from common precursors, there are probable differences between various species in the predominant metabolites. Some xenobiotic substances have demonstrated endocrine disruption in humans. Although clearly defined endpoints have not been established, some human health effects include disruption of normal sexual differentiation, fertilization and pregnancy in females, endometriosis in women, human breast cancer and human prostate cancer, reduction in human sperm production and higher than normal incidence of genital defects in human males. Only one chemical has been proven to cause serious endocrine disruption in humans. Diethylstilbesterol (DES) is a pharmaceutical product that was administered to pregnant women from 1948 to 1972 to prevent miscarriages. Offspring of these women demonstrated reproductive abnormalities and increased incidence of reproductive system cancers. Environmental effects of endocrine disrupting compounds are well documented. Sumpter and Jobling have conducted extensive research on the Thames receiving environment and found induction of vitellogenin, an egg precursor protein not usually in elevated concentrations in male fish, in both bioassay and in situ exposure of fish to municipal sewage (Sumpter & Jobling, 1995). Laboratory and field experiments with tributyl tin (TBT) have suggested that environmental exposure of some snail species to antifouling paints containing TBT can induce significant hormonal responses. A number of mollusk species have been used in dose-response 3 exposure to sublethal levels of TBT and it appears these compounds cause irreversible induction of male sex characteristics in females (condition known as imposex). Among the chemicals that have been demonstrated to disrupt the normal functioning of the endocrine system in a variety of organisms is 4-nonylphenol. 4-Nonylphenol (4-NP) is a precursor in the production of nonylphenol ethoxylates (NPE). These are nonionic surfactants found most often in cleaning products and industrial processes. Release of effluents containing NPE contaminants is of particular concern since their breakdown products include 4-nonylphenol, octylphenol, and a wide variety of carboxylates and ethoxylates which may also have significant endocrine disrupting potential to organisms. Primary end uses of NPE compounds include cleaning products, plastic and elastomer manufacturing, textile processing, pesticide emulsifiers, pulp and paper production, and personal care products (Talmage, 1994). The principal pathway of 4-nonylphenol to the aquatic or marine environment is from industrial and municipal wastewater discharges. Since some effects of exposure of organisms to endocrine disrupting substances are not observable until considerably later in their life cycle, it is impossible to reliably track effects of these substances in the human population with the exception of studying the medical records of individuals that have been subjected to accidental occupational exposure. For the purposes of this study, the monitoring of the responses of organisms with shorter life cycles is the most practical source of information which can be used as a model to elucidate effects in organisms with longer life cycles but this approach can never be extended as a definitive description of general biological behaviour. Fish, amphibians and rodents have proven to be the most practical organisms for which biological models of responses to contaminants can be constructed. Accidental exposure to wild populations of organisms that have longer life cycles of contaminants provides strong supporting evidence for validation of the effects demonstrated under laboratory conditions. It is curious that although considerable effort on the part of the scientific community is focused on examination of the effects of endocrine disruption, the public remains either largely unaware or ambivalent to the risks posed by voluntary or involuntary exposure to endocrine disrupting substances. Media attention cyclically addresses concerns linked to human health or ecological risks, but the transmission of the most current scientific knowledge on this topic is sporadic and unsustained. 4 Frequently, media coverage of any information connected to the endocrine disruption controversy emphasizes the, yet unproven, hypothesis that there are temporal declines in human sperm quality and number. This raises some important questions regarding public perception of a potentially devastating risk, the trust in scientific knowledge when considerable uncertainty surrounds many aspects of the endocrine disruption controversy, and the way in which policy makers respond when faced with a man-made risk which has large scale implications on human reproductive health. In the experiment of this thesis, I examine the potential effects of one chemical that has previously been demonstrated to elicit endocrine disruption in fish. Unlike other studies in fish of this compound, 4-NP was administered via the diet since salmon in nature could be ingesting prey containing the contaminant. 4-nonylphenol was selected for this study because of its ubiquitous existence in aquatic ecosystems and the coho salmon is a particularly important native species. Hence, the information generated in this thesis could be valuable to fisheries managers. The second part of the thesis serves as a focus from which to explore some aspects of the nature of controversy linked to endocrine disruption. Also it enables the development of some hypotheses that help to explain the apparent imbalance between public perception of the risk of exposure to endocrine disrupting compounds and the degree of seriousness now accepted by the scientific community. 5 BACKGROUND "My biographical situation defines the way in which J locate the arena of action, interpret its possibilities and engage in its challenges. " Alfred Schutz Since the early 1990s a substantial body of scientific evidence has been published to support the observations that many environmental contaminants can interfere with normal developmental or reproductive processes in organisms by interacting with estrogen receptors. There is an ever-increasing weight of evidence that natural or synthetic estrogenic compounds are responsible for effects seen in wild populations and in laboratory experiments. Some chemicals are now known to elicit estrogenic response in several species and risk assessments involving a wide variety of environmental contaminants are currently being conducted to screen those compounds whose tendency to be hormonally active is yet to be determined. The term "hormonally active agents" has been proposed to describe the classification of endocrine disrupting compounds without regard to their mechanism of action (US National Research Council, 1999). Although compounds that demonstrate estrogenic activity dominate the very long list of substances, this classification also includes chemicals which cause anti-estrogenic, androgenic or secondary hormonal responses in organisms. Universally, however, the scientific community accepts the term "endocrine disrupting compounds or EDCs" to describe the very large category of natural or synthetic substances which have been implicated to demonstrate some sort of hormone-like behaviour in organisms. 2.1 Endocrine Disrupting Compounds in the Environment Description of the historic evolution of the science supporting the endocrine disruption hypothesis is incomplete without citing the well-known experimental evidence provided by Lake Apopka alligator studies (Guillette et al., 1994, 1995, 1996b, 1996c; Guillette and Crain, 1996a). Louis Guillette Jr. has championed the cause of bringing the EDC hypothesis to public attention and is frequently quoted in the popular press on the subject of reproductive abnormalities seen in wild male alligators in a previously contaminated lake. Studies of Lake Apopka alligators during 6 the past decade have revealed that gonadal abnormalities and abnormal sex steroid concentrations in the blood of mature alligators may be the result of accidental environmental exposure to o,p-DDT when the alligators were juvenile. Observations seen in the wild population of alligators were reproduced under laboratory experimental conditions (Guillette et al. 1995) and since the life cycle of alligators is relatively long, these studies are often cited as a basis for the extension of exposure to endocrine disrupting compounds as a threat to human male reproductive health. 2.2 Implications of E D C Exposure on Human Health The pharmaceutical use of the synthetic estrogen diethylstibestrol (DES) is often cited as a model for effects of endocrine disrupting compounds on human health and environmental exposure. For nearly twenty years DES was administered to pregnant and post-menopausal women. Prior to the 1971 ban by the US Food and Drug administration, DES had found extensive use in agriculture for over thirty years. The intergenerational health effects of this chemical prompted considerable research related to possible risk to human health of endocrine disrupting substances. Human health effects of in utero exposure to DES included development of clear-cell carcinoma of the vagina and although DES exposed children were born without apparent reproductive abnormalities, upon reaching puberty or adulthood, these children developed evidence of carcinomas and reproductive dysfunction (Sharara et al., 1998). Studies were conducted over twenty years to examine the health, mortality and reproductive outcomes of US veterans of the Vietnam War involved in Operation Ranch Hand, the unit responsible for the aerial spraying of herbicides, including Agent Orange (Michalek and Ketchum, 2000). Agent Orange contains a significant concentration of 2,3,7,8-tetrachlorodibenzo-p-dioxin (dioxin). Occupational exposure to dioxin has led to a small but significant increase in human mortality due to prostate cancer (Keller-Bryne et al., 1997). Based on these findings, it has been hypothesized that personnel exposed to high concentrations of dioxin during the Vietnam conflict may have higher mortality due to prostate cancer. Consensus has not been reached within the scientific community as to whether exposure to high concentrations of dioxin through aerial spraying and maintenance of aircrafts containing Agent Orange is related to the increased incidence of prostate cancer in the US male Vietnam veteran population. 7 Some persistent organic pollutants have been demonstrated to compromise reproductive, immune and neurological systems of fish and wildlife but one ongoing study provides some interesting observations regarding human exposure to pesticides known to elicit endocrine disruption in organisms. Elizabeth Guillette et al. (1999) conducted a study of an indigenous population of people residing in the Yaqui Valley, Sonora Mexico. In an ecosystem that was traditionally a desert, technological farming was introduced in the early 1940's with the coming of the green revolution to Mexico. The local community was given the option of adopting the practices of technological farming which included government subsidized application of commercial pesticides (primarily DDT) or maintaining the traditional farming and ranching lifestyle. Some of the Yaqui Valley Indian people moved away from the immediate vicinity that employed pesticide application as part of their agricultural practices but both groups, being of the same cultural and genetic history retained the same diet, social customs and infrastructure although in different locations in the same valley. Since 1995, E. Guillette et al. (1999) had been conducting interviews and examining the health of the population of people exposed to pesticides and they compared these observations to those of the population who retained traditional farming practices and were not excessively exposed to pesticides. In this regard, they observed behavioural differences (i.e. coordination, memory and endurance effects could be detected), especially between 4 and 5-year-old children, as well as higher incidences of birth defects colds, allergies, diarrhea and headaches in the exposed population when compared to the control population located about 100 km from the exposed group. In Canada, particularly in Arctic ecosystems, there is concern that the presence of environmental contaminants that have been shown to exert endocrine disruption effects in fish and wildlife may pose a potential threat to the health of the people consuming traditional foods in northern communities (Phare et al., 2001). Further, sufficient scientific evidence has been collected that shows that contaminants in the Great Lakes basin have the potential to negatively affect the health of First Nation people who have a culture and dietary reliance on freshwater fish for subsistence. Moreover, it has been observed that the children of women in Inuit communities in Northern Quebec that consume fish containing measurably higher levels of PCBs have significantly higher incidence of respiratory and ear infections than children in Southern Quebec (Guidotti & Gosselin, 1999). It has been hypothesized that endocrine disrupting compounds such as PCBs can suppress immune response in humans. Hence, there is particular concern that 8 many First Nations communities which are located in rural areas that close to pulp and paper mills, mines and smelters may be at higher risk of being exposed to endocrine disrupting compounds. 2.3 Environmental Effects of EDCs on Other Organisms Exposure of marine mammals to persistent organic pollutants, many of which are organochlorine compounds, is a particular concern in protecting marine ecosystem health. Some of the organochlorine contaminants of concern that have been demonstrated to bioaccumulate in dolphins and whales are endocrine disrupting compounds. Studies of striped dolphins and whales living in the North Pacific Ocean have found that tissue concentrations of some organochlorine compounds have been 10 million times greater than the respective concentrations of these contaminants in local seawater (Tanabe, 1998). Expanded tropical use of organochlorine pesticides increases pollution load to oceans and the pollutants, in turn, are transported by currents as well as through atmospheric deposition to oceans throughout the world. This situation exacerbates the possibility that marine mammals will be exposed to significant concentrations of these contaminants. Marine mammals thick layers of adipose tissue that tend to accumulate lipophilic contaminants. Further, they also tend to transfer significant concentrations of contaminants to the next generations through lactation and the bioaccumulation effect is magnified since they appear to have a low ability to degrade some of the contaminants. For these reasons, whales, dolphins and other marine mammals are thought to be especially vulnerable to the risks posed by exposure to contaminants that have demonstrable endocrine disrupting capabilities. Estrogenic contaminants represent an important class of toxicants to birds since differentiation of the avian reproductive system is estrogen dependant. Phenotypic differentiation of an embryo into a male occurs unless specific female genes are translated and 17-(3-estradiol is synthesized to cause development of the gonad into an ovary. Much of the initial data that led to the endocrine disruption hypothesis was based on studies of birds in the Great Lakes region (Colburn, 1996). A n increased incidence of female-female pairing of gulls that was observed in California and the Great Lakes was thought to be related to alterations of the sex ratio observed in birds following exposure to some organochlorine contaminants such as DDT, PCBs and dioxin. Supernormal clutch size was also hypothesized to be related to the apparent sex skew in several bird species. 9 Concerns regarding declines in the populations of salamanders, frogs and toads have intensified the research focusing on endocrine disrupting effects of contaminants on amphibians. Amphibians are particularly vulnerable to contaminant exposure due to the compounding effect of multiple pathways of exposure from terrestrial and aquatic sources. Gross deformities and mortalities of wild amphibian populations in North America has been well documented (Schmidt, 1997). The role of hormones such as 17-B estradiol, testosterone and corticosterone is under continued investigation but correlation between observed deformities and hormonal modification required further research (Hayes et al., 1997). 2.4 Assessing Estrogenicity of Contaminants Since the 1991 first Wingspread meeting (Berne et al., 1992), one of the key mandates of global research on endocrine disrupting substances has been the development and validation of reliable tests to evaluate the estrogenicity of contaminants. To date, several in vitro and in vivo methods have been under investigation for screening contaminants for endocrine disrupting potential. Although some of these methods are accepted as indicators of estrogenic activity of compounds, at present these tests are limited largely to the research community and there are extremely few analytical laboratories that offer any of the accepted screening tests on a commercial basis. Since regulatory mechanisms dealing with compounds that are suspected to cause endocrine disrupting effects are in their infancy, standard procedures for the testing of estrogenic activity of contaminants are not universally accepted yet. Based on best available science to date, some procedures have earned global agreement as acceptable screening tests for endocrine disrupting potential of new chemicals where registration is being sought. Research continues to focus on the investigation of new tests for estrogenic activity and the practical application of these assessment tools in a regulatory context. Assessment of vitellogenesis has become a popular method to detect exposure of oviparous organisms to estrogenic contaminants to the extent that it is a protocol recommended for Tier 1 screening by the US E P A (Christiansen et al, 2000). The measurement of plasma vitellogenin, an egg precursor protein, has been shown by a number of researchers to a sensitive biomarker of exposure of organisms to estrogens (Sumpter & Jobling, 1995; Kime, 1999; Sherry et al., 1999). Biosynthesis of vitellogenin occurs in the liver where interaction of estrogen with the estrogen receptor (ER) causes initial production, release into the bloodstream, modification and then its deposition as yolk in oocytes. Normally, vitellogenin is present in very small concentrations or 10 absent altogether in the plasma of male fish (Sumpter and Jobling, 1995). Male fish carry the gene responsible for production of vitellogenin but male fish usually lack sufficient endogenous estrogens to trigger the expression of this gene to produce vitellogenin. Exposure of male fish to exogenous estrogens or compounds that mimic the action of estrogen can markedly induce vitellogenin in the plasma where its level remains elevated due to lack of proteins to clear the protein (Nicolas, 1999; Mommsen and Walsh, 1988). Assay of plasma vitellogenin (VG) is accepted as a biomarker for exposure of fish to estrogenic substances and several studies have documented a correlation between vitellogenin induction and adverse reproductive effects in male and female fish (Harries et al., 1997; Janssen et al., 1997; Panter et al., 1998). In addition to assessment of vitellogenin induction, the OECD working group on Endocrine Modulators and Wildlife (1997) recommended several endpoints that could evaluate estrogenic activity using fish. These endpoints could be applied to short term bioassays and longer term sublethal bioassays using species such as rainbow trout (Oncorhynchus mykiss), goldfish (Carassius auratus), carp (Cyprinus carpio), channel catfish (Ictalurus punctatus), Japanese medaka (Oryzias latipes), Zebrafish (Danio rerio), and fathead minnow (Pimephales promelas). For short term tests, sex steroid levels (17p-estradiol (E2), Testosterone (T), and 11-ketotestosterone (11-k-T)), vitellogenin and gonadosomatic index were recommended as providing valuable information about exposure of early life stages of fish to estrogenic substances. Endpoints that are acceptable indicators of exposure of mature fish to endocrine disrupting substances include mating behaviour, fish colouration, time to first hatch, fecundity, sperm production, sex steroid concentrations, gonad histopathology and vitellogenin induction in males. The E-screen is a well-known cell proliferation assay that was introduced by Soto et al. (1995). This assay uses the human breast cancer cells, MCF-7, which are estrogen sensitive. Thus a substance that induces the mitotic division leading to cell proliferation in these cells displays estrogenic activity. When MCF-7 cells are inoculated in ovariectomized hosts, they remain quiescent or when these cells are cultured in media supplemented with charcoal-stripped human serum, their growth is inhibited. However, when these cells are cultured in media in the presence of estrogen or an estrogen-like substance, the cells proliferate. Although this test is sensitive and based on a relatively simple concept, its species specificity and lack of predictiveness for in vivo effects in wildlife has been criticized (US National Research Council, 1999). This test has recently been demonstrated to a sensitive method for quantitative 11 determination of total estrogenic activity in effluent samples from municipal wastewater treatment plants (Korner et al., 1999). A bioassay using yeast estrogen system (YES) involving the human estrogen receptor (hER) is now often used to evaluate the estrogenic activity of substances. This method has been proposed as sensitive and applicable for environmental monitoring since it can be quantitative and satisfies several criteria for acceptance as a regulatory test (Graumann et al., 1999). This assay technique used radio-ligand binding of human estrogen receptor expressed in specific yeast strains that are transfected with the appropriate yeast expression plasmid. The yeast cells are cultured, harvested, measured spectrophotometrically, and compared against a standard protein calibration curve. 2.5 Contaminants that Demonstrate Estrogenicity A very wide variety of compounds both natural and manmade have been demonstrated to be weakly estrogenic. These include natural estrogens, phytoestrogens, mycoestrogens, certain pesticides and herbicides, some polychlorinated biphenyls (PCBs), certain combustion pollutants, some plasticizers and some surfactants and their breakdown products. Of the more than 60,000 man-made compounds that are routinely found in municipal wastewater, many substances have been identified as having endocrine disrupting potential. Particular attention has been focused on xenobiotic chemicals which demonstrate estrogenic and/or androgenic biochemical behaviour and the synthetic estrogens that have been derived from the production of oral contraceptives. Compounds such as alkylphenol ethoxylates and their derivatives, phthalates, pesticides, polychlorinated biphenyls, chlorinated dibenzodioxins and difurans, tributyl tin and bisphenyl A have been demonstrated to modulate the endocrine system and affect reproduction of organisms. Hence, the main compounds that exert this effect include: • natural animal hormones (includes estrone, 17p-estradiol,) • synthetic hormones (eg DES and ethynl estradiol) • organochlorine pesticides (PCBs, dioxins, DDT and metabolites) and organotins • bisphenol A • alkylphenol ethoxylates and degradation products 12 • phytoestrogens, phytosterols and mycoestrogens • some phthalates and phthalate esters 2.6 Endocrine Disruption in Fish Many members of the scientific community have examined the effects of endocrine disrupting substances on a variety of fish species. Some of the best known research concerning the effects of endocrine disruption on fish was conducted by researchers from the United Kingdom (Sumpter and Jobling, 1995, Harries et al., 1996, 1997; Purdom et al., 1994). They hypothesized and subsequently demonstrated that the presence of estrogenic substances in riverine ecosystems exerted adverse effects on fish populations. The United Kingdom has been at the forefront of EDC research for the past ten years and much public attention regarding the potential for municipal wastewater effluents to cause endocrine disruption can be credited to their systematic investigation of possible estrogenic or androgenic effects of contaminants found in receiving streams, effluents and sludges. Tyler and Routledge (1998) cited a study that provided evidence over twenty years ago of intersex condition (male fish with eggs present in testicular tissue), or hermaphroditism, in wild populations of roach living downstream of sewage treatment facilities. Jobling et al. (1998) have since demonstrated an apparent relationship between the concentration of sewage effluents in U K rivers and the percentage of intersex fish in wild populations. Feminization of young male carp (Cyprinus carpio) was observed when they were exposed to 4-terf-pentylphenol (TPP) during sexual differentiation, 50 days post hatch (Gimeno et al., 1998). Histopathological analyses determined that the average number of primordial germ cells in the testes of male carp was reduced (PO.001) when they were exposed to 17p-estradiol and that the extent of reduction was dose dependent when they were exposed to a concentration range of TPP. Spermatogenesis was found to be impaired in the testes of the TPP exposed fish and vitellogenin was significantly elevated in the plasma of male carp exposed to the highest concentration (256 u.g/1) of TPP. After 30 days of exposure of male carp to TPP, oviduct formation was noted instead of induction of a male vas deferens and 17p-estradiol was found to exert the preceding effect in a shorter time (20 days). Perhaps the best documented Canadian studies related to environmental exposure effects to endocrine disrupting substances are those of Munkittrick, Servos, and colleagues (1997-2001). They investigated the effects of pulp and paper effluents on the reproductive health of wild fish 13 populations. McMaster and Munkittrick (2001) summarized the findings of this research. Over the past decade it was determined in both laboratory and field experiments that certain chemicals in pulp mill effluent have the potential to affect endocrine homeostasis. Several studies that have monitored a series of Canadian pulp mills in the Moose River basin, the Ottawa River and the large bleached kraft pulp mill at Jackfish Bay on Lake Superior, have shown a correlation between improved fish performance over time and adoption of improved process changes and water treatment practices by the mills (Servos et al., 1997; Van der Kraak et al., 1998). Although these process changes have resulted in the improvement of the reproductive health of wild fish at some sites, persistent responses such as reduced gondal development and decreased steroid levels in some species continue to be seen at several Canadian sites. A three-year, multi-species study that was undertaken to examine the reproductive-endocrine responses in fishes exposed to a New Zealand pulp mill effluent has suggested that this effluent is eliciting effects that may be androgenic rather than estrogenic (Van den Heuval, 2001). Short-term bioassays using mosquito fish, goldfish and immature rainbow trout demonstrated no significant vitellogenin induction in some species at concentrations varying between 10 and 70% effluent but longer-term mesocosm studies with mature rainbow trout have suggested that an androgenic mechanism may be responsible for the observed endocrine disruption effects. Experiments with mosquito fish demonstrated that the particular pulp mill effluent was capable of inducing male secondary sexual characteristics in females. These investigations continue to examine endocrine disrupting effects that may be linked to androgenic responses possibly caused by wood sterols present in the effluent. In a study that was conducted to assess the toxicity of alkylphenols to killifish (Fundulus heteroclitus), Kelly and Di Guilio (2000) demonstrated that waterborne nonylphenol (NP) and 4-terr-octylphenol (A-tert -O) caused both lethal and sublethal developmental abnormalities. The larval exposure of the fish to the NP and A-tert-0 determined that the lethal concentrations (NP larval: 96 hr LC50 = 204 pg/L) were an order of magnitude less than those concentrations that were lethal to embryos (NP embryo: 96hr LC50 = 5 mg/L). These studies also indicated that tamoxifen, an estrogen receptor antogonist was effective in preventing embryo lethality for NP and A-tert -O. It has often been suggested that there is a critical window in the life history of animals where they are most sensitive to exposure to endocrine disrupting compounds. Indeed, exposure of the 14 fish to these compounds during the early part of their life cycle may result in the greatest adverse effects on the development and reproductive capacity of the fish in later life. Maack and Segner (2001) used zebrafish (Danio rerio) to look for a particularly sensitive period when estrogen exposure would cause deleterious effects. They observed histopathological differences and induction of vitellogenin in fish exposed to estrogen between 40 to 70 days post fertilization. In a multigenerational study of fathead minnows (Pimephales promelas) exposed to Bisphenol A, the British research team led by Sumpter et al. (2001) suggested that the population success of fathead minnows and similar aquatic organisms may not be affected by their exposure to the concentrations of the contaminant that currently exist in the environment. They found effects of Bisphenol A on fish survival, growth, reproductive fitness and hatchability at concentrations of 640 pg/l and higher. Spermatogenesis was impaired at lower concentrations although hatchability was affected only at 160 ug/L and greater. They determined that the no observed effect concentration of this chemical (NOEC) for the purpose of risk assessment was 16 pg/l. Despite effects on spermatogenesis, environmental concentrations of bisphenol A are typically in the range of 0.001 to 0.1 pg/l, which are considerably less than the observed NOEC. This study stresses the importance of interpreting risk assessment values such as NOEC in light of the more ecologically relevant effects at the organism and/or population level. Gonadal development and endocrine responses have been demonstrated in Japanese Medaka (Oryzia latipes) exposed to o,/?'-DDT in water and through maternal transfer. Since various isomers and metabolites of DDT including o,p'-DDT had been shown previously to be relatively potent estrogen agonist, studies by Metcalfe et al. (2000) found that continuous waterborne exposure of medaka to nominal concentrations of o,p'-DDT (5, 10 and 50 pg/L) over the period of gonadal differentiation induced intersex condition. They also found that when medaka were exposed to DDT metabolites by maternal transfer and then they were later exposed to 17(3-estradiol at 10 months of age, that hepatic vitellogenin induction in DDT exposed males was significantly greater than in control males. This finding suggests that exposure of medaka to estrogenic chemicals at a critical early life history stage may potentially induce vitellogenin following exposure of the fish to estrogenic compounds much later in the life of the fish. Juvenile medaka (Oryzias latipes have also been used in studies that have examined the reproduction, sexual differentiation and aromatase gene expression following exposure of the fish to environmentally relevant concentrations of 17-a-ethinylestradiol (Scholz and Gutzeit, 15 2000). At low 17-a-ethinylestradiol concentrations (1 and 10 ng/L), Scholz and Gutzeit (2000) found no evidence of sex reversal or detectable alteration of testicular tissue as well as no apparent changes in male fertility but at 100 ng/L exposure concentration, all X Y fish were sex reversed and they had developed an ovary. Aromatase, which is usually only present in ovaries was detected in the testis of X Y males exposed to 10 ng 17-a-ethinylestradiol/L. These results suggest that multiple biological endpoints such as gene expression must be included in the biological evaluation of the efficacy of endocrine disrupting compounds since morphological deficiencies may not always be apparent. One of the key challenges of conducting laboratory experiments on fish is directly related to maintaining control conditions so that the responses of contaminant-exposed laboratory fish can be interpreted properly. Ishibashi et al. (2001) demonstrated this fundamental difficulty in their studies of vitellogenin induction and yeast two-hybrid assays when using ovariectomized goldfish (Carassius auratus) fed a commercial diet with potential estrogenic activity. They observed vitellogenin induction in male goldfish in vivo after they consumed a commercially prepared carp diet and a trout diet. The trout diet contained 4% soybean oil whereas the carp diet contained an unspecified amount of soybean meal. A control diet was formulated without soybean bean and this was based on casein, flour, vitamins, minerals alfalfa and fish meal. Estrogenic activity in the fish was less when they were fed the soy-free diets instead of the commercially prepared diets as assessed by the yeast two-hybrid assay. This work reinforces the need to address the challenges associated with maintaining negative control groups in laboratory experiments when the fish need to be fed. In early summer 1999, van den Briel conducted a preliminary exploration of the effects of waterborne exposure of coho salmon (Oncorhynchus kisutch) to 4-NP on smoltification. This trial performed at the DFO West Vancouver Laboratory was not completed due to complications encountered with the husbandry of the coho salmon test stock. This investigation however, yielded valuable information related to optimal feeding condition for the fish and other factors that directly influenced the outcome of the diet experiments conducted in the indoor aquarium facility (van den Briel, 1999). 2.7 Summary of Coho Salmon Life Cycle Salmon represent an important traditional and recreational resource of food that forms the basis of a multimillion-dollar recreational, commercial and aboriginal fishing industry in Canada. 16 Production of farmed Pacific salmon has played an essential role in the growth of the aquaculture industry in Canada, the United States and Japan as well as in southern hemisphere Nations, such as Chile and New Zealand, to which Pacific salmon species have been introduced. In order to examine the impact of environmental pollutants on early life stages of salmon, it is necessary to have a general understanding of the salmon life cycle. The following is such an overview. Of the seven species of Pacific salmon, coho salmon known scientifically as Oncorhynchus kisutch, occur in smaller numbers than other salmonid species and are estimated to represent about 10% of the total catch (Sandercock, 1991). However, coho salmon have demonstrated particular ease to domesticate and as such, they have been readily adapted to the hatchery environment making them very suitable for salmon enhancement and fish farming. They utilize a wide variety of coastal stream habitats for spawning and are frequently found in areas that are inaccessible to other salmonids with the exception of migrant steelhead or resident cutthroat trout. Coho salmon spawn in coastal streams and the smaller tributaries of larger rivers. They seldom migrate more than 240 km upstream. They tend to migrate further than pink salmon (Oncorhynchus gorbuscha) and chum salmon (Oncorhynchus keta) but usually not as far as sockeye (Oncorhynchus keta) or chinook (Oncorhynchus tshawytscha) salmon. It has been suggested by Milne that there are two distinct types of coho in British Columbia. The "ocean" type has been described as preferring outer coastal or offshore waters, whereas the "inshore" type remains within inside waters during the saltwater phase of their life cycle (Milne, 1950). The two types are characterized by body morphological differences and it is possible that gene flow may be greater among the ocean type coho. Most coho salmon mature in their third year of life. Their incubation phase usually lasts from four to six months and up to fifteen months of rearing in fresh water is followed by a sea water growing period of about sixteen months. There are, of course, variations in this time sequence. For example, some males precociously mature and return to spawn after only four to six months in sea water and these are known as "jacks" whereas other fish who remain in fresh water for two winters. Males are usually larger than females and although the sex ratio is assumed to be 1:1 at the time of migration, differential mortality may be linked to the strong selection of larger females by the commercial and recreational fisheries. 17 In the early freshwater and marine stages of the coho life cycle, there are no apparent external phenotypic differences between males and females. Considerable differences in secondary sexual characteristics evolve with the onset of maturity. Incubation time for coho salmon eggs deposited in gravel nests, known as redds, is largely dependent on temperature. At temperatures between 2 - 5 °C, the coho egg incubation period is anywhere between 90 to 140 days and this depends on the conditions of the redd as well. Alevins are the post-hatch larval fish and these have a well-defined yolk sack. Thereafter, these alevins develop into fry whose emergence is also dependent on temperature and, to a lesser extent, on dissolved oxygen concentration. Newly emerged fry are usually about 30 mm in length. As the fry become older, they become territorial and fairly nonselective in the variety of food items that they consume. Abundant food supply and moderate water temperatures favor optimal growth of coho salmon fry. There are generally two temperature dependant growth spurts that follow a period of no growth in mid winter. Stream dwelling juveniles, known as parr, become pre-smolts and these complete their final growth phase just prior to beginning their seaward migration. Since predation is a major component in the mortality of juvenile coho, body colour plays an important role in influencing mortality due to capture by larger fish, birds and small mammals. Understanding the implications of the environmental influence on the complex physiological, biochemical and behavioral changes that constitute parr-smolt transformation requires more in-depth discussion of these processes. Anadromous salmonids undergo distinct morphological and physiological transformations prior to their seaward migration. In many salmonid species such as coho, the last intense phase of growth of parr prior to seaward migration is coupled with changes in cryptic colouration and streamlining of the body. As these individuals become physiologically adapted to life in the marine environment, a silvery layer of purines (guanine and hypoxanthine) forms in the integument, the weight per unit length (condition factor) decreases, and the fins, particularly the pectoral and caudal, develop black margins. During the smoltification process, there are attendant is responsible for elevations in plasma T4, enhanced gill ATPase activity and increased osmoregulatory ability in seawater. Seasonal changes in vertebrate physiology are timed and regulated by the neuroendocrine system (Hoar & Randall, 1988). Environmental cues related to photoperiod, lunar cycles and flooding 18 streams trigger the onset of endocrine activity in the hypothalamus, which in turn, stimulates the secretion of hormones from the anterior pituitary that control the secretion of hormones from the thyroid and interrenal glands. The hormones thought to be most involved in parr-smolt transformation are the thyroid hormones, prolactin, growth hormone, corticosteriods and possibly gonadal steroids. The most obvious morphological change that occurs during parr-smolt transformation is the disappearance of the brightly coloured pigment spots and narrow bars crossing the lateral line of the body (referred to as parr marks). These markings are formed by a variety of chromatophores in which the distribution of pigment granules can change cryptically. In contrast to the body shape of parr and pre-smolts, the smolts have relatively streamlined bodies and they weigh more per unit length. During smoltification, body lipids decrease quantitatively and they undergo qualitative changes as well. The reflective silver appearance of smolts is due to the synthesis of two purines: guanine and hypoxanthine. Purines are present in two distinct skin layers: one immediately beneath the scales and one layer deeper in the skin closer to the muscle tissue. These layers containing both purines occur in parr but they thicken considerably in smolts and the ratio of guanine to hypoxanthine sharply increases over the course of smoltification. Parr-smolt transformation involves changes in purine nitrogen metabolism that results in the silvery fish that is better adapted for survival in marine habitats. The physiological and biochemical basis for these observations is not clearly understood although the silvering process of smolts appears to be related to increased secretion of thyroid hormones. The series of coordinated physiological, biochemical, and behavioral changes that facilitate the transition of juvenile salmonids from freshwater to marine life are interrelated in a complex fashion. A minimum size, age or growth phase appears to be required before this process can begin. Thorpe et al. (1998) described bimodality in growth of Atlantic salmon from Scottish rivers. In populations of hatchery-raised fish, individuals grow at similar rates until late summer or autumn when distinct growth bimodality becomes apparent. This bimodality does not appear to be related to sex or precocious male maturity although it does seem to be linked to genetic factors of Atlantic salmon (Thorpe et al., 1982, Gunnes and Gjeredrem, 1978, Thorpe & Morgan., 1980). Growth bimodality also appears to be the normal pattern in laboratory stocks of Atlantic salmon in Eastern Canada, where growth conditions are similar to those in Scotland 19 (Bailey et al., 1980; Kristinsson et al., 1985). It is very difficult to assess growth bimodality in wild populations due to overlapping year classes of fish and local environmental effects. Smolts are biochemically, physiologically and metabolically very different from the parr from which they originated. These changes that accompany smoltification also involve a decidedly altered pattern of endocrinology. The rate of oxygen consumption increases with changes in the metabolism of carbohydrates, fat and protein. The gill enzyme systems responsible for osmoregulatory adaptation adjust to enable the fish to tolerate marine salinity. Changes in plasma glucose, amino acid nitrogen and free fatty acids are observed as glycogen and lipid reserves are depleted and moisture content increases. Growth hormone, prolactin, thyroid hormones and Cortisol are all thought to be involved in lipid metabolic transformations accompanying smolting (Sheridan, 1986). The smoltification process is accompanied by an increase in the complexity of hemoglobins, with the appearance of two new anodic and four cathodic components added prior to seaward migration (Giles & Vanstone, 1976). The changes that occur in the complexity of hemoglobins appear to be linked to thyroid hormones. Hemoglobin is particularly important in transferring oxygen to tissues when juvenile fish may be exposed to variations in temperature or pH or may need sudden bursts of swimming activity. Adaptive changes in gas-transport proteins occur at the time of parr-smolt transformation in some species of salmon and these changes appear to take place well in advance of changing habitat (Bradley & Rourke, 1984). A very important component of parr-smolt transformation is related to gill enzyme systems and maintaining ionic equilibrium. In early life stages of salmon, life in a hypoosmotic environment encourages the removal of essential ions as water flushes these from the kidneys. In later life phases, when salmon are exposed to a hyperosmotic environment, drinking seawater and excreting salts must replace loss of water through osmosis. Gills, opercular epithelia, kidneys, urinary bladder and intestinal epithelia are the more important organs involved in this process. The gill cells play an essential role in absorbing salts from freshwater and excreting ions, especially sodium and chloride, when exposed to more saline conditions. Parr-smolt transformation is accompanied by increases in Na + , K +-ATPase, an important gill enzyme that is involved in changes in osmosregulation and salinity tolerance. Other important enzymes are linked to the processes of gas exchange, ion regulation and acid-base regulation as the fish 20 experience higher salinity conditions, although the increase in Na + , K +-ATPase activity is a key adaptation found in typical parr-smolt transformation. The changes in endocrinology that accompany parr-smolt transformation appear to be associated primarily with thyroid hormones, prolactin, growth hormone, corticosteroids and to a lesser extent, gonadal steroids. Thyroids hormones do not seem to control the development of hypoosmoregulatory capacity required for smoltification, but they have some influence on behaviour, growth and morphological development (Folmar & Dickhoff, 1980). There is a connection between triiodothyronine (T3) as demonstrated by studies in which dietary T3 administration significantly enhanced growth in juvenile coho salmon (Higgs et al., 1979). Prolactin is associated with sodium-retaining activity and thus influences salinity tolerance in smolts. Although growth hormone is not thought to be an important osmoregualtory hormone, it has been recognized for some time that there is an association between rapid growth and smolting (Hirano et al., 1990; Donaldson et al., 1979). Smoltification is accompanied by increased secretion (and metabolic clearance) of growth hormone. Growth hormone stimulates the production of insulin-like growth factor (IGF-1) in the liver, gills and kidney which, in turn, causes increases in gill Na + , K +-ATPase and stimulates conversion of T 4 to T 3 . The smolting process is also linked to elevated Cortisol levels possibly in concert with changes in the activity of Na + , K +-ATPase in the gills. The role of sex steroids in smoltification and the development of hypoosmoregulatory capacity is not well understood although it is known that changes occur in plasma estrogens, testosterones and other gonadal steroids in mature fish as they prepare to enter lower salinity environments of their spawning streams (Hirano et al., 1990). The complex nature of the hormonal interactions accompanying parr-smolt transformation presents a myriad of challenges for future studies. There are at least a dozen hormones that are thought to be involved either directly or indirectly in the process of smoltification, but many of the metabolic pathways and their sensitivities to environmental factors are yet unknown. Coho smolts grow very rapidly once they enter the marine environment. Survival depends largely on the abundance of prey or the salmon successfully avoiding predation. Smolting in salmonids is a reversible process and it is important to understand environmental conditions that delay or prevent desmolting. Survival of smolts to the adult stage in the ocean is variable but it usually from 5 to 10%. 21 Coho salmon begin to mature after one winter at sea and return to their rivers of origin during the late summer or autumn. Successive generations of each stock return to estuaries and spawning streams at approximately the same time each year (Royce et al., 1968). The duration of spawning migration for most stocks is about three months or more. Generally migration coincidences with particular stream flow conditions that are influenced by flow velocity and depth. Goundwater seepage has a significant influence on habitat conditions. Coho salmon prefer to migrate when the water temperature ranges from 7.2 - 15.6 °C, the minimum depth is 18 cm and water velocity is below 2.44 m/s (Sandercock, 1991). They actively migrate during daylight hours rather than at night and travel quickly through shallow riffle areas, seeking deeper, quieter pools to rest and avoid predation. In British Columbia, coho salmon are widely dispersed throughout the province, occurring in small numbers in most areas compared to other Pacific salmon species. The abundance of coho salmon in British Columbia has declined considerably since the mid-1960's and the trend of declining spawning escapement of coho salmon appears to be continuing in recent years (Henderson and Graham, 1998) Cumulative deleterious effects of pollution and alteration of spawning and rearing habitat can effect this already sensitive phase of survival for coho salmon. Given the importance of salmon fisheries and the aquaculture industry as a contributor to the global supply of protein for the human population, it is important to understand the natural life cycle of salmon and to recognize the implications of deleterious environmental effects. Research continues to probe the complex biological processes that are linked to salmon survival in the wild and optimal rearing of salmon under farmed conditions. The connections between effects of environmental contaminants on populations, individuals, individual physiological systems, individual organs and cellular response are a continual research challenge. Examination of effects of one contaminant of interest, 4-nonylphenol, on a critical life phase of salmon, smoltification, is a valuable exercise to gain information about these important connections. 2.8 Basic Nutrition of Coho Salmon In order to ensure successful growth and survival throughout the life cycle of all salmon species, energy derived from food intake must be equilibrated with the energetic costs of existence in freshwater and marine habitats. Insufficient reserves of protein and energy are thought to exacerbate incomplete physiological transformation of parr into smolts and thus indirectly contribute to mortality of anadromous salmon species. There has been much research conducted 22 on the nutritional requirements and feeding habits of Pacific salmon (Higgs et al., 1979, Cho & Kaushik, 1985, Healey et al., 1991 & MacDonald et al., 1987). This discussion will focus on the basic nutritional requirements of coho salmon with particular emphasis on composition of formulated diets as opposed to the diet composition of wild prey based diets. Growth of the aquaculture industry to augment the global supply of salmon has been accompanied by refinements in the design of formulated diets that promote the farming of nutritionally optimal fish in a cost effective fashion. Modern analytical instrumentation and knowledge make determination of nutrient composition, and thus protein and lipid quality, a fairly straightforward task. This evaluation of proximate constituents (protein, lipid, carbohydrate, chitin or fibre and inorganic components as ash) provides the means to estimate the nutritional status and the energetic requirements of fish. The gross energy content of salmon diets or prey can be determined by bomb calorimetry or by estimating calorific equivalents for proteins, lipids or carbohydrates. The chemical energy of food is measured directly in terms of heat units in analyses by bomb calorimetry whereas the values of calorific equivalents for protein, lipids and carbohydrates are based on previous experimentation where the heat of combustion is measured relative to the constituent weight. The currently accepted gross energy values are 23.6 kJ/g for protein, 39.5 kJ/g for lipid and 17.2 kJ/g for carbohydrates (Teskeredzic et al, 1995). Two key factors can influence the nutritive value of a salmon diet. The protein quality depends on the match between available quantities and balance of amino acids in food and the optimal amino acid needs of the fish. Lipid quality, defined as the match between essential fatty acid content of food and the fatty acid requirements of salmon, is also critically important. Therefore any external factor, such as environmental exposure to 4-NP, that could have a deleterious influence on protein or lipid quality can, in turn, affect the efficiency of conversion of diet constituents into useful survival and reproductive energy. Environmental factors that could be responsible for negative impacts on salmon growth, marine survival and pre-spawning mortality has direct economic and protein source effects on the human population. Proteins account for the largest proportion of organic compounds in the body of fish (between 53-78% of the whole-body dry matter in most salmonid species). They are important in fish species for the enzymatic activity and the cell structure. Proteins also play a role in meeting energy requirements of the fish, repairing tissue, synthesis of antibodies and the production of 23 hormones such as growth hormones, insulin and thyroid hormones. Proteins are composed of essential and non-essential amino acids joined by peptide bonds between the amino group of the amino acid and the neighboring carboxyl group of the next. Ten of the approximately 25 amino acids found in proteins of salmon diets are classified as essential amino acids (those that salmon cannot synthesize to sustain maximum growth), while the rest are non-essential amino acids. The fatty acid composition of salmon diets includes polyunsaturated fatty acids that can be divided into three families: the oleic (n-9) series, the linoleic (n-6) series and the linolenic (n-3) series where the individual members can be differentiated by the number of carbon atoms and double bonds in the chain and by the position of the first double bond, counting from the terminal methyl (CH3) group carbon to the carbon atom of the first double bond. These fatty acids in addition to some highly unsaturated fatty acids have particular nutritional importance in a balanced salmon diet that promotes optimal energy conversion and growth. The availability of n-3 series of highly unsaturated fatty acids in salmon diets is important for optimal fish health. Salmonids derive most of their energy for life from protein and lipids. The energy requirements of the fish must fulfill the needs for standard metabolism (minimum rate of energy expenditure to stay alive) and then the needs for routine metabolism (additional energy requirements for movement) before growth can occur. Standard metabolism is largely governed by water temperature and fish size. The energetic cost of routine metabolism must also balance energetic needs of feeding, activity, somatic growth, gamete production and excretion. Since a considerable part of available dietary energy is intimately linked to daily intake of proteins, it is essential that salmon diets satisfy the nutritional requirements of the fish for digestible protein in relation to digestible energy to enable optimal growth and energy conversion at the various life stages of the fish. Salmon growth and protein utilization can be adversely affected by a deficiency of dietary energy, excess amounts of protein energy relative to dietary energy or excessive amounts of non-protein (mainly lipid) energy in relation to protein energy. Any biotic or abiotic stress that could affect protein utilization can seriously disrupt the normal course of growth and development for salmonids. Much of the available (digestible) dietary energy requirements of salmon are supplied by their daily intake of protein. Dietary protein intake must be optimal for maximum fish growth and this is highly dependent on fish species and stage of the life history of the fish. The transfer of 24 salmon to sea water involves different energy requirements for the process of osmoregulation and thus, changes in dietary protein content are linked to optimal fish growth after smoltification. Evidence to date suggests that best likelihood of salmon survival require the fish to consume diets that are high in digestible protein and lipid content throughout their transfer to sea water (Higgs etal., 1995). Lipids are an important source of digestible energy per unit weight in salmonid diets. These compounds contain essential fatty acids that are critical to promote growth, health, survival and reproduction. In addition to playing a role in proper membrane structure and function, lipids transport fat-soluble vitamins A , E, D and K and carotenoid pigments. Phospholipids are especially important in membrane structure, fluidity, flexibility and permeability in salmon. As such these compounds can influence ionic transport across membranes, the rates of membrane-based enzymatic reactions and mitochondrial oxidation in the fish (Higgs et al., 1995). The study of nutrient and energy requirements of salmon species, the bioenergetics and digestibility of diets in relation to exposure of fish to contaminants is particularly valuable to understanding best practices of fisheries risk management. 2.9 Risk Assessment of Endocrine Disrupting Compounds Ecological risk assessments are conducted to determine the responses of organisms to various levels of contaminants such as endocrine disrupting compounds and the risk they may pose to human and ecosystem health. The general framework for conducting such an assessment involves the examination of the defined problem as based on the most current scientific information. Once the nature of the problem has been determined, analyses of sources, observed exposure levels and effects of contaminants on the organism are rigorously studied to provide the best possible scientific characterization of the risk given the best available technologies to conduct these experiments. If the results of these studies characterize the contaminant as posing a threat to the ecosystem or human health, steps must be taken to manage the associated risk. Much of the responsibility for enforcing risk management practices rests however, with government institutions while the private sector and civil society can also share the role of risk management when effective risk communication enables the environmental problem to be understood and addressed by voluntary choices at the corporate and individual level. 25 Ecological Risk Assessment/ Management Process 1. Principles, Objectives & Policies 2. Problem Formulation 3. Communication with Risk Assessors 4. Risk Management Decisions Figure 2.1: The General Framework for Ecological Risk Assessment The formulation of the problem and the characterization of the risk for several contaminants demonstrating endocrine disruption capabilities in various organisms has resulted in a substantial body of scientific evidence that supports the need to manage the risks associated with environmental exposure to these contaminants. Many examples of the observed effects of endocrine disruption in humans and other organisms have been cited above. Many of these compounds have established exposure limits that have been defined in terms of their acute toxicity and sub-lethal toxicity values. The sources of endocrine disrupting compounds in the environment include: • Pesticide application - agricultural run-off • Industrial effluents: pulp and paper, textile and steel production and plastic manufacture • Municipal wastewater - sewage, and • Plasticizers, surfactants, cleaning products, adhesives, personal care products, lubricants, resins, chemical stabilizers, antioxidants. Despite the substantial weight of evidence that supports the concerns regarding the adverse consequences of exposing animals and humans to endocrine disrupting compounds, formulating a strategy to manage this risk is problematic since there are many knowledge gaps and uncertainties in the risk characterization. Very little is understood about the synergistic and antagonistic effects of combinations of endocrine disrupting compounds. Moreover, there are many gaps in our understanding of the normal functions of the endocrine systems of organisms 26 and our knowledge of the environmental fate and effects of contaminants is incomplete. Also, for most contaminants, there are gaps in our knowledge with respect to these effects at the cellular level within an organism or even their effects on the organism as a whole or on the population of organisms and how these may affect the interaction of organisms within an ecosystem. A l l of these factors make risk management nearly impossible without some sort of precautionary judgment. As a specific example of ecological risk assessment, nonylphenols are contaminants for which there is a considerable body of scientific evidence that has been developed to characterize its risk as an endocrine disrupting compound. Nonylphenols are C9 alkyl phenols that are composed of both straight and branched chain isomers. Over 90% of the industrial formulations of nonylphenol are para isomers with the other 10% being ortho isomers. Small amounts of 2,4 dinonylphenol can also be present in nonylphenol mixtures. There are 211 pesticides registered in Canada which contain nonylphenol although 95% of these products contain less than 20% NP or NPE. In 1996, 17,200 tonnes of NP plus NPEs were available for domestic use in Canada although the relative proportion of NP to NPE was not stated (CEPA, 2000). The industrial use of NP in Canada in 1996 was 5000 tonnes. The risk assessment evidence usually includes nonylphenol and its ethoxylates since nonylphenol ethoxylates are often the primary substances that are disposed of in wastewater whose degradation products include the more environmentally persistent 4-nonylphenol. Figure 2.2: Structures and abbreviations of nonylphenol polyethoxylates. NP nonylphenol n=0 N P 1 E O nonylphcnof^m n=l NT^EO nonylphenol n=0 N P 1 E C nonylphenoxy acetic acid n=l JST^ES norrylplwn nonylphenol and estrogenic metabolites of 27 Acute adverse effects of nonylphenols have been reported in several species of invertebrates, fish, mammals and algae. Some acute toxicity and sublethal endpoints (not including estrogenicity) are provided in Table 2.1. Since most nonylphenol and ethoxylates are disposed of in wastewater, the most important media of concern are water and sediments. The relatively hydrophobic nature of 4-nonylphenol (KQ W ~ 4.5) promotes its sorption to particulate matter or sludges and this compound demonstrates a moderate potential to bioaccumulate in aquatic organisms. Soil is also of concern as a terrestrial medium of exposure since sludge is often disposed on land and thus growth of vegetation is an important endpoint for assessing toxicity of these substances. For the most part, the distribution of nonylphenol and the degradation products of nonylphenol ethoxylates are localized to areas near to their point of discharge. Table 2.1: Toxicity Endpoints for Nonylphenol to Fish, Invertebrates and Algae. (UK assessment report, 1997). Species Endpoint Value ( u g / L ) Reference Freshwater Fathead minnow (Pimephales promelas) 96 hr L C 5 0 135 Holcombe et al., 1984 Fathead minnow (Pimephales promelas) 33 day LOEC survival 14 Ward&Boeri , 1991a Daphnia magna 21 day NOEC survival of offspring 24 Comber etal., 1995 Ceriodaphnia dubia 96 hr LC 5o 69 England, 1995 Ceriodaphnia dubia 7 day NOEC reproduction 88.7 England, 1995 Algae (Selenastrum capricornutum) 72 hr EC 10 cell growth 410 Ward&Boeri , 1990a Algae (Scenedesmus subspicatus) 500 Huls, 1996 Lemna minor 4 day LOEC growth 0.125 Prasad, 1989 Marine Sheepshead minnow (Cyprindon variegatus) 96 hr L C 5 0 310 Ward&Boeri , 1990b Mysid shrimp (Mysidopsis bahia) 96 hr L C 5 0 43 Ward&Boeri , 1990c Mysid shrimp (Mysidopsis bahia) 28 day LOEC 6.7 Ward&Boeri , 1991b Algae (Skeletonema costatum) 96 hr E C 5 0 27 Ward&Boeri , 1990d 28 Nonylphenols appear to be more toxic to aquatic organisms than nonylphenol polyethoxylates (NPEOs) or their other degradation products. In municipal wastewater plants, the treatment process can rapidly degrade long chain nonylphenol polyethoxylates into shorter chain nonylphenol polyethoxylates and nonylphenol polyethoxycarboxylates (NPECs) and ultimately to nonylphenol (Ahel et al., 2000). Although these transformation products are formed in the treatment process, the nature of the inputs and the method and degree of treatment strongly influence the final effluent concentrations. NPECs are considerably more water-soluble than NPEOs and are therefore found in the aqueous phase of the final effluents where nonylphenol is generally associated with suspended organic particles. Octylphenol is among the degradation products of the municipal wastewater treatment process that is of concern since its toxicity is relatively similar to nonylphenol. Indeed, some studies have indicated that it may have greater relative potency to than that of 17-0 estradiol and estrone (Blackburn & Waldock, 1995). 2.10 Environmental Relevance of 4-Nonylphenol in British Columbia Watersheds The experiment selected in this case study investigated the response of juvenile coho salmon to a contaminant of concern that is thought to be present in appreciable concentrations in many aquatic and marine water systems. In Canada, and specifically in British Columbia, protection of salmon habitat requires extensive evaluation of environmental effects of pollution or habitat degradation since salmon play a vital in the environmental, economic and cultural health of watersheds. One contaminant of specific current concern is nonylphenol and its ethoxylates. This investigation is designed to provide a small contribution of sound scientific evidence that relates to the effects of dietary exposure of a juvenile salmon species to 4-nonylohenol. The Canadian Environmental Protection Act, 1999 proposed that nonylphenol and its ethoxylates be declared toxic and as such, appropriate control measures be taken to prevent harm to human and ecosystem health (CEPA, 2000). The management strategy as defined by this act involves identifying those substances that may be toxic, assessing their toxicity as defined in Section 64, and, for those substances found to be toxic, establishing and applying control measures to exposure pathways. The Toxic Substances Management Policy (TSMP) produced by Environment Canada in 1995 also guides the management of toxic substances. This policy adopts a preventative and precautionary approach to deal with potentially toxic contaminants that enter the environment and pose a risk to human and/or ecosystem health. The policy provides the framework for 29 making decisions about effective management of toxic substances based on sound scientific evidence. Using this framework, the key management objectives are to 1) virtually eliminate from the environment any toxic substances that result predominantly from human activity and that are persistent and bioaccumulative (referred to in the policy as Track 1 substances) and 2) manage other toxic substances or substances of concern, throughout their life cycles, to prevent or minimize their release into the environment (referred to as Track 2 substances). A substance categorized as either a Track 1 or Track 2 substance will thus require different strategies of management. The Canadian Environmental Protection Act, 1999 requires that the Ministers of the Environment and Health compile and publish a list of substances that it considers a priority for assessment to determine i f they should be defined as being "toxic" under the Act. Among the contaminants on this list which is known as the Priority Substances List (PSL), is nonylphenol and its ethoxylates. Nonylphenol and its ethoxylates, are discharged in large amounts in Canada from pulp and paper mills, textile mills and municipal wastewaters. These compounds are on the Second Priority Substances List. Environment Canada and Health Canada proposed that these compounds be declared toxic under CEPA through the publication of the Toxic Assessment Report in the Canada Gazette Part 1, April 1 2000. On June 23, 2001. nonylphenol and its ethoxylates were declared "toxic" in Canada under CEPA on the basis of assessments conducted as part of the PSL 2 program (Servos et al., 2001). This designation requires that this group of compounds be subjected to the risk management process. As a result of the risk characterization and comparisons with environmental concentrations of these compounds that have been implicated as having impacts on ecosystem health, the Minister of the Environment requires the establishment of a regulation or other instrument that respects prevention and control actions in relation to nonylphenol and its ethoxylates. Following this declaration, 18 months are provided to finalize the regulations and/or instrument considering Best Available Technologies Economically Achievable (BATEA) and best management strategies developed through consultation with all interested parties. 4-Nonylphenol (4-NP) is a precursor in the production of nonylphenol ethoxylates (NPE), which are nonionic surfactants found most often in cleaning products and industrial processing. Nonylphenol ethoxylates (NPEs) are high volume chemical members of a broader group of 30 compounds known as alkylphenol ethoxylates. These NPEs have been extensively used as detergent, wetting agents, emulsifiers and dispersing agents for more than forty years. Release of effluents containing NPE contaminants is of particular concern since breakdown products include 4-nonylphenol, octylphenol, and a wide variety of carboxylates and ethoxylates which may also have significant endocrine disrupting potential to organisms. The presence of 4-NP and NPEs in the environment is exclusively the result of anthropogenic activities. Primary end uses of NPE compounds include cleaning products, degreasers, paints, resins and protective coatings, plastic and elastomer manufacturing, textile processing, pesticide emulsifiers, pulp and paper production, and personal care products (Talmage, 1994). These products are found in many other sectors such as oil and gas recovery, steel manufacturing and power generation since their varied applications include control of deposits on machinery, equipment cleaning ,and product finishing. Key environmental sources of these contaminants are linked to discharge of municipal wastewater, effluents from pulp and paper mills, effluents from textile mills and agricultural runoff to the receiving environment. 4-Nonylphenol is one of the primary degradation products of nonylphenol ethoxylates and its chemical stability in the environment exceeds that of its parent compounds. 2.11 Estrogenic Effects of 4-Nonylphenol Previous research has demonstrated that exposure of Atlantic salmon (Salmo salar) to exogenous estrogenic and/or androgenic substances during the period of parr-smolt transformation may compromise subsequent seawater adaptability and survival (Fairchild et al., 1999). The latter study revealed a significant relationship between the timing of historical insecticide applications containing 4-nonylphenol as an adjuvant in Atlantic Canada and impairment of smoltification in Atlantic salmon. A Danish study that involved 30-day exposure of Atlantic salmon to either 17 P-estradiol or 4-nonylphenol by injection suggested that the presence of estrogenic substances in the environment may have a deleterious effect on smoltification and migration of wild stocks of salmon (Madsen et al. 1997). Further, the study demonstrated that treatment of salmon with 17 P-estradiol and 4 nonylphenol activated the vitellogenin system and inhibited the process of smoltification. Sea water challenge tests were used to assess osmoregulatory ability during smoltification and the observation of impaired salt water tolerance correlated with significantly reduced gill Na + , K"-ATPase of treated fish when they were compared to control fish. The results of this experiment suggested that 17 p-estradiol and 4-nonylphenol may have a qualitatively similar inhibitory effect on smoltification and hypoosmoregulatory physiology of 31 Atlantic salmon. Brown et al. (2001) hypothesized that alkylphenols such as 4-nonylphenol and nonylphenol-1-carboxylate (NP1EC) may have deleterious effects on parr-smolt transformation and subsequent sea water growth of Atlantic salmon (Salmo salar). They investigated this hypothesis through laboratory exposure of Atlantic salmon to environmentally relevant concentrations of 4-NP, NP1EC and 17 p-estradiol (E2) during the latter stages of parr-smolt transformation. During the experiment they evaluated the ability of the smolts to withstand sea water and their subsequent growth and they also assessed the status of some osmoregulatory, hormonal, and histological parameters. They found no treatment related mortalities following seawater challenge tests immediately after exposure although they observed subsequent impairment of Atlantic salmon growth in sea water by 30-40% in fish in the various treatment groups relative to controls. Some treatment groups also displayed changes in indices of growth, thyroidal status, plasma steroids, vitellogenin and histology but not osmoregulatory parameters. Liber et al. (1999) conducted a comprehensive series of studies using littoral enclosures to assess the persistence and distribution of 4-NP in a littoral ecosystem. They evaluated the effects of the compounds on resident aquatic biota and found that zooplankton was the most sensitive group of organisms that were studied with significant reduction in copepod taxa. 4-nonylphenol was observed to persist a long time in sediments which were identified to be a primary sink for this compound. Hale et al. (2000) conducted a study of nonylphenols in sediment and effluents associated with a number of diverse wastewater effluents was undertaken by in the Virginia river ecosystems. Their results collected from 7 different sites in the mid-Atlantic US region indicated that nonylphenols may be released from diverse sources, concentrate in the sediment and persist for extended periods. One of the highest concentrations of 4-NP measured was 14,100 pg/kg and this level was detected in the stormwater discharge near a federal facility. Although many studies focus on evaluating 4-NP near sewage treatment plants, significant concentrations of these compounds in the sediment have been found near many diverse outfalls. It is now known that different isomers of 4-nonylphenol elicit differing degrees of estrogenicity (Servos, 2000). 4-NP is a mixture of various straight and branched chain alkylphenols used commonly in the production of nonionic surfactants. In addition, nonylphenols are now widely distributed in marine and freshwater receiving environments. 32 Lussier et al. (2000) determined acute toxicity values for early lifestage exposure of four species of marine invertebrates to 4-NP. A narrow range of toxicity values was observed for the species and this implies that exceeding a threshold concentration could endanger a larger portion of the marine or aquatic community. The well know team from Michigan State University led by Dr. John Geisy have conducted many studies on the effects of 4-NP on fish. One of their recent published studies (Geisy et al., 2000) examined the water borne exposure of adult fathead minnows (Pimephales promelas) to 4-NP both at the beginning and the end of the breeding season. The experiments did not demonstrate a dose-dependant effect of 4-NP on production of vitellogenin in male fish, but significant effects of 4-NP on plasma vitellogenin were detected in female fish. Plasma estrogen levels were significantly elevated in both sexes. It appears that the effects of 4-NP on fathead minnows does not result from a direct acting estrogen antagonist mechanism but rather from changes in plasma estrogen concentrations by an indirect mode of action. The few examples of effects of 4-NP described above illustrate the diversity and complexity of establishing an environmental cause and effect relationship. Complicating these issues is the fact that the very physical nature and purpose for which the chemical was produced, i.e. its lipophilicity and viscosity, make design of experiments troublesome. 33 METHODOLOGY "There is no science without fancy and no art without facts. " Vladimir Nabokov 3.1 Experimental Design 3.1.1 Animals Juvenile coho salmon (Oncorhynchus kisutch) were obtained from Capilano Hatchery in January 2000 as swim-up fry and they were reared in 1100 1 tanks in the indoor aquarium of the Department of Fisheries and Oceans West Vancouver laboratory. The animal care committee of the laboratory approved the rearing facilities and experimental protocols for the fish. 3.1.2 Test Conditions The experiment was divided into four 28-day periods in order to monitor the growth and performance of the fish. Figure 3.1 summarizes the task versus time matrix for the experimental phase of the research. Sample chemical analyses of 4-NP was conducted by Dr. Natasha Hoover between June and December, 2000. A l l hormone assays were conducted at the CCIW laboratory in Burlington, Ontario. Proximate analyses of five fish per treatment tank in duplicate were conducted between January 24-September 20, 2001 at the DFO West Vancouver Laboratory. 34 May June 2000 July Aug. Sept Acclimation o f f i sh Fed treatment diets Samples for weight & length 1 r Seawater challenge test L r Samples for proximate Analyses L J J L L r T Fish sexed, livers weighed, Plasma and tissue sampled for 4NP determination J L Pi Sea water acclimation Growth in sea water June 26 Figure 3.1: Task versus time matrix of experimental phase of research. On May 10, 2000, eighty (80) juvenile coho salmon between 13-18 g were randomly distributed into each of 12 oval fiberglass 1100 1 tanks in an indoor aquarium facility. Flow through conditions of approximately 15 1/min and 6 °C Cypress creek water were provided for each tank. Photoperiod within the indoor aquarium approximated the normal seasonal diurnal cycle and was later adjusted to match seasonal change over the course of the experiment (May 10 - September 12, 2000, Vancouver, Canada, 49°15'N. 123°10' W). Oxygen saturation in the water was maintained above 85% by supplemental aeration in each of the tanks. The experimental design was a two-way randomized block, with six tanks per block. The treatment diets were randomly assigned to tanks within each block. Experimental fish were first sedated with clove bud oil and then anesthetized with a 1:1 solution of MS222 (tricaine methanesulfonate) to allow weighing and measuring, and vaccination against vibrio and furunculosis prior to their distribution to their assigned tanks. The fish were fed a basal diet as they acclimated to the experimental conditions for 14 days prior to receiving their prescribed test diet. 35 3.1.3 Diet Composition and Administration of 4-NP The treatment diet was prepared February 16, 2000 using the coho basal diet formulation provided in Table 3.1. The steam pellets were prepared using a California model CL-Type 2 pellet mill that was equipped with a 2-38 mm die. Samples of each of the test diets were removed for chemical determinations of their respective contaminant concentrations. Table 3.1: Ingredient composition of the basal diet that was supplemented with either 4-nonylphenol (4-NP; 0, 0.002, 0.20, 20 or 2000 mg/kg air dry basis) or 17-(3 estradiol (E2; 31 mg/kg air dry basis).17 Ingredient g/kg dry weight LT-Anchovy meal 524.0 Blood flour; spray-dried 50.0 Squid meal 70.0 Wheat gluten meal 50.0 Pre-gelatinized wheat starch 85.8 Raw Wheat Starch 30.0 Vitamin supplement2/ 20.0 Mineral supplement37 20.0 Anchovy oil; stabilized47 109.7 Soybean lecithin 10.0 Choline chloride (60%) 5.0 Vitamin C, monophosphate (42%) 3.6 Permapell 10.0 Finnstim™ 10.0 DL-Methionine 2.0 L 1 / The different amounts of 4-NP, or E2 were dissolved in ethanol and then included in the oil that was added to the mash before the diets were steam pelleted. The basal (control) diet received the same quantity of ethanol alone. All diets were coated with 27g/kg (air-dry basis) of spray-dried hydrolyzed krill in a marine oil carrier (20 g/kg diet on an air-dry basis). 2 / The vitamin supplement provided the following amounts/kg of diet on a dry weight basis: vitamin A acetate, 5000 IU; cholecalciferol (D3), 2400 IU; DL-a- tocopheryl acetate (E), 300 IU; menadione, 18 mg; D-calcium pantothenate, 168 mg; pyridoxine HCI, 49 mg; riboflavin, 60 mg; niacin, 300 mg; folic acid, 15 mg; thiamine mononitrate, 56 mg; biotin, 1.5 mg; cyanocobalamin (B12), 0.09 mg; inositol, 400 mg, BHT, 22 mg. 3 / The mineral supplement provided the following (mg/kg diet on a dry weight basis): manganese (as MnS04 • H2O), 75.0; zinc (as ZnS04 • 7H20), 90; cobalt (as C0CI2 • 6H2O), 3; copper (as CuS04 • 5H20), 7; iron(as FeS04 • 7H20), 100; iodine (as KIO3 and KI, 1:1), 10; fluorine (as NaF), 5; selenium (as Na2Se03), 0.2; sodium (as NaCl), 1419; magnesium (as MgSCH • 7H20), 400; potassium (as K2SO4 and K2CO3, 1:1), 1500. 4 / Stabilized with 0.5 g santoquin/kg oil. 36 Fish were fed their prescribed diet twice daily by hand to satiation. Records of water temperature, salinity, dissolved oxygen and fish mortality was maintained on a daily basis. Feed was withheld for 24 hr prior to each sampling. The randomized distribution of the treatments within the two blocks of six tanks was: Tank Treatment Tank Treatment 223 2000 mg 4-NP/kg 229 2000 mg 4NP/kg 224 E2 230 20 mg 4NP/kg 225 0.2 mg 4NP/kg 231 0.2 mg 4NP/kg 226 20 mg 4NP/kg 232 0.002 mg 4NP/kg 227 0.002 mg 4NP/kg 233 E2 228 Control 234 Control The experiment was divided into two exposure periods. The 28-day fresh water phase while the fish consumed the treatment diets was the first period. After June 21, 2000, a gradual 5-day changeover to seawater conditions began the second period of the experiment. Fish were held in flow through seawater with salinity between 25-32 °/ 0 0 and fed untreated basal diet for the remaining experimental time. During the freshwater phase of the experiment, the temperature of the Cypress Creek water ranged between 4.9 - 8.2 0 C from May 12 to June 19, 2000. The dissolved oxygen concentration of the same water at this time ranged between 11.1 - 12.8 mg/1. After the June 21, 2000 sampling day, water supply was gradually adjusted to 100 % filtered seawater over a five-day period. Throughout the seawater exposure period of June 26 - September 12, 2000, the temperature of the water ranged between 8.7 - 12.2 0 C, the dissolved oxygen ranged between 7.8 - 12.2 mg/1 and the salinity was between 25 - 32 °/ 0 0 . During the first fresh water phase of the experiment, the Cypress Creek water was often turbid and thus, estimation of wasted feed was difficult and sometimes not possible. The fish were fed to apparent satiation as determined by observing their feeding behaviour. Rainfall events influenced the turbidity of the water conditions often making it difficult to observe the fish 37 within their tanks at all. During these times, fish were fed on the basis of the time to consume their food when less turbid water allowed behavioural observation on other days. The 4-NP diets that were supplemented with 2000 mg/kg and 20 mg/kg were prepared by weighing the appropriate amount of 4-NP (reagent grade, Aldrich, primarily branched chain) or 17-p estradiol (Sigma Chemicals) directly into 20 g of marine (anchovy) oil and then dispersing the emulsion for 3 minutes by sonication. Diets supplemented with the lower concentrations (0.2 mg/kg and 0.002 mg/kg) of 4-NP were prepared by adding an appropriate volume of a stock solution of 10.0 ml of 95% ethanol (reagent grade, Anachemia Science). Since the solubility of 4-NP is accepted as being 4.9 ± 0.04 mg/1 in water at 25 °C (Brix et al., 2001), the weights of 4-NP used for the lower doses were likely be completely dissolved in the 10.0 of ethanol used to prepare the stock solution. The stock solution was appropriately diluted and delivered into 20 g of marine oil before sonication. This oil solution was then thoroughly mixed for a total of 20 minutes in the diet mash using a Hobart mixer prior to the pelleting process. A l l diets were coated externally with 27 g of marine euphausids (ground krill, Marine Specialty Products) dispersed in 123.1 g of marine (anchovy) oil to improve diet palatability. Fish were fed to satiation two times per day every day between May 10, 2000 and September 12, 2000. Each day, the diet containers were weighed at the beginning and the end of the day to enable calculation of the daily food intake of per group. The pellets were dispensed to the fish in each tank using a plastic teaspoon. Initially at each feeding time, the lids of all twelve tanks were opened and the fish were fed for 1-1.5 hours or until they refused to consume any more food with minimum pellet wastage. Later during the experiment, the tank lids could not be left open when feeding successive tanks since the enthusiastic feeding behaviour of the fish on a couple of occasions resulted in fish jumping out of their tanks. 3.1.4 Treatment On the initial day of the experiment, approximately 20 individual coho were sampled randomly. Subsequently they were sacrificed and frozen for later chemical and proximate analyses. The fish were fed to satiation their prescribed contaminant-treated diet for twenty-eight (28) days. Each group was fed twice daily by hand using a plastic spoon to distribute about 10 pellets each time until the fish showed signs of losing their enthusiasm for eating. Once uneaten pellets began to accumulate at the bottom of the tank, the feeding session was discontinued. 38 At the conclusion of 28-day feeding period in which the fish were administered 4-NP or E2, the fish were weighed, measured and returned to their respective tanks on June 21, 2000. On day 28, 12 fish were sacrificed for analyses of vitellogenin, T 3 and T 4 in blood plasma and 4-NP in fish tissue, gall bladder, liver and blood plasma. A l l fish were weighed and measured in each treatment tank. Health observations were recorded and the sex of all.dissected individuals was also recorded. Livers were weighed and pooled for 4-NP analyses. Blood was collected from fish using a 1000 pi heparinized vacutainer™ syringe. A l l syringes and equipment used to sample plasma for Vg were kept refrigerated before use. Sample vials were pre-labeled and refrigerated prior to plasma collection. Samples were centrifuged at 1800 rpm at 4 °C and immediately frozen on dry ice upon collection and stored at -40 °C. Following the weighing and sampling of the fish on day 28, the water flowing into each tank was changed from Cypress Creek water to ambient temperature, oxygenated (80% saturation), filtered sea water. Also, all groups were fed the control diet for the remainder of the study. After six weeks of growth in seawater on August 6, 2000, all fish were first sedated with clove bud oil and then anesthetized with a solution of MS222 (tricaine methanesulfonate), weighed, measured and returned to flow through seawater in their respective 1100 L tanks. Six weeks later, the fish were once again weighed and measured. 3.2 Analyses 3.2.1 Fish Weighing and Sampling A l l fish in each group were individually weighed and measured at 28-day intervals throughout the 112 day study after they were subjected to the duel anesthetic procedure described earlier. The experiment continued until September 12, 2000 (day 112). At this time, all fish in each group were first sedated with clove bud oil and then anesthetized as described above and then they were individually weighed and measured. Thereafter, five fish from each group were randomly selected and sacrificed for proximate analyses. Also, 12 fish from each group were randomly selected, sacrificed and dissected for determination of plasma hormone titres and 4-nonylphenol concentrations in the liver, gall bladder, blood plasma and fish muscle tissue. Fish muscle tissue was sampled from the region immediately below the dorsal fin of the fish. 39 3.2.2 Hematocrit Measurement The erythrocytes of fish blood, like mammalian counterparts, contain hemoglobin and are essential for the transport of oxygen to all tissues and organs of the body (Hesser, 1960). Hematocrit measurements give an indication of he general health of the fish at the time of sacrifice. The term hematocrit is used to define the instrument that measuring the amount of plasma and corpuscles in the blood. It is essentially a measure of the volume occupied by packed erythrocytes in a given volume of blood and is usually measured as the percentage of erythrocytes in blood. The limited volume of blood available in small fish is overcome by using the micro-hematocrit method in which small (1.1-1.2 mm inside diameter and 75 mm length, 40 pi S/P Brand Baxter Healthcare Corp) heparinized capillary tubes are filled with one or two drops of blood, sealed with Hemato-seal™ (Fisher Scientific) and centrifuged for three minutes at 12,000 rpm. Microhematocrit capillary tubes were then measured to the nearest mm with a hematocrit-measuring ruler. Hematocrit of eight randomly selected fish for seawater challenge test was measured by Dr. John Blackburn on June 22, 2000 at the DFO West Vancouver Laboratory. The increase in osmality following seawater challenge may cause dehydration of blood cells and thus reduced hematocrit. Hematocrit measurement gave an indication of the degree of dehydration of the blood cells at the time of analyses. Cells did not demonstrate apparent lysis. 3.2.3 Seawater Challenge Test The seawater challenge test is a physiological test that evaluates the osmoregulatory competence and seawater adaptability of juvenile salmon as they undergo the process of smoltification (Blackburn & Clark, 1987). The test involves measurement of plasma sodium levels by flame photometry and comparison of group mean plasma sodium levels to the control group. The methodology involves immediate transfer of 12 fish from each treatment directly into 29-30°/ o o seawater. After 24 hr in sea water, the fish are immobilized in a MS222 solution made with seawater. Then the tail of each fish is cut off and blood from the caudal artery is collected in heparinized microhematocrit tubes. These tubes are kept refrigerated until plasma sodium measurement. 40 Plasma sodium is measured with a Turner (Case Instruments) clinical flame photometer using a commercial reference human serum sodium standard. Aliquots of 5 pi of fish plasma are diluted to 1.0 ml. Three replicates of each standard are run to construct the calibration curve and sample plasma sera are compared to the resulting standard curve. 3.2.4 T 3 and T 4 Hormone Measurement Since the smoltification process is linked to thyroid hormone levels in salmonids, analysis of T 3 and T 4 could give an indication of the endocrine status of the treatment fish. Plasma samples from twelve fish per treatment tank were collected on June 21 & September 12, 2000. Blood was withdrawn from the caudal vessels of each fish using chilled heparinized syringes (21 gauge disposable needles). A l l syringes, collection tubes and cyro-vials were kept refrigerated until use. Plasma was collected after cold centrifugation of the blood samples at 1700 rpm for 10 minutes. Samples were immediately frozen and stored at -20 °C prior to overnight shipment, on dry ice, to the Canada Centre for Inland Waters in Burlington, Ontario. Radioimmunoassay procedures are the most sensitive analytical method for determination of L-thyroxine and 3,5,3'-triiodo-L-thyronine. Dr. Scott Brown determined the plasma concentrations of T 3 and T 4 hormones using a radioimmunoassay method described by Brown and Eales (1976). The small size of the June 21 fish resulted in only small volumes of the plasma for the hormone assay and in several cases, where there was insufficient plasma to reliably conduct the assay, pooled samples were assessed within each treatment. Although the larger September 12 fish provided more plasma per fish, some of these samples were too small and hence pools of plasma for analysed for each treatment. In relation to the radioimmunoassay procedure, barbital buffer was first prepared by dissolving 15.6 g of sodium barbital in 900 ml of deionized H2O. Then the pH was adjusted to 8.6 with 6N HCI (about 2-2.5 ml) and the solution was diluted to one litre. [ 1 2 5]T 4(T 4*) and [125] T 3(T 3*) that were phenolically labeled and had initial activities of about 725 and 500 mCi/mg respectively were obtained from Industrial Nuclear, St. Louis M O in 50% aqueous propylene glycol. The T 4* or T 3 * solutions were diluted with 0.1N NaOH such that 0.1 ml aliquots of the stock solution yielded activity of 3500-4500 counts per minute (cmp) in a gamma well detector at about 50%. Standard stock solutions of T 4 and T 3 (10% pg % as free acid) were prepared by dissolving T 4 tablets (Eltroxin, sodium L-thyroxine pentahydrate) or the anhydrous sodium salt of T3 (Sigma 41 Chemicals) in 0.1N NaOH. Working standards of 0, 50, 100, 200, 400 or 800 ng/lOOml were prepared by further dilution of the stock solution with 0.1N NaOH. Lyophilized rabbit antisera to T 4 human serum albumin ( T 4 antibody) or to T 3 human serum albumin (T 3 antibody) were obtained from K & T Biological Services Ltd, Edmonton, Alberta. The raw T 4 serum was diluted 1:7000 and the raw T 3 serum was diluted 1:22,000 with barbital buffer. Standard sized molecular sieve Sephadex columns (Tetralute) were prepared, conditioned with human plasma (1:20 in barbital buffer), rinsed with 0.1N NaOH and stored capped at room temperature containing l-2ml of 0.1N NaOH for use in the separation of sample plasma. Assay Procedure Reagents, samples and columns were allowed to come to room temperature. Columns were drained and the cap was placed on the bottom of the column. Aliquots of 0.1 ml of T 3 or T 4 standard or 0.1 ml of sample plasma were pipetted onto the columns. Duplicate standards of each concentration were run. Aliquots of the T 3 * or T 4 * solutions were added to the columns. Total reference count standards were prepared by pipetting 0.1 ml of T 3 * or T 4 * in duplicate directly into counting tubes and adding 3.9 ml (for T 4 RIA) or 2.9 ml (for T 3 RIA) of barbital buffer. The tracer and standards or sample plasma were drained through the columns and the eluates discarded. The process was repeated with 4.0 ml (for T 4 RIA) or 3.0 ml (for T 3 RIA) and this solution discarded as well. After the buffer solutions had drained through the columns, 1.0 ml of the T 3 or T 4 antibody reagent was added to the columns and allowed to equilibrate for 90 minutes with collection counting tubes placed beneath each column. At the conclusion of the equilibration phase, 3.0 ml (for T 4 RIA) or 2.0 ml (for T 3 RIA) of barbital buffer were run through the columns and eluent was collected in counting tubes (4.0 ml total volume for T 4 RIA or 3.0 ml for T 3 RIA). Samples and standards were counted for 10,000 cpm or 10 mins in a Nuclear Chicago Automatic Gamma System using a 2 in (DS 202) Nal crystal. The resulting standard curve was plotted with T 3 or T 4 antibody-bound cpm on the y-axis against hormone concentration in ng/100ml on the x-axis. Detection limits of the method are 12.5 ng/100 ml for T 4 and 9.5 ng/100 ml for T 3 . Standard procedures for QA/QC were conducted throughout the analyses. 42 3.2.5 Vitellogenin Assay Measurement of vitellogenin in the blood plasma of fish has been demonstrated to be a useful biomarker for determining estrogenic activity of contaminants (Sumpter and Jobling, 1995). Vitellogenin (Vg) is a phospholipoprotein that plays an important role in the reproduction of oviparous vertebrates. Thus evaluation of plasma Vg in both female and male organisms can indicate exposure to estrogenic substances. Vitellogenin is not usually induced in relatively high concentrations in male oviparous organisms so comparatively high Vg concentrations found in males suggest atypical exposure to estrogenic substances. The enzyme linked immunosorbent assay (ELISA) method for determining V g in plasma is now accepted for the detection of exposure of fish to environmental estrogens and estrogen mimics. Plasma collected from twelve fish per treatment on June 21 & September 12, 2000 was separated into 1.0 ml cryo-vials for Vg analyses. Blood was withdrawn from the fish using chilled heparinized syringes (21 gauge disposable needles). A l l syringes, collection tubes and cyro-vials were kept refrigerated until use. Plasma was collected after cold centrifugation of blood at 1700 rpm for 10 minutes. Samples were immediately frozen and stored at -20 °C prior to overnight shipment, on dry ice, to the Canada Centre for Inland Waters in Burlington, Ontario. Analyses of V g were to be conducted by Jim Sherry at the Canada Centre of Inland Waters laboratory in Burlington Ontario according to the methodology described by Sherry et al. (1999). The enzyme linked immunosorbent assay (ELISA) based on polyclonal antibodies used purified Vg from 17P-estradiol induced coho as the competing antigen. Reagent Vitellogenin Reagent Vg was isolated from coho plasma by a triple precipitation procedure (Norberg & Haux, 1985). A l l reagents were used at 4 °C. Four ml of 0.2M NaiEDTA containing TITJ/ml aprotinin were added to 1.0 ml of plasma. The solution was mixed and then 0.3 ml of 0.53M MgCk containing aprotinin (4 TITJ/ml) were added and again mixed. Ten ml of sterile distilled H2O were added and the resulting Vg precipitate was allowed to sit for 15 min. The precipitate was centrifuged for 15 min at 1700 rpm and the Vg pellet that collected was re-dissolved in 1M NaCl containing (4 TITJ/ml) aprotinin. This solution was dialyzed in two stages with dialysis tubing of molecular weight cut-off of 1000 (Cole-Parmer Instruments, Vernon Hills, IL). The first dialysis was for 24 hours using 1M NaCl 43 in 0.05M sodium phosphate buffer (pH 7.5) to remove the Na2EDTA and the second dialysis was for 24 hours using only 0.05M sodium phosphate buffer (pH 7.5) to remove the MgCb-Part of the clarified supernatant was diluted 1:10 with 0.05M sodium phosphate buffer (pH 7.5) and the absorbance of this diluted stock was measured at 280 nm (Varian Canada DMS100 spectrophotometer, Mississauga, ON). A n absorbance of 0.660 units corresponded to 1 mg/ml from which calibration curves could be constructed. The Vg stock solution was diluted with 50% glycerol (to prevent freezing at -20 °C) and 0.05M sodium phosphate buffer (pH 7.5) and then it was stored at -20 °C in cryo-vials until use. Antibodies A monoclonal antiserum (anti-Vg) that was induced in rabbits against trout was used. A rabbit was injected with fish serum to induce production of the anti-Vg antibody. Rabbit serum containing the monoclonal antibody was then collected, separated and used for the Vg determination of the fish plasma. The anti-serum was diluted 1:1000 in antibody diluent (100 mg bovine gamma globulin and 20 mg rabbit gamma globulin dissolved in 100 ml of phosphate buffered saline (0.01M Phosphate, pH 7.4, containing 850 mg NaCI)). Diluted antibody sera were stored at -85 °C until use. E L I S A Procedure A checkerboard titration was used for determination of the working proportions of coating Vg and the primary antibody for use in the ELISA. Twofold serial dilutions of the purified Vg were prepared in 0.05M carbonate buffer (pH 9.6) and this was delivered in 200 pi aliquots to each well of 96-well microtitration plates (Immulon 4, Dynatech, Chantilly, V A ) . The coated plates were covered and incubated overnight at 4°C. The coating reagent was removed by inversion. Excess binding sites were blocked by adding 300 pi aliquots of 1% (w/v) bovine serum albumin (BSA) in carbonate buffer to prevent anti-Vg and dissolved Vg from binding to the wells. At the conclusion of a 1 hour incubation period, this solution was removed by inversion and the plates were rinsed three time with PBS-T (phosphate buffered saline containing 500 pl/1 Tween-20 to prevent nonspecific binding of antibodies to sites in the micro wells). A dilution series of the anti-Vg in PBS-T was added column-wise to the plate in 200 ul aliquots. Each plate was sealed and allowed to incubate at room temperature for 2 hours following which, it was washed three times with PBS-T. A second antibody (goat anti-rabbit IgG labeled with alkaline phosphatase 44 (Zymed, San Francisco, CA)) was diluted 1:2000 in PBS-T and added in 200 ul aliquots to each well of the plates. After incubation at room temperature for 1 hour, this solution was removed and the plate washed with PBS-T. The enzyme substrate was then prepared (20 mg p-nitrophenyl phosphate (Sigma Chemicals) in 22 ml of diethanolamine buffer (10% v/v diethanolamine in distilled H2O, pH 9.6) and added in 200 pi aliquots to each well. The plate was gently agitated and incubated at room temperature for 45 min prior to reading the absorbance. The wells of 96-well microtitration plates were coated with Vg (125 ng/ml) with four wells per plate left uncoated to correct for non-specific binding. Coated plates were stored at 4°C until use. A standard solution of 1000 ng/ml Vg in PBS-T was twofold serial diluted to 0.98 ng/ml. Each 400 pi aliquot of Vg standard was mixed 1:1 with diluted anti-Vg (1:240,000) and incubated overnight at 4 °C in a capped vial (12x74 mm). Several Vg-free vials were included in each assay as zero binding references. These pre-incubation mixtures brought to room temperature and delivered using triplicate 200 pi aliquots per well. Immediately prior to analyses by ELISA, the plasma samples were quickly thawed and serial diluted with PBS-T. The sample dilutions were analysed in triplicate by ELISA and a calibration curve was included on each plate. The absorbance of the wells was read at 405 nm (Biotek Model EL312, Winooski, VT). The plasma dilution curve was plotted as % binding against dilution factor. Standard practices of QA/QC were conducted through the analyses. The assay's working range was 25-500 ng/ml and method detection limit was 10.5 ng/ml. 3.2.6 Proximate Analyses Proximate analyses, or the measurement of the amounts of moisture, protein, lipid and ash in diet and fish samples, are routinely conducted in nutrition studies. The information provided by proximate analyses of the fish allows assessment of the nutritional status of the fish and the subsequent calculation of the efficiency of transfer of nutrients such as protein from the feed to the fish when used in concert with data for fish growth and level of intake of the nutrient of interest. Proximate analyses were conducted at the Department of Fisheries and Oceans West Vancouver Laboratory on Day 0 (May 10, 2000) using composite samples each comprised of 5 fish. The limited size of the fish initially prevented individual determinations of proximate composition. 45 Five fish from each replicate group per diet treatment were randomly selected for individual determinations of their proximate compositions from the fish collected on the June 21, 2000 (day 28) and September 12, 2000 (day 112) sampling (total of 120 fish analyzed). Each fish was ground to a homogenous paste using a small food processor before analyses. A sample of each of the six treatment diets was also analyzed for proximate composition. Moisture and Ash Determination Moisture content was determined by weighing approximately 1.2 g of sample into a pre-weighed porcelain crucible. Subsequently, this was dried in an oven at 105 °C for 16-20 hours. A l l treatment samples were analyzed in duplicate. Following drying, desiccation and re-weighing, each of these samples were ashed at 600 °C for two hours. Protein Analyses Protein determination was based on analyses for Total Kjeldahl Nitrogen using a Technicon™ AutoAnalyzer II. A sample of 0.5g + 0.01 was weighed onto nitrogen-free weighing paper and placed into 75 ml digestion tubes containing one boiling chip each. A l l treatment samples were analyzed in duplicate. One Kjeltab™ tablet (89.7% Potassium sulphate and 10.3% cupric sulphate in approximately 3.9 g tablets; ProPac™ Powder No. CT-37) was crushed and distributed into each digestion tube. Ten ml of concentrated sulphuric acid (H2SO4, Anachemia Science) were then added to each tube and then each was allowed to stand for at least one half hour. Appropriate safety measures were taken throughout the process when using concentrated reagents and high heat. Samples were digested by carefully adding a total of 4 ml (1 ml at a time) of 30% hydrogen peroxide (H2O2, Anachemia Science) and heating to 410-430 °C in a Technicon™ BD-40 digestion heating unit. Following digestion, samples were allowed to cool for at least an hour before making up their volume to 75.0 ml. Since addition of de-ionized water following the digestion process is highly exothermic, adding approximately half the necessary volume of water and allowing at least one half hour before making up to final volume prevented the formation of a troublesome precipitate that must be completely dissolved before further analyses. Cooled digested samples were then analyzed by colourimetric measurement using the Technicon™ Auto Analyzer II using four reagent streams: sodium hypochlorite 0.315% (prepared fresh daily), working buffer solution (composed of 20% sodium hydroxide (NaOH), sodium potassium tartrate (NaKC 4H406 4H20) and sodium phosphate (Na2HP04 7H2O) solutions), sulphuric acid/sodium chloride solution and sodium salicylate/ sodium nitroprusside 46 solution. Results were calculated from comparison to a reagent standard of 50% protein and blank as measured from peaks on the strip chart record. Lipid Analyses A modification of the Bligh and Dyer (1959) method was used for total lipid determination. Duplicate 4 g sub-samples of each treatment sample were each mixed with 10 ml of chloroform (Reagent Grade, Anachemia Science) and 20 ml of methanol (Reagent Grade, Anachemia Science) and then each were mixed for 120 seconds using an Omni-mixer Homogenizer model 17105. An additional 10 ml of chloroform were added to the mixing vessel and mixed for 30 seconds. Eight ml of de-ionized water were then added and mixed for 30 seconds. Filtrate was collected by filtration under vacuum into a 500 ml filtration flask through qualitative filter paper (Whatman No. 1, Qualitative, 70 mm diameter circles). Filters were rinsed with 1:1 chloroform/methanol solution prior to decanting the filtrate into 50 ml graduated cylinders. Solutions were allowed to settle for at least one hour to ensure complete separation of layers. At the conclusion of the settling period, the volume of the bottom chloroform layer was recorded and the top layer was removed by vacuum suction. Five ml aliquots of the bottom layer were removed and placed in tared (heat 1 hour, cool, pre-weigh) foil dishes (VWR Aluminum weighing dishes No. 25433-088). The solvent was evaporated off over low heat on a hot plate in an organic tolerant fume hood. Sample dishes were dried in the drying oven for 1 hour at 105 °C, desiccated during cooling and re-weighed. Gross Energy Value Determination The gross energy value of the prepared diet samples was determined by bomb calorimetry using an I K A - W E R K E Model 5001 adiabatic bomb calorimeter with a IKA Model 5003 computer control system. Samples weighing approximately 0.5 g were analyzed using an 8 minute ignition time and a total analyses time of 17 minutes. Two bombs were calibrated prior to analyses and used alternately for sample calorific value determinations and the results were expressed as cal/g. 3.2.7 Determination of Concentrations of 4-nonylphenol Mass spectrometry is perhaps the most specific technique for detection and identification of organic compounds. It provides information about the molecular weight and structural detail and this gives a unique fingerprint to each analysis. The dynamic combination of gas 47 chromatography and mass spectrometry yields the most specific and sensitive method to characterize components of complex volatile mixtures. HPLC or L C is a powerful method for separating components that are insufficiently volatile to be separated by gas chromatography. Most organic compounds fit into this category. The specificity of mass spectrometry detection can differentiate between co-eluting components. Electrospray ionization (ESI) is the softest ionization technique available for LC-MS and it allows large labile molecules to be studied in tact. In ESI, sample molecules are simultaneously nebulized and ionized at atmospheric pressure. The sample solution is passed through a steel capillary tube situated a few millimeters away from a hollow counter electrode. A potential difference of several thousand volts is maintained between the capillary and the counter electrode so that solvents emerging from the capillary form an electrostatic spray towards the counter electrode. Gas phase ions formed by ion evaporation at atmospheric pressure are then sampled through a two-stage momentum separator into the high vacuum of the mass analyzer. Chemical Analyses of 4-Nonylphenol Dr. Natasha Hoover at the Institute for Ocean Sciences (IOS) at Sidney, British Columbia conducted all chemical analyses of project samples. A l l samples were stored at -20 °C prior to their transport on dry ice to IOS for analyses. Normal-phase liquid chromatography with electrospray mass spectrometric analyses was used to analyze for 4-nonylphenol, heptaphenol, octaphenol and decaphenol in samples of fish muscle tissue, fish livers, diets and water samples. The analytical procedure was based on the method developed at the IOS laboratory for detecting nonylphenol polyethoxylate surfactants by normal-phase LC-ESI-MS (Shang et al., 1999). Collection and Analyses of Water Sample Although the principle route of exposure of the juvenile coho to 4-NP was through the diet, it was important to quantify the water-borne concentrations of 4-NP originating from waste feed and feces. Also, it is important to know when water concentrations of 4-NP reach steady state and the degree of disturbance in the equilibrium concentration of 4-NP during routine tank cleaning. To accomplish this task, 4 litres of tank water were collected from one of the groups of fish receiving the highest diet 4-NP concentration treatment tanks every day for one week during the 48 4-NP administration period. These samples were collected from the outside 8drain of the tank. In order to determine i f exposure concentrations differed with water depth or proximity to organic matter accumulated on the bottom of the tanks, each week during the 4-NP treatment period, 4 litre samples were collected by carefully siphoning tank water from the top 1/3 depth from the water surface and from the tank drains. In the highest 4-NP concentration tank, 4 litres of water were sampled prior to tank cleaning in this fashion and after cleaning by collection through the drain. A l l samples were collected in hexane rinsed and analytically clean brown glass bottles and they were held at 4 °C until transport to IOS for analyses. Sample Preparation Treatment diets, water samples, livers and gall bladders were analysed as they were received at IOS. Frozen samples (gall bladders, livers and fish muscle tissue) were thawed to room temperature prior to analyses. Fish tissues were thawed and homogenized prior to weighing for analyses. Each batch of samples that was extracted by steam distillation at one time consisted of five samples and one blank. Between 0.1-0.5 g of each of the six treatment diets were weighed and steam distilled in 1 1 of double Mil l i -Q filtered (Millipore) water (DMQ). Limited sample weight of liver samples and gall bladders required that the whole sample be used for analyses. These samples were cut into small segments using a scalpel or homogenized where possible to increase surface area. Fish muscle tissue was collected from the dorsal region of each treatment fish and of the twelve homogenized fish tissues; three were selected and weighed in 10.0 g samples for 4-NP determination. Glassware Preparation Since many glassware cleaning products contain 4-NP, the compound of interest for analysis, every attempt was made to ensure that the glassware that was used in dosing the diets, preparing feed and collecting water samples was washed using the IOS analytical chemistry approved protocol. No soap was used in the glassware cleaning process. Glassware was rinsed thoroughly in hot tap water, scrubbed if necessary with a soap free scrubbing brush and rinsed three times with deionized water. Glassware was then rinsed at least three times with hexane (hexane rinse was retained for soaking aluminum foil). Any aluminum foil used in diet preparation and dosing, sample weighing etc was soaked in hexane in a 2 1 graduate cyclinder, rinsed and dried at 110 °C prior to use. A l l glassware used in the IOS analytical chemistry lab was cleaned using the glassware washing procedure approved for use for analyses of 4-NP and related compounds. 49 Steam Distillation Extraction Each of six flasks of the clean steam distillation apparatus was charged with measured sample, NaCl (reagent grade, BDH), boiling chips, 3 ml H 2S0 4 and 1000 ml of D M Q H 2 0. Each flask was spiked with 50 pi of C 1 3 ring labeled 4-nonylphenol internal standard (Cambridge Isotopes). Three ml of D M Q H2O and 10 ml cyclohexane were then added into each condenser. A l l solvents used were HPLC grade from Mallincrodt. Samples were boiled for 1.5 hr and the cyclohexane fraction was collected with wet Na2S04 (reagent grade, Mallincrodt). Another aliquot of 3 ml D M Q H2O and 10 ml of cyclohexane was added to the condenser and the solution was boiled for another 1.5 fir. The steam distillation process was the same for all treatment diets, fish tissue samples and water samples. Sample Clean-up The total cyclohexane fraction, approximately 20 ml, was transferred to clean glass tubes without the Na2S04 and evaporated to about 1.0 ml over a period of 3-4 hours. The extract was cleaned up using 6 amine (NH 2) solid phase extraction cartridges (500 mg stationary phase, Varian) connected in parallel to a vacuum manifold solid phase extraction apparatus. Each amine cartridge was covered with 1 g of Na2S04 and conditioned by running 3 vials of dichloromethane (DCM) and 2 complete vials of hexane that were drained to just above the Na2S04 surface. The sample was run through the column and rinsed with three 5 ml aliquots of hexane. The hexane vials were then replaced with new vials and the sample was rinsed three times with acetone. The final preparation for LC/MS analyses involved allowing the acetone fractions of the extract to evaporate under N 2 to about 1 ml over a period of about 3-4 hours. The remainder was allowed to air dry and then it was reconstituted using T D M (toluene: dichloromethane: methanol 6:2:2). The recovery standard that was taken through the analytical procedure was 2-n-butyl-3-(4-hydroxybenzoyl) benzo [b] furan (reagent grade, Aldrich). The clean-up procedure was the same for all treatment diets, fish tissue samples and water samples. Normal Phase H P L C Normal phase HPCL analysis was based on the method of Ahel and Giger (1985) with some slight modifications. A Beckman System-Gold, Model 126, HPLC system was used for the 50 chromatographic segment of the analyses (Fullerton, CA, USA). The samples were introduced using a Rheodyne sample loading injector (Model 7225) with a 3 pi loop and then they were eluted through a 25 cm x 4.6 mm inside diameter (ID) column packed with 5 pm dp NH2 Hypersil APS-1 from Phenomenex. Chromatographic analyses were conducted under isocratic conditions using the mobile phase of toluene: methanol: D M Q H2O, (10:88:2) at a flow rate of 0.28 ml/min. The system was equilibrated for at least 15 min between injections in order to achieve constant retention times. Liquid Chromatography Electrospray Ionization Mass Spectrometry Determination A l l treatment diets, fish tissue samples, water samples and recovery standards were analyzed for 4-NP, octylphenol, heptaphenol and decaphenol in the same fashion. Following elution through the HPLC system described above, samples were introduced to a triple quadrupole V G Quattro tandem mass spectrometer equipped with an electrospray source from Micromass (Manchester, UK). Only the negative ion mode was used for selected-ion monitoring (SIM) not full scan mode. The ESI probe is a pneumatically assisted system that uses N2 as the nebulizing gas at approximately 80 psi and flow rate between 6-7 ml/min. The ion source and lenses settings used were: source temperature 80 °C; drying gas flow rate 0.3 ml/min; ESI capillary voltage 2.85 kV; H V lens at 520 V (negative ion mode); cone voltage between 29-35 V ; skimmer offset at 0 V ; lens-3 between 20-35 V and ion energy between 3.4-4.1 V . A calibration standard solution of 4-NP was prepared and analysed at various concentrations with the resulting calibration curve used for quantitative determination of samples by comparison. An internal standard of C-labelled 4-NP was taken through the procedure in all sample runs. A l l samples and standards were run in duplicate. The recovery standard for analytical determination is most often a compound that is not common in nature so that interference from sample constituents would be minimal. For all 4-NP analyses, the recovery standard was 2-n-butyl-3(4-hydroxy-benzoyl) benzo[b] furan. The detection limit of the method varied depending on the sample matrix and was based on a 3:1 signal to noise ratio. For most of the samples, the detection limits were between 3-4 ng/g. 3.3 Data Calculated and Statistical Analyses A l l proximate calculations were made on a dry weight basis. Dry food intake data (DFI g/fish) were used to calculate feed efficiency (FE; weight gain (g)/DFI) and the gross energy utilization (GEU = gross energy gain (MJ/g) X 100/gross energy intake (MJ/g). The values determined 51 through whole body proximate analyses for protein and lipid composition were used to estimate body energy content. The gross energy gain was estimated by ascribing 0.0236 MJ/g protein and 0.0395 MJ/g lipid (Higgs et al., 1995). Specific growth rates (SGR = % wet weight/day) were derived from the covariate slopes for each diet as follows: SGR = loge[weight(fmai)] -loge[weight(initiai)] X 100/interval in days. The protein efficiency ratio (PER) was calculated from the wet weight gain/protein intake. Percent protein deposited (PPD) was calculated as protein gain X 100/protein intake. Condition factor was calculated as weight/length3 2 5 and the hepatosomatic index was calculated as the liver weight/fish weight. Frequency histograms of weigh distributions were constructed from statistical comparison between the weight measurements of treatment and control groups was performed using one way A N O V A . The experiment was designed as a randomized block design. Each block constitutes a replicate of the treatments. The treatments were assigned to 6 randomly chosen tanks within each of 2 blocks. The analysis of variance method used to assess dietary performance parameters was a two-way procedure used for randomized complete block designs. The source of the variance for the two-way classification within the experiment can be summarized as: Source Degrees of Freedom Block 1 Treatment 6-1=5 Experimental Error (2-1 )(6-1) = 5 Sampling interaction 2 x 6 (80-1) = 948 Version 2000 of the SigmaStat™ Statistical computer software program was used to conduct statistical analyses. In order to ensure equality of variance in conducting multi-comparison tests of means, data represented by proportions were transformed before statistical analyses. Since the variance of a proportion is dependant on the proportion itself, the variance can be represented by replacing the original measurement with an arcsine square root transformed value and comparison of means can be performed without the use of non-parametric tests. In order to minimize the risk of committing a Type I error in comparing the means of observations at the P = 0.05 level, Student Newman-Keuls test was used. Where recommended 52 by the SigmaStat™ Program, Bonferoni adjustment was used as a more conservative comparison between paired observations when it was not appropriate to use Student Newman-Keuls test. Newman-Keuls test is a conservative method in which means differ less than the student t-value. 53 RESULTS "It is a capital mistake to theorize before one has data. Insensibly, one begins to twist facts to suit theories instead of theories to suit facts. " Sherlock Holmes 4.1 Fish Growth - Lengths and Weight Measurement A summary table of the lengths, weights, sexes, liver weights, condition factor and hepatosomatic index of the twelve fish used for analyses of 4-NP and for which plasma was reserved for vitellogenin analyses is found in Appendix B. In general, normal weight and length distributions were observed for each replicate group per diet treatment on days 28, 56, 84 and 112 of the study (Figures 4.1-4.4). No statistical significant difference were found between the mean fish weights among the treatments at the conclusion of the beginning of the trial and at the end of the freshwater and sea water phases of the experiment for all groups of fish. 54 (fl 30 25 20 2 15 | 10 3 Z 5 Control (1) 11 12 13 14 15 16 17 18 19 18+ Fish Weight (g) 30 SI 25 (fl iZ 20 >*-o i— 15 o Si 10 E 3 Z 5 Control (2) 1 r—i 1 1 11 12 13 14 15 16 17 18 19 18+ Fish Weight (g) 30 25 w i l 20 o 1— 15 CU Si 10 E 3 z 5 0.002 mg 4NP/kg (1) ---— r - i 11 12 13 14 15 16 17 18 19 18+ Fish Weight (g) 0.002 mg 4NP/kg (2) 11 12 13 14 15 16 17 18 19 18+ Fish Weight (g) 30 SZ 25 CO i l 20 o 15 0) SI 10 £ 3 5 Z 0.2 mg 4NP/kg (1) j t—l 11 12 13 14 15 16 17 18 19 18+ Fish Weight (g) 0.2 mg 4NP/kg (2) 30 s: 25 jfl iZ 20 o L _ 15 V Si 10 E Nu 5 1 , n 11 12 13 14 15 16 17 18 19 18+ Fish Weight (g) Figure 4.1: Fish Weight Distribution Histograms for the May 10, 2000 Sampling 55 30 SZ 25 .52 iZ 20 -o 15 Q) .Q 10 E 3 5 Z 0 J 20 mg 4NP/kg (1) X L 11 12 13 14 15 16 17 18 19 18+ Fish Weight (g) 30 J C 25 co u_ 20 o I — 15 <D 10 E Z3 Z 5 20 mg 4NP/kg (2) X L 11 12 13 14 15 16 17 18 19 18+ Fish Weight (g) 30 sz 25 co u_ 20 n-O 15 JH 10 E zz 5 ~Z-0 2000 mg 4NP/kg (1) £L 11 12 13 14 15 16 17 18 19 18+ Fish Weight (g) 2000 mg 4NP/kg (2) -?n s: 25 £ 20 -2 15-- ° 10 = 5 1 0 - n 11 12 13 14 15 16 17 18 19 18+ Fish Weight (g) 30 sz 25 CO UL 20 X -O 15 i — CD Si 10 E 3 5 Z E 2 - 31 mg/kg (1) X L 11 12 13 14 15 16 17 18 19 18+ Fish Weight (g) 30 sz 25 to ii 20 Z 15 CD • | 10 Z 5 0 E 2 -31 mg/kg (2) — n 11 12 13 14 15 16 17 18 19 18+ Fish Weight (g) Figure 4.1 (con't): Fish Weight Distribution Histograms for the May 10, 2000 Sampling 56 in 30 25 •20 £15 e J310 E 3 5 Control (1) & N f e N% # «(V ^ <$> <$> <§> ^ Fish Weight (g) 30 SI 25 .2 ^20 f 15 E 3 5 Control (2) ^ <V K > N<b N% ^ ^ ^ ^ ^ x Fish Weight (g) 30 s: 25 ^20 4 -°15 (l> -Q 10 E 3 5 0.002 mg4NP/kg (1) XLXLXL •$> <v ^  sfc ^  c(y ^ <$> ^ ^ ^x Fish Weight (g) E 3 30 25 20 15 10 5 0 0.002 mg 4NP/kg (2) • — m n II n \^ O \* \% <£> ^ ^ «$> Fish Weight (g) 30 .C 25 £2 ^-20 »*-°15 a) J3 10 E 3 5 0 0.2 mg 4NP/kg (1) XL H X I N f e N% 0> ^ qfe ^ ^ ^ Fish Weight (g) (A 30 25 20 0.2 mg 4NP/kg (2) - 1 5 <D •a 10 E 3 5 0 a X L Fish Weight (g) Figure 4.2: Fish Weight Distribution Histograms for the June 21, 2000 Sampling 57 30 I" "-20 4 -a> -Q 10 E = 5 20 mg 4NP/kg(1) XL XL •\° \% <yr <y> ^ n<b n <b oSl Fish Weight (g) 30 I 2 5 "-20 4 -f 1 5 a> -Q 10 E = 5 0 20 mg 4NP/kg (2) £ <V N * Ncb N % ^ ^ ^ ^ ^ ^ Fish Weight (g) 30 "-20 4 -2 1 5 •Q10 2 5 2000 mg 4NP/kg(1) XL ^ & $. $ <£> e$ $> <$> „p Fish Weight (g) 30 "-20 ° 1 5 •Q10 2 5 0 2000 mg 4NP/kg (2) X L Fish Weight (g) 30 "-20 f 15 CD . 0 1 0 E = 5 E2-31mg/kg (1) — — — n n ^ ^ K V \ f e \ " f V f f ®"„sr K<b „ < i ,s\, nv n <b n <b Fish Weight (g) 30 •=25 .2 "-20 2 is CD -Q 10 E = 5 z 0 E 2 - 31 mg/kg (2) n II II II $ <v Nv sfe ^, ^ ^ £ ^ ^ ^ ^ x Fish Weight (g) Figure 4.2 (con't): Fish Weight Distribution Histograms for the June 21, 2000 Sampling 58 co 30 25 •20 -Q10 E 3 5 Z 0 Control (1) ..LT n ^ ^ # # ^ uv ^ # <£> X Fish Weight (g) 30 ^ 2 5 .52 " - 2 0 « 10 E = 5 0 Control (2) pa n . n n • & <$> <& & & & & & & & Fish Weight (g) 30 I 2 5 " - 2 0 4 -f 15 a> -Q10 = 5 0.002 mg 4NP/kg (1) T T , TH n ^ >^ # n£ ^ ^ £> # # # Fish Weight (g) 30 ! " - 2 0 a> •Q10 E = 5 0 0.002 mg 4NP/kg (2) <P tf> & & & # tf> <£> ^ Fish Weight (g) co 30 i 2 5 •20 0.2 mg4NP/kg(1) - 1 5 QJ •a 10 E 3 5 0 E L ^ & ^ t?> * $ # Fish Weight (g) 30 I 2 5 " - 2 0 n-S« e - ° 1 0 z 5 0 0.2 mg 4NP/kg (2) X L . ^ <#> $ # £ # #> # # # ^ Fish Weight (g) Figure 4.3: Fish Weight Distribution Histograms for the August 02, 2000 Sampling 59 30 •=25 v> ^20 215 0) • O 1 0 E z 5 0 20 mg 4NP/kg (1) i l l . M o r£> & r§= £ # ^ # # Fish Weight (g) 30 I25 "-20 4 -215 o -|10 = 5 20 mg 4NP/kg (2) n • LT r-i n & ^ <§> <§> £ & & # # ^ Fish Weight (g) 30 i 2 5 •20 2000 mg 4NP/kg(1) 2«H a* ^Q10 E = 5 0 n r£> ^ <£> # # S> «$V g$> Fish Weight (g) 30 i 2000 mg 4NP/kg (2) » 2 5 "-20 2 « o -o 10 z 5 0 X L <V* <v* & <§> £ # # «?> # ^ Fish Weight (g) 30 -,: •=25 S2 "-20 215 s •Q10 E => 5 E 2 -31 mg/kg (1) $ ^ ^  ^  # # # $ & # $ . Fish Weight (g) 30 25 •20 - 1 5 ^ 2 1 0 E 2 - 31 mg/kg (2) XL XL n? <v* # •#> tf> * # # # ^ Fish Weight (g) Figure 4.3 (con't) : Fish Weight Distribution Histograms for the August 02, 2000 Sampling 60 30 "•20 i— 215 <D JO10 E 3 5 Control (1) ^ <$> # ^ $ # ^ Ntf Fish Weight (g) sz to szt E 3 30 25 20 15 10 5 Control (2) X L n? <$> tf> <£> <§> <v* <§> <§> ^ ^ Ntf Fish Weight (g) 30 •=25 1 1 2 0 n— 215 a> J O 10 E 3 5 0 0.002 mg 4NP/kg (1) XL <§> $> <§> ^ <$> <$> * ^ Fish Weight (g) 30 •=25 1 1 2 0 4— £ 1 5 o J Q 1 0 E 3 5 0.002 mg 4NP/kg (2) * r MM J - . ^ . . . , .H B _ <£> # <§> ^ # ^ ^ A* Fish Weight (g) 30 J = 2 5 .52 ^20 «-215 JQ10 E 3 5 0.2 mg 4NP/kg (1) . C L £ «P ^ <$> j§> K& Ktf Fish Weight (g) co 30 25 •20 Z15 co JO10 E 3 5 0 0.2 mg 4NP/kg (2) " — • n 11 i—i (—i T? «S> <S> 5? ^  9? # # Ntf \ Fish Weight (g) Figure 4.4: Fish Weight Distribution Histograms for the September 12, 2000 Sampling 61 30 20 mg 4NP/kg (1) 30 tf>" " - 2 0 215 a -Q10 E 20 mg 4NP/kg (2) • = 2 5 "-20 -x-° 1 5 o J 3 1 0 -E 3 5 0 i — — mm n n ™* 0 z 0 -1 " • n n ? <§> & <p <§> # # ^ Fish Weight (g) 5 ^ ^ 9? c$> »$> K t f Fish Weight (g) 30 .I25 " - 2 0 215 <o •Q10 E 3 5 2000 mg 4NP/kg (1) XL $ $ <§> & \* & <§> ^ ^ ^ Fish Weight (g) 30 -=25 tf> ^ 2 0 2« JD10 E = 5 2000 mg 4NP/kg (2) n XL «P <P & & *P <§> & ^ ^ Ntf Fish Weight (g) 30 • = 2 5 " - 2 0 t-215 a> •Q10 E 3 5 E 2-31mg/kg(1) - n IT n n? # ^ <p ^  >tf Fish Weight (g) 30 • = 2 5 tf>" " - 2 0 . Q 1 0 E 3 5 E 2 -31 mg/kg (2) m n IT n i—i i—i # £ ^ 9? # . <J> K ^ Fish Weight (g) Figure 4.4 (con't): Fish Weight Distribution Histograms for the September 12, 2000 Sampling 62 None of the dietary concentrations of 4-NP had any significant effect on coho salmon weight gains or growth rates, dry feed intake, conversion of dietary protein into flesh (PER) or body protein (PPD), conversion of dietary energy into body energy (GEU) or percent survival during the freshwater phase of the study. Moreover, treatment of the fish with 4-NP before their transfer to sea water, irrespective of the dietary concentration, did not influence their subsequent performance in sea water when all fish were fed the control diet. Dietary performance data is presented in Tables 4.1 and 4.2. Table 4.1: Summary of dietary performance and survival - fresh water phase. Treatment Initial Fish Wt. Gain SGR DFI FE PER PPD GEU Percent mg/kg 4NP Wt. G g/fish g/fish Survival Control 15.15 5.77 1.24 6.99 0.83 1.67 35.65 30.80 99 0.002 15.22 6.45 1.36 6.57 0.98 1.97 42.89 32.89 98 0.2 15.28 6.17 1.31 6.63 0.93 1.87 44.67 33.55 98 20 15.35 5.87 1.24 6.15 0.96 1.92 46.10 31.09 99 2000 15.21 5.75 1.23 6.82 0.85 1.69 40.76 31.73 98 E2 15.21 7.93 1.61 6.79 1.17 2.34 48.33 31.60 100 Dietary treatment of the fish with E2 did, however, result in enhanced fish growth and feed efficiency as well as protein (PER) and energy utilization during the freshwater phase of the study. However, the subsequent growth, feed and energy utilization of these fish was noted to be poorer relative to all other groups during the sea water phase of the study when all fish were given the control diet. The differences in the preceding performance parameters among the groups during the sea water phase were not found to be significant. Table 4.2: Summary of dietary performance and survival - sea water phase. Treatment Initial Fish Wt. Gain SGR DFI FE PER PPD GEU Percent mg/kg 4NP Wt. G g/fish g/fish Survival Control 20.92 43.40 1.37 45.66 0.95 1.93 30.33 29.27 96 0.002 21.68 43.37 1.34 44.30 0.98 1.96 31.38 29.49 94 0.2 21.40 44.09 1.36 43.93 1.00 2.02 33.29 32.01 97 20 21.22 44.38 1.37 43.06 1.03 2.07 33.32 29.37 100 2000 20.97 43.08 1.36 43.48 0.99 1.97 32.23 28.90 94 E2 23.14 36.47 1.15 42.89 0.85 1.71 28.77 31.97 98 63 Percent survival during the sea water phase ranged between 94 and 100% compared to the percent survival of 98=100% in the fresh water phase (illustrated in Figure 4.5). Figures 4.6 -4.8 illustrate the comparison of performance parameters during the freshwater and seawater exposure phases for the various treatments. 4.2 Dietary Performance Results In general, juvenile coho salmon growth, dry food intake, feed and protein efficiency, percent survival, gross energy utilization and whole body proximate compositions were not influenced by dietary exposure to 4-nonylphenol at the concentrations selected and under the given experimental conditions. There was an increase in wet weight gain, specific growth rate, feed efficiency, protein efficiency ratio and percent protein deposited during the time that the estrogen (E2) treatment groups were being fed the treated diet however, this trend appeared to be reversed during the sea water growth phase when all treatment groups were being fed basal diet. These lower dietary performance parameters observed only for the estrogen (E2) treatment fish in sea water were not statistically significantly lower at P = 0.05. Percent Survival O Fresh Water Phase • Sea Water Phase T r e a t m e n t Figure 4.5: Percent survivals of the fish during the fresh water and sea water phases of the study in relation to dietary treatment. A l l fish were fed the control diet during the sea water phase. 64 Dry F o o d Intake a Fresh Water Phase • Sea Water Phase Control 0.002 0.2 20 treatment 2000 E2 1.20 r i Feed E f f i c i ency • Fresh Water Phase • Sea Water Phase Control 0.002 0.2 20 2000 treatment E2 Figure 4.6: Dry food intakes and feed efficiencies of the groups of coho during the fresh water and the sea water phases of the study in relation to diet treatment. A l l fish were fed the control diet during the sea water phase. 65 G r o s s Ene r gy Utilization Figure 4.7: Specific growth rates and gross energy utilization in the groups of coho salmon during the fresh water and sea water phases of the study in relation to dietary treatment. A l l fish were fed the control diet during the sea water phase. 6 6 Percent Protein Deposition 50.00 45.00 40.00 35.00 30.00 c <D O 25.00 Q Q . 20.00 15.00 10.00 5.00 0.00 I Fresh Water Phase I Sea Water Phase Control 0.002 0.2 20 t reatment 2.50 2.00 1.50 1.00 0.50 0.00 Protein Efficiency Ratio Control 0.002 0.2 20 2000 E2 t reatment I Fresh Water Phase I Sea Water Phase Figure 4.8: Values for protein efficiency ratio and percent protein deposition for the groups of coho salmon during the fresh water and sea water phases of the study in relation to dietary treatment. A l l fish were fed the control diet during the sea water phase. 67 Figure 4.9 illustrates the variation of condition factors of the groups at the end of the freshwater and sea water phases of the study in relation to the diet treatment. The significant difference among the groups for condition factors reflected the estrogen (E2) treatment group. Fish treated with E2 had higher condition factor than those noted for other groups during the freshwater phase. During the fresh water phase there were no significant differences in the condition factors for the fish given the 4-NP treatments. Also, there were no significant differences in the condition factors of the treatment groups during the sea water phase of the study. Figure 4.9 also illustrates the hepatosomatic indices of the fish in relation to the diet treatment, as calculated liver weight/fish weight. During the fresh water phase, the estrogen (E2) treatment group had a statistically significantly higher hepatosomatic index than the other treatment groups. There were no significant differences in the hepatosomatic indices for any of the treatment groups during the sea water phase. 68 Condition Factor 0.0060 0.0060 0.0059 0.0059 0.0058 0.0058 0.0057 0.0057 0.0056 0.0056 Control 0.002 • Fresh Water • Sea Water 0.2 20 Treatment 0.035 0.030 0.025 0.000 Control 0.002 Hepatosomat i c Index 0.2 20 Treatment 2000 O Fresh Water • Sea Water E2 Figure 4.9: Condition factors and hepatosomatic indices of the coho salmon at the end of the fresh water and sea water phases of the study in relation to diet treatment. 69 4.3 Hematocrit Measurement No significant differences were found among the groups for hematocrit at the end of the fresh water phase of the study using a two-way randomized block A N O V A (Figure 4.10). Hematocrit Index-June 21 ,2000 50 i Treatment Figure 4.10: Values for hematocrit (%) of the groups at the conclusion of fresh water phase of the study in relation to diet treatment. 4.4 Seawater Challenge Test and Fish Appearance Potassium concentrations are measured at the same time as sodium concentrations when clinical flame photometers are used in the plasma analyses. Lysed samples often have high potassium concentrations and low sodium concentrations (Blackburn & Clarke, 1987). A l l samples had low potassium concentrations (Figure 4.12), between 4.00 - 6.00 m M when they were compared 70 to the sodium concentrations (Figure 4.11) and this indicated that the samples were not lysed at the time of analyses. No significant differences were found between groups at the end of the freshwater phase for either blood sodium or potassium concentrations using a two-way randomized block A N O V A . Sodium - Plasma June 21/2000 180.00 Treatment Figure 4.11: Plasma sodium concentrations - June 21/2000 sampling. 71 Potassium - Plasma June 21/2000 6.000 , Control 0.002 0.2 20 2000 Treatment Figure 4.12: Plasma potassium concentrations - June 21, 2000 sampling Although differences were not observed for plasma sodium concentration of the fish at the end of the fresh water phase, some trends were noted with the number of fish with parr marks among groups. One particular observation regarding coloration of fish was noted during the August 6, 2000 measurement of fish. The characteristic dark striped markings or "parr marks" observed in all the fish during the June 21, 2000 sampling were retained in some of the fish on August 6 whereas some of the fish had undergone the obvious morphological changes that accompany smoltification and appeared silvery with no dark stripes crossing their lateral lines. The ability to distinguish two different sub-groups of fish within the treatments based on their colour led to percent estimates of the fish that retained parr marks (Figure 4.13). Qualitative observation of the fish suggested that one of the estrogen (E2) treatment groups and both of the 0.002 mg/kg 4-NP treatment groups had retained higher percentages of fish with parr marks. The average percentage of parr marks retained in other 4-NP treatment groups seemed similar between 72 groups, although the control groups appeared to have a lower percent retention of parr marks in both duplicate treatment groups. Since coloration and morphological changes during smoltification are intimately linked to hormonal changes within the fish, it could be possible that exposure to 4-NP had a subtle effect on this process. Since this observation is highly subjective, without quantitative determination of a pigment compound or an enzyme linked to the colour change at parr smolt transformation, it is impossible to correlate parr mark retention with diet treatment. Fish With Parr Marks Aug 2/2000 Control E2 0.002 0.2 20 2000 Treatment Figure 4.13: Retention of parr marks in coho salmon after four weeks of growth in sea water. A l l fish were fed the control diet following sea water transfer. 4.5 Proximate Analyses Table 4.3 provides the concentrations of moisture, dry matter (DM), ash, protein, lipid and gross energy in the treatment diets. A l l diets contained similar levels of proximate constituents and gross energy. 73 Table 4.3: Proximate composition of treatment diets. Treatment Moisture DM Ash Protein Lipid Total Gross Energy (•%) (%) (%) (%) (%) (%) (MJ/kg) Control 9.4 90.6 8.8 49.5 18.0 85.7 21.2 0.002 mg/kg 4NP 9.3 90.7 8.8 49.8 19.5 87.4 21.2 0.2 mg/kg 4NP 9.3 90.7 8.8 49.7 19.4 87.2 21.0 20 mg/kg 4NP 9.0 91.0 8.8 49.9 19.2 86.9 21.2 2000 mg/kg 4NP 8.9 91.1 8.8 50.1 18.2 86.0 21.3 E 2 - 31 mg/kg 9.0 91.0 8.9 49.9 20.1 88.0 21.2 Table 4.4 provides the concentrations of moisture, ash, protein and lipid analyses in the fish at the conclusion of the freshwater phase (June fish) and Table 4.5 provides the same parameters at the end of the seawater growth (September fish) period. A l l analytical results of proximate composition determinations can be found in appendix B. Table 4.4 Proximate composition of fish - end of the fresh water phase Treatment Protein % Lipid % Moisture % Ash % Control 15.7 7.1 74.4 2.6 0.002 mg/kg 4NP 16.0 8.0 74.1 2.5 0.2 mg/kg 4NP 16.5 8.0 73.7 2.5 20 mg/kg 4NP 16.4 6.9 74.3 2.5 2000 mg/kg 4NP 16.4 7.2 73.9 2.5 E 2 - 31mg/kg 16.0 7.4 73.6 2.5 74 Table 4.5 Proximate composition of fish - end of the sea water phase Treatment Protein % Lip id % Moisture % Ash % Control 16.4 6.9 74.0 2.2 0.002 mg/kg 4NP 16.0 6.0 75.4 2.0 0.2 mg/kg 4NP 16.1 6.6 74.6 2.2 20 mg/kg 4NP 15.8 6.4 75.0 2.2 2000 mg/kg 4NP 16.4 7.1 73.8 2.1 E 2 - 31 mg/kg 16.0 5.5 75.4 2.2 A l l proximate constituent parameters were expressed as percent and thus all statistical analyses were conducted using the arcsine square root transformed data. No significant differences were found between any of the treatment groups in terms of their concentrations of moisture, ash, protein or lipid at the end of the fresh water and sea water phases of the study. Figures 4.14 and 4.15 are graphical representations of the aforementioned proximate composition data and they provide no indication of a treatment effect. 75 Composit ion - Protein 20.00 15.00 Q. 5.00 0.00 Control 0.002 0.2 20 2000 E2 Treatment Composit ion - Lipid 10.00 Control 0.002 0.2 20 2000 E2 Treatment o S e a Water B Fresh Water a Sea Water • Fresh Water Figure 4.14: Proximate composition of the fish at the conclusion of the fresh water and sea water phases of the study - protein and lipid. 76 Composit ion - Moisture 100.00 80.00 60.00 0) u a 40.00 20.00 0.00 • Sea Water • Fresh Water Control 0.002 0.2 20 2000 E2 Treatment Composit ion - Ash 3.00 2.50 2.00 I o 1— 1.00 0.50 0.00 • Sea Water • Fresh Water JIB Control 0.002 0.2 20 Treatment Figure 4.15: Proximate composition of the fish at the conclusion of the fresh water and sea water phases of the study - moisture and ash. 77 4.6 T 3 and T 4 Hormone Measurement Due to volume constraints, some plasma samples were pooled to obtain sufficient volume for the assay. For some of the June fish plasma samples, up to five separate plasma samples were combined. In most cases, single samples of the twelve individual plasma samples provided enough volume for the assay. Many of the September fish plasma samples comprised enough plasma volume such that sample pooling within the treatment groups was unnecessary. Table 4.6 provides the plasma T 3 and T 4 concentrations in coho salmon following the fresh water phase and Table 4.7 illustrates the same parameters after the sea water phase. Table 4.6: Plasma T 3 and T 4 concentrations in coho salmon in relation to diet treatment at the end of the fresh water phase. Fresh Water Phase Sample T 3 nmol/L Ave. T 4nmol/L Ave. Control #9 1.038 1.450 1.937 1.569 (Tank 228) #3,4 1.876 0.734 #2,10 1.434 2.036 Control #2 2.242 1.259 2.760 2.701 (Tank 234) #3 1.387 2.927 #6 0.149 2.417 0.002 mg 4NP/kg #6 1.794 1.298 0.891 1.257 (Tank 227) #10 1.951 1.781 #3,6 0.149 1.097 0.002 mg 4NP/kg #3,10 0.149 0.790 1.354 1.774 (Tank 232) #9,12 2.074 2.037 #7,11 0.149 1.930 0.2 mg 4NP/kg #5,6,8,9 2.637 1.828 1.664 1.258 (Tank 225) #1,2,3,11 1.019 0.852 0.2 mg 4NP/kg #5 1.737 1.052 1.730 2.922 (Tank 231) #11 0.742 3.974 #12 0.676 3.062 20 mg 4NP/kg #3 1.905 2.307 2.255 1.715 (Tank 226) #4 2.972 2.010 #11,12 2.045 0.879 20 mg 4NP/kg #9 1.816 1.580 10.053 2.780 (Tank 230) #3,11 1.351 3.650 #2,6,7 1.575 1.911 2000 mg 4NP/kg #5 1.909 2.024 1.577 2.624 (Tank 223) #6 4.013 5.665 #11 0.149 0.630 2000 mg 4NP/kg #4 2.233 1.624 1.418 1.405 (Tank 229) #7,11 1.258 1.161 #1,3,5,6,8 1.380 1.635 E 2 31 mg/kg #3 0.149 0.324 0.945 0.442 (Tank 224) #4 0.149 0.269 #8 0.673 0.112 E 2 31 mg/kg #2 0.149 0.149 0.797 0.938 (Tank 233) #4 0.149 1.060 #10 0.149 0.957 78 Table 4.7: Plasma T 3 and T 4 concentrations in coho salmon in relation to diet treatment at the end of the sea water phase. A l l fish were fed the control diet in the sea water phase. Sea Water Phase Sample T 3 nmol/L Ave. T 4 nmol/L Ave. Control #7 4.124 2.502 4.316 3.734 (Tank 228) #9 1.608 3.131 #11 1.773 3.754 Control #2 2.206 1.834 2.998 3.275 (Tank 234) #6 1.747 2.419 #11 1.548 4.408 0.002 mg 4NP/kg #1 4.343 3.787 2.258 3.495 (Tank 227) #2 4.037 5.485 #10 2.979 2.741 0.002 mg 4NP/kg #1 2.438 1.681 5.441 4.831 (Tank 232) #2 0.874 4.449 #8 1.731 4.604 0.2 mg 4NP/kg #4 2.509 2.131 3.159 3.081 (Tank 225) #8 1.342 2.657 #12 2.542 3.428 0.2 mg 4NP/kg #5 1.914 1.234 4.121 3.062 (Tank 231) #9 1.469 2.925 #11 0.318 2.139 20 mg 4NP/kg #2 1.574 2.297 2.743 4.471 (Tank 226) #3 2.307 5.599 #7 3.009 5.071 20 mg 4NP/kg #3 3.353 2.349 3.495 4.363 (Tank 230) #8 1.650 2.860 #9 2.045 6.735 2000 mg 4NP/kg #3 2.518 1.466 2.948 2.756 (Tank 223) #5 0.242 1.795 #10 1.638 3.526 2000 mg 4NP/kg #2 0.169 2.038 4.101 5.007 (Tank 229) #5 3.298 6.546 #12 2.645 4.373 E 2 31 mg/kg #2 3.821 3.251 5.027 5.536 (Tank 224) #6 2.979 3.519 #8 2.953 8.061 E 2 31 mg/kg #2 1.210 1.866 2.479 6.991 (Tank 233) #4 1.712 4.094 #6 2.675 14.399 Results for pooled samples of T 3 and T 4 for both the freshwater and sea water periods were statistically analyzed by two-way randomized block A N O V A . At P=0.05, there was no significant difference between all treatment groups for the seawater phase for both T 3 and T 4 . As well at P=0.05, there was no significant difference among treatment groups for plasma T 4 concentrations during the fresh water period of exposure to the diet treatments. However, 79 plasma T 3 concentrations in the estrogen (E2) treated fish were lower than those found in the plasma offish given the other treatments at the end of the fresh water exposure phase. 4.7 Vitellogenin Assay Plasma vitellogenin concentrations were to be measured in twelve fish from each replicate per treatment by Dr. Jim Sherry at the Canadian Centre for Inland Waters at Burlington, Ontario. At the time of completion of this thesis, no vitellogenin results were available (analyses still pending). 4.8 Chemical Analyses of 4-Nonylphenol Concentrations of 4-nonylphenol were determined in the treatment diets, fish dorsal muscle tissue, livers, gall bladders and water samples by ESI LC-MS. Decylphenol (DP), heptylphenol (HP), octylphenol (OP) and 4-nonylphenol (NP) results were recorded on each selected ion chromatogram. The recovery standard for the analytical procedure was 2-n-butyl-3(4-hydroxy-benzoyl) benzo[b] furan and more than 80% recovery was obtained for all analyses. Failure to obtain at least 80% recovery meant that the analyses were repeated. The internal standard, 1 3 C labeled 4-NP, was taken through the extraction process and analytical determination for every five samples. The relative recovery factor for all analyses was within the range of 80-85%. Figure 4.17 illustrates the selected ion chromatogram of analyses for one recovery standard, the accompanying internal standard and the accompanying calibration 4-NP standard. Figure 4.18 is a sample selected ion chromatogram from one water sample (June 14, 2000 - drain) with accompanying internal and calibration standard scans. 80 AUG0OI3 100 15:55:49 08-Aug-2000 SIR of 6 Channels ES-293:20 5.4Se5 Area a) Recovery Standard 2-n-butyl-3(4-hydrpxy-henzyl)benzo[b]furan IAUG0013 100-, 4.92 100844" JAUG0013 100-2.012.68 0.99 84675360,3^9 2496 ^ V 2 7 7 4 2 ' ' ' ' ' ' SIR of 6 Channels ES-225.10 1.20e6 Area b) Internal Standard 1 3C-labelled 4-NP 4.92; 144785* SIR of 6 Channels ES-219.00J l:l4e« Area! c) Calibration Standard 4-NP Time! 'l-W " 1' W • j.bo ' io'.oo 12-QQ Ijjjg ; ; i ^ L Figure 4.16: Selected ion chromatogram for recovery, internal and calibration standards. 81 CDW1 rCohFDiit;r?CWDiet Proj.), NP-anaiysis,:"flH2-col( CN,guard-col),80o, 28 ml/mi, 3 ul «j47^0.^Odj20Oq OCT0017 9 2 1 1 0 0 n 79560-1 1.45: „ „ 3.29 13348 2 78 4242 9076 SIR o f 6 Channels ES-293.20 4.00e5 Area a) Water Sample June 14,2000 - drain IOCT0017 100-, locroon 100-4.87 109232"! Slk o f 6 Channels: ES-225.10) 7.99e5 Area! b) Internal Standard 1 3C-labelled 4-NP 4.87 328682" Slk o f 6 Channels ES-219.00, 2.25e<S Areal 0.43 f i l l 2.53 2409 c) Calibration Standard 4-NP ' ' ' 6.66 ' 4.60 Timer ' 10.00 12.00 14.00 Figure 4.17: Example selected ion chromatogram for one water sample with internal and calibration standards 82 Table 4.8 provides the analytical results for 4-NP in the treatment diets and selected dietary components. A l l diet samples were analyzed twice (set A and B) with the exception of the 2000 mg 4-NP/kg diet sample. In all tables of analytical results n/d means that a sample peak was not detected above the background noise signal for the instrument. The detection limit for all samples was usually above a 3:1 signal to noise ratio (between 3-4 ng/g). The first analysis of the samples revealed that the 20 mg/kg dose was 13 mg/kg. Since this dose was less than the desired diet dose, a second preparation of this diet was made. This second analysis of the 20 mg/kg diet gave 43-44 mg/kg 4-NP (indicated on table as 20 mg/kg (dup)). The fish were fed this diet concentration. The 2000 mg/kg 4-NP diet sample was analysed once. Since all other samples being run for this experiment and other research experiments concurrently conducted at IOS were expected to contain 4-NP concentrations orders of magnitude less than 2000 mg/kg, this sample was run once to prevent gross contamination of the analytical equipment that required rigorous cleaning after the analysis of the very high concentration was completed. The 2000 mg/kg 4-NP sample was determined to contain 1355 mg/kg on an air-dry basis or 1489 mg/kg on a dry weight basis (diet samples were 9.1% moisture). Although the actual concentrations were not the expected doses of 4-NP with all the low doses containing 1-5 mg/kg 4-NP, a broad concentration range of three orders of magnitude in the diets was achieved. Initially, krill and marine samples were not anticipated to contain any measurable concentrations of 4-NP or the other three alkylphenols which were simultaneously determined. Decylphenol, heptylphenol and octylphenol are usually determined at the same time as nonylphenol although these compounds are generally taken to be sample contaminants. Analyses were repeated using different sample sizes in the extraction procedure to confirm the results first obtained. The repeated analyses confirmed the relative concentrations of 4-NP, DP, HP and OP in the diet samples, krill and marine oil in the first reported results allowing for variation between extraction runs. Although detectable concentration for 4-NP and other alkylphenol contaminants were present in the estrogen (E2) treatment, control, 0.002 mg/kg and 0.2 mg/kg 4-NP doses, these concentrations were small compared to the highest diet 4-NP dose which failed to produce a statistically significant difference in the performance parameters of the fish as previously mentioned. 83 Table 4.8: Nonylphenol and related alkylphenols found in fish food pellets and diet components (duplicate analyses except for the 2000mg 4-NP/kg sample), (n/d = none detected). Sample (Set A) Blank 0.0 mg/kg 4-NP 0.002 mg/kg 4-NP 0.2 mg/kg 4-NP 20 mg/kg 4-NP 20(dup) mg/kg 4-NP 2000 mg/kg 4-NP E 2 31 mg/kg Krill Marine Oil Sample Size (g) 0.54 0.553 0.481 0.543 0.121 0.124 0.110 0.498 0.194 0.581 Compound DP (mg/kg) 0.2 N/d 0.2 0.3 0.8 2.1 71.0 0.3 n/d 0.2 HP (mg/kg) n/d N/d n/d N/d n/d n/d n/d n/d n/d n/d OP (mg/kg) 0.2 0.3 0.4 0.4 1.1 1.1 80.0 1.5 0.6 0.7 NP (mg/kg) 0.06 1.09 2.00 1.76 12.85 43.99 1355 5.01 4.52 0.34 Sample (Set B) Blank 0.0 mg/kg 4-NP 0.002 mg/kg 4-NP 0.2 mg/kg 4-NP 20 mg/kg 4-NP 20(dup) mg/kg 4-NP 2000 mg/kg 4-NP 31 mg/kg Krill Marine Oil Sample Size (g) 1.0 1.07 1.01 1.03 0.121 0.124 - 0.498 0.94 1.03 Compound DP (mg/kg) 0.1 N/d 0.1 0.1 0.6 1.8 - 0.3 0.2 0.1 HP (mg/kg) n/d N/d n/d N/d n/d n/d - n/d n/d n/d OP (mg/kg) 0.1 0.3 0.4 0.2 0.7 6.0 - 2.5 0.2 0.6 NP (mg/kg) 0.07 1.26 2.26 1.44 12.51 42.42 - 5.18 5.31 0.64 Table 4.9 illustrates the difference that was found between sampling tissue very near the undersurface of the fish skin and in the deep with dorsal muscle tissue. Early in the course of sample preparation for 4-NP determinations of the fish tissue, significant differences in concentration results were observed in tissues sampled immediately below the dermal layer and those sampled from more internal fish tissue. Of the two analysts who collected fish tissue samples for extraction, one analyst dissected all tissue immediately beneath the skin surface of 84 the dorsal region of the fish, whereas the other analyst dissected tissue closer to the internal organs of the fish and selected tissue from at least 1-2 mm under the skin layer. The extraction of tissue samples from immediately beneath the skin surface of the fish yielded considerably higher concentrations of 4-NP. In order to minimize variations in surface concentration effects, the same analyst collected all tissue samples in a similar fashion from the dorsal section of the fish although using tissue more proximal to the internal organs of the fish than the skin layer and these were used for the reported results. Future experiments must consider that the lipid layer immediately beneath the skin of the fish may accumulate different concentrations of 4-NP than internal tissue and therefore sampling consistency (or lipid normalization of the muscle tissues) is necessary. Table 4.9: Concentration of 4-NP just underneath the surface of the skin and deep within the dorsal muscle tissue (n/d = none detected). Sample Location Treatment Underneath Skin Internal Tissue 223 - 2000 mg 4-NP/kg diet 3593 -2 2 4 - E 2 - 12.8 225 - 0.2 mg 4-NP/kg diet 424 20.5 226 - 20 mg 4-NP/kg diet 485 15.9 227 - 0.002 mg 4-NP/kg diet - 19.7 228 - control 102.8 -229 - 2000 mg 4-NP/kg diet 3437 526 230 - 20 mg 4-NP/kg diet 484 n/d 231 - 0.2 mg 4-NP/kg diet 347 25.1 232 - 0.002 mg 4-NP/kg diet - 16.7 2 3 3 - E 2 - 11.3 234 - control 115 -Table 4.10 provides the analytical results for 4-NP in fish dorsal muscle tissue, livers and gall bladders at the conclusion of the fresh water exposure period. At the end of the fresh water phase, only liver tissue had detectable concentrations of the contaminant decylphenol (DP) and these values were relatively close to the concentration determined in the method blank sample. 85 Heptylphenol (HP) was not detected in any of the samples (including the method blank). Octylphenol (OC) was detected in fish muscle of the 20 mg/kg and 2000 mg/kg 4-NP treatments, in all treatment livers samples and in the gall bladders of the 0.002 mg/kg, 20 mg/kg, 2000 mg/kg of 4-NP and in the E2 treated fish at the end of the fresh water phase. 4-NP was detected in all samples of fish muscle, all liver samples and the gall bladders of the 20 mg/kg and 2000 mg/kg 4-NP treatment groups. The concentrations of 4-NP were higher in liver and gall bladder samples than in fish muscle samples. Table 4.11 provides the results for the same parameters at the conclusion of the growth phase in seawater. Observable concentrations of 4-NP were only detected in the livers of the 0.2 mg/kg and 2000 mg/kg 4-NP treatment samples despite evidence of a low concentration of 4-NP in the basal diet that was fed to all fish until the termination of the experiment. The values reported for these 4-NP concentrations described the presence of small chromatogram peaks that were only slightly above the detection limit concentration. Octylphenol was detected in all 4-NP treatment samples with the exception of the control and the 2000 mg/kg 4-NP gall bladder samples. Estrogen (E2) samples were not analysed at the conclusion of the sea water growth phase and this is described in Table 4.11 as n/a. From the analyses of the tissue concentrations of 4-NP at the conclusion of the freshwater and sea water phases, it is apparent that only trace amounts of 4-NP were detected in the livers of the 0.2 mg/kg and 2000 mg/kg 4-NP treatments after the sea water growth phase, thus suggests that the bilary fecal pathway effectively eliminated 4-NP from all groups by the end of the sea water growth phase. 86 Table 4.10: Concentration of 4-NP in fish tissue and organs (muscle, liver and gall bladder) at the end of the fresh water phase of the study in relation to diet treatment (n/d = none detected). Treatment Parameter Blank Control 0.0 0.002 0.20 20 2000 mg/kg 4-NP mg/kg 4-NP mg/kg 4-NP mg/kg 4-NP mg/kg 4-NP M usele Tissue Sample 10.0 10.3 10.83 10.4 10.5 10.4 10.3 size (g) 10.4 10.76 10.56 10.6 1 Compound DP (ng/g)1' n/d n/d n/d n/d n/d n/d n/d HP (ng/g)2/ n/d n/d n/d n/d n/d n/d n/d OP(ng/g)3/ n/d n/d n/d n/d 35.7 33.8 n/d 14.5 NP (ng/g)4/ n/d 102.8 19.7 25.1 15.9 526 12.8 115 16.7 20.5 11.3 Li\ cr Sample 3.0 2.5 3.2 2.7 2.17 2.10 3.06 size (g) 2.72 2.86 2.7 2.82 2.92 3.75 Compound DP (ng/g) 12 15.5 16.2 13.9 22.4 22.2 9.2 n/d n/d n/d 24.4 47.8 n/d HP (ng/g) n/d n/d n/d n/d n/d n/d n/d n/d n/d n/d n/d n/d n/d OP (ng/g) 10.1 26.9 55.9 158.8 124.7 85.4 43.8 30.0 22.3 30 71.0 117.2 14.2 NP (ng/g) 22.6 36.6 26.9 47.0 67.1 1183 46.6 32.3 28.7 32.1 65.7 1251 37.9 Call Bladder Sample 0.30 0.402 0.455 0.350 0.458 0.339 0.37 size (g) 1 Compound DP (ng/g) n/d n/d n/d n/d n/d n/d n/d HP (ng/g) n/d n/d n/d n/d n/d n/d n/d OP (ng/g) n/d n/d 118 266 227 770 162 NP (ng/g) n/d n/d n/d n/d 395 71475 n/d DP= decylphenol;11 HP = heptylphenol;3 / OP = octylphenol; NP = nonylphenol 87 Table 4.11: Concentration of 4-NP in fish tissue and organs (muscle, liver and gall bladder) at the end of the sea water phase of the experiment, (n/d = none detected). A l l fish were fed the control diet after sea water transfer. Treatment Parameter Blank Control 0.0 mg/kg 4-NP 0.002 mg/kg 4-NP 0.20 mg/kg 4-NP 20 mg/kg 4-NP 2000 mg/kg 4-NP Sample size (g) 10.0 10.4 Muscle Tissue 10.4 10.5 10.2 10.3 n/a Compound DP (ng/g)" HP (ng/g)2/ OP (ng/g)3/ NP(ng/g)4/ n/d n/d n/d n/d n/d n/d 5.8 n/d n/d n/d 31.3 n/d n/d n/d 8.2 n/d n/d n/d 18.1 n/d n/d n/d 43 n/d Liver Sample size (g) 3.0 5.9 6.13 5.12 5.53 4.86 n/a Compound DP (ng/g) HP (ng/g) OP (ng/g) NP (ng/g) n/d n/d 10 n/d n/d n/d 20.2 n/d n/d n/d 21.3 n/d n/d n/d 31.8 6.7 n/d n/d 30 n/d n/d n/d 16.7 7.7 Gall Bladder Sample size (g) 0.80 1.091 0.759 0.847 1.454 0.714 n/a Compound DP (ng/g) HP (ng/g) OP (ng/g) NP(ng/g) n/d n/d n/d n/d n/d n/d n/d n/d n/d n/d 35.5 n/d n/d n/d 31.3 n/d n/d n/d 61.7 n/d n/d n/d n/d n/d 1 7 DP= decylphenol; 21 HP - heptylphenol;3 / OP = octylphenol; NP = nonylphenol To determine whether there would be a difference in the concentration of 4-NP with the depth of water column of the fish tanks, two separate samples were taken before and one after siphoning of the tanks to remove accumulated organic matter in the highest 4-NP diet treatment tanks on three separate occasions. In the normal daily maintenance of the fish tanks, a siphon suction hose was used to clean the bottom of the fish tanks. Each tank is equipped with a special sample drain which is separate from the major tank drain. Each sample labeled "drain" was collected 88 from the sampling drain. The samples labeled "scoop" were collected from the tank water column at about one third of the total water depth beneath the surface of the water. The idea was to differentiate between concentrations near the bottom of the tank (and the organic layer) and concentrations higher up in the water column which could be at a different concentration. Water concentrations of 4-NP are provided in Table 12. One sample of Cypress Creek water was spiked with 10 mg of 4-NP in 4L to assess recovery from the time of sample collection to determination of the water 4-NP concentration. In this process, 10 mg 4-NP was dissolved in 25 ml of 95% ethanol and then mixed by sonication for three minutes before transfer to a clean 4L brown bottle. The bottle was then filled with Cypress Creek water with care not to overfill the bottle. The recovery of 4-NP in this sample was 83.1%. Discrepancy from higher recovery concentrations could be due to 4-NP remaining adsorbed to the inner surface of the sample bottle or adsorption to any organic particulate matter suspended in the sample water. The concentration of 4-NP determined in clean unspiked Cypress Creek water was detected to be 43 ng/L although the concentration value assigned to the blank was 40 ng/L. It would be impossible to distinguish between the concentration difference given by the sample signal to that given by the background noise, thus the concentration of 4-NP in the Cypress Creek water was probably negligible. Water samples were analysed in two runs and therefore Table 4.12 identifies the blank value that corresponded to each series of analyses. The June 20, 2000 (2000 mg/kg 4-NP) scoop was taken before the drain sample was collected and was run in duplicate as required by standard QA/QC protocol for the analyses. The reproducibility of the duplicate was good. A lower concentration of 4-NP was found in the scooped samples compared to those samples collected from the drain. Considerably higher concentrations of 4-NP were determined after siphoning of the tank. The concentration may have been higher after siphoning the tanks if physical disturbance of the water by the action of vacuuming dislodged 4-NP that would have otherwise absorbed to the sides of the tanks and the fish. Further, the action of siphoning tanks increased fish activity due to avoidance of the siphon tube and disturbance of the organic layer at the bottom of the tank may have re-suspended particulate matter. The 4-NP could then adsorb to the re-suspended organic fraction. 89 Only the highest concentration tank, 2000 mg 4-NP/kg provided detectable concentrations of 4-NP in the water column on the sampling date. Concentrations of 20 mg 4-NP/kg 4-NP and below did not yield measurable concentrations of 4-NP in the water column. Table 4.12: Determination of 4-NP in water samples (n/d = none detected). Water Sample Analyses 4-NP, ng/ L Sample # Blank Ng/L Cypress Creek control Cypress Creek Spiked 4 NP 10 mg in 4 L Before Siphoning 2000 mg 4 NP/ kg Drain (June 7/2000 ) 2000 mg 4NP/ kg Drain (June 8/ 2000) 2000mg 4NP /kg Drain (June 9/2000) 2000 mg 4NP /kg Drain (June 10/2000) 2000 mg 4NP /kg Drain (June 11/2000) 2000 mg 4NP /kg Drain (June 13/2000) 2000 mg 4NP /kg Drain (June 14/2000) 2000 mg 4NP /kg Drain (June 15/2000) Sampling Day Analyses 0.002 mg 4NP/ kg diet (June 20/2000) 0.2 mg 4NP/ kg diet (June 20/ 2000) 20 mg 4 NP/ kg diet (June 20/2000) 2000 mg 4 NP/ kg diet (June 20/2000) Water Column Profile 2000 mg 4 NP/ kg diet Scoop before siphoning (June 7/2000) 2000 mg 4 NP/ kg diet Scoop before siphoning (June 20/2000) 2000 mg 4 NP/ kg diet Drain after siphoning (June 7/2000) 2000 mg 4 NP/ kg diet Drain after siphoning (June 15/ 2000) 2000 mg 4 NP / kg diet Drain after siphoning (June 20/ 2000) 40 207,750 n/d 1151 5178 1339 770 1145 1512 n/d n/d n/d 994 261 437 242 2064 2183 421 6 7 16 17 2 3 4 13 14 15 18 9 11 8 12 10 43 43 n/d n/d n/d n/d 43 43 43 n/d n/d n/d n/d n/d n/d n/d n/d n/d The intent of sampling the water from the tank where the highest diet concentration of 4-NP was dispensed to the fish each day for one week was to establish the approximate time at which an equilibrium steady state was reached in the experimental tank system. Extreme variability was observed in the water concentration profile Of 4-NP over the seven days. This was consistent with variability of turbidity of the Cypress Creek water. Relatively heavy rainfall events increased the concentration of suspended solids in the Cypress Creek water supply. June 9 was a day of fairly high precipitation and on June 11 the turbidity of the water in the tanks was so high that it prevented routine siphoning of tanks since the fish were not visible. It is possible that the 90 organic particle fraction and 4-NP were not uniformly distributed in the water column. This heterogeneity of distribution would yield non-uniform absorption of 4-NP throughout the water column and collection of the sample through the drain would not be consistent from day to day. Ideally, future experiments should use filtered fresh water to minimize the possible effects of 4-NP adsorbing onto the suspended organic particles in the water column. 91 DISCUSSION OF RESULTS " ...the trade of clothing facts in words is bound by its very nature to fail. " Primo Levi The facilities and access to specialized knowledge that is available at the Department of Fisheries and Oceans West Vancouver Laboratory provided a unique opportunity to collect research information and data. The indoor aquaria and vast expertise in fish husbandry allowed the bioassay to be conducted under controlled experimental conditions and with minimal unexpected complications during the experiment. 5.1 General Observations of Fish Feeding and Behaviour In the course of the freshwater exposure phase of the experiment, there were several days where rainfall events made the flow through Cypress Creek water very turbid. High suspended solid concentrations in all tanks made direct fish observation impossible and turbidity prevented accurate counts of wasted diet pellets to be counted. The physical properties of 4-nonylphenol suggest that high suspended organic particle concentration in the water column of the tanks will effect the actual exposure of the fish to of the contaminant. 4-nonylphenol tends to associate with organic matter or sediments. The experimental fish quickly adopted a feeding pattern that ensured that the treatment diets were, indeed, being ingested. During the first days of feeding the fish, there appeared to be excessive waste of diet pellets although slightly turbid water made accurate waste estimates difficult. To ensure that the fish began to accept the treatment diets early in the 28-day freshwater phase, it was decided (based on past experience of indoor aquarium experiments with coho salmon) to coat the diet pellets with krill. This was effective in improving palatability and the fish began feeding to satiation early in the experiment. This coating procedure was not discontinued once the fish began to consume their diet but rather all treated and untreated diet pellets were coated prior to feeding the fish. There were no overtly obvious block effects 92 observed during the feeding of the fish. As a general rule, the fish in all treatment tanks did not immediately regain their appetites after the weighing and sampling days but they resumed their usual feeding behaviours within one or two days post sampling. Over the course of the experiment, there were only occasional fish mortalities in each treatment tank and percent survival ranged between 98 and 100% during the fresh water phase of the experiment. Some fish in various treatment tanks died on days after sampling but these numbers did not exceed the typical mortality due to the stress of handling during the sampling process. During the seawater growth phase of the experiment, it was possible to accurately assess feed waste since this water supply was filtered and thus free of suspended organic matter. Also, there were a few occasions of fish mortality and these fish often displayed excessively distended abdomens. The pattern of fish mortality occurred randomly. Dissection of the dead fish did not reveal abnormalities within the fish with the exception that the stomach cavity was very full. There were no evidence of fish disease in the dead fish. Hematocrit is one of the indicators to assess tolerance limits of fish to biotic or abiotic stress factors. The hematocrit values for all fish fell within the range of 36-42 %. The normal range of hematocrit for coho salmon is between 32-52 (Schreck and Moyle, 1990). 5.2 Observations of fish growth Over the course of the experiment during both the fresh water and sea water phases of the study, normal growth distribution of fish in each of the treatments was observed. This observation contrasts the results of Brown et al. 1 who found bimodal growth distribution in Atlantic salmon after they were fed diets supplemented with 10 pg/g and 50 pg/g 4-NP immediately prior to parr-smolt transformation. During their experiment, however, a fixed ration of the daily diet with 4-NP was fed to 100 experimental fish per day. Throughout the present experiment, the fish were fed to satiation with diets containing up to 2000 mg/kg of supplemental 4-NP and the calculated dry feed intake was found to be 6 - 7 g of food per fish during the fresh water exposure phase. Brown et al. (2001) reported that water-borne exposure of juvenile Atlantic salmon to environmentally relevant concentrations of 4-NP and 17-p estradiol did not appear to influence fish mortality of the fish during sea water challenge immediately after exposure but subsequent 1 Dr. Scott Brown. Canadian Centre for Inland Waters. 2000. Personal communication. 93 growth in sea water was impaired by 30-40%. In the present experiment, treatment diets supplemented with as much as 2000 mg/kg of 4-NP did not influence the mortality of coho salmon immediately after exposure nor did their growth during the sea water phase appear to be affected by the 4-NP treatments given prior to sea water transfer. Qualitative observation of the 17-P estradiol (E2) treatment groups during the sea water phase in this experiment suggested a lower specific growth rate and wet weight gain although these lower values were not statistically significantly different from the block and treatment means. The exposure pathway of 4-NP was not confined to the dietary route only since appreciable concentrations of 4-NP were detected in the water column of the tank where fish received the 2000 mg/kg 4-NP diet treatment. Also, differences in the tissue concentrations of 4-NP which varied depending on the depths from the skin undersurface from which the sample was taken suggests that exposure to 4-NP could also be via diffusion processes through the skin. Despite three possible exposure pathways of the fish to 4-NP before their transfer to sea water, statistically significant reduction of coho salmon growth was not observed during the sea water growth phase. Geisy et al. (2000) used fathead minnows (Pimephales promelas) in their 42-day experiment to examine the effect of water borne 4-NP. The concentration range of 4-NP used in their experiments was 0.05 - 3.4 pg NP/L. The waterborne exposure pathway on the fathead minnows produced no significant acute lethality and hematocrit measurements suggested that all fish were not suffering from general stress in the range of 4-NP concentrations to which they were exposed. Their experiments revealed that although plasma estrogen (E2) concentrations were not affected by exposure of the fish to high and very low doses of waterborne 4-NP, they were, however, affected by the mid-range concentrations (shown by an inverted U-shaped dose -response curve). They concluded that overall reduction of egg production in the fathead minnow was not likely the result of 4-NP acting as an estrogen-receptor antagonist. Our experimental findings using coho salmon also demonstrated no significant acute lethality of 4-NP at dietary concentration ranges where the measured attendant concentrations of 4-NP in the water column were comparable to the waterborne concentration range tested by Geisy et al. (2000). The hematocrit values of the experimental fish suggested that they were not suffering from general stress in the concentration range of 4-NP to which they were exposed even though they were exposed to 4-NP by three different routes, i.e. the diet, water (possibly across the gills) 94 and adsorption through the skin surface. Based on the determination of the 4-NP concentration of the highest diet dose (2000 mg/kg 4-NP), the water borne concentration range to which my experimental fish were exposed to was probably between 0.2-2.0 pg of 4-NP/L. It would be very interesting to learn i f plasma vitellogenin levels in the juvenile coho salmon in the present study followed the similar inverted U-shaped dose-response curve that Geisy et al. (2000) observed in fathead minnows. Madsen et al. (1997) found that 4-NP and 17-0 estradiol (E2) had qualitatively similar inhibitory effects on smoltification and the hypoosmoregulatory physiology of Atlantic salmon. In their experiment, 23 g salmon were subjected to interperitoneal injection of either 4-NP or E2 dissolved in peanut oil once per week for thirty days. The treatment endpoints they measured were plasma vitellogenin and osmosregulatory parameters of the fish at the time of sea water challenge. Unlike the Madsen et al. (1997) study, our results could not validiate inhibition of the process of smoltification in coho salmon based on the parameters measured in my experiment. At the time of sea water challenge in the present experiment, plasma sodium levels in coho salmon were not elevated and there was no suggestion that the osmoregulatory ability of the fish had been compromised by any of the diet treatments of 4-NP unlike the situation observed by Madsen et al. (1997) where 4-NP was given to Atlantic salmon by injection. In the present experiment, statistically significantly higher hepatosomatic indices were observed for the estrogen (E2) treated fish. This observation supports the effect also seen by Madsen et al. (1997) although, in their experiment, they observed elevated hepatosomatice index for both 4-NP treated fish (3 mg injected in 23 g fish) and the estrogen (E2) treated fish. The treatment dose of 17-(3 estradiol selected for our experiment was based on the dose injected (50 pg for 23 g fish) in the Madsen et al. (1997) study. The dietary dose range of 4-NP used in this experiment was intended to cover the concentration administered in the Madsen et al. (1997) experiment although analytical determination of 4-NP in my experiment demonstrated considerably higher doses than intended in the low 4-NP dietary treatments. Nevertheless, increased hepatosomatic indices were not observed in the 4-NP treatment groups an the end of the during the fresh water phase of the study. The mechanism of action of estrogen and other sex steroids on the hypoosmoregulatory system in salmonids is largely unknown. Although until vitellogenin activation can be confirmed by 95 assay results in this experiment, it is not possible to explain the increase in the hepatosomatic index of the fish treated with estrogen (E2) treatment. If, however, vitellogenin is found to be increased in the E2 fish in the present study, this would support the suggestion by Madsen et al. (1997) that increased vitellogenin and sex steroid activity in the liver is linked to larger livers in the estrogen (E2) treated salmonids. Although the combined exposure pathway of the fish to 4-NP in this study was very different from direct exposure of the fish to 4-NP through injection, it is possible that the effects noted for incresed liver size of the estrogen (E2) treated coho were similar to the responses observed by Madsen et al. (1997) in Atlantic salmon . 5.3 Effects of Estrogen (E2) Treatment on Growth It has been known for many years that estrogenic steroid hormones play an important role in cellular regulation in all vertebrate species (Zuckerman, 1940). Optimal growth of vertebrates requires a combination of estrogens and androgens (male hormones). Growth promoting effects are suspected tb originate from ability of estrogen/androgen combinations to increase the retention of dietary nitrogen through protein synthesis through several mechanisms (Davies & Danzo, 1981). Following the second world war, it was recognized that estrogens acting alone or in combination with androgens could enhance growth of cattle, sheep or poultry and this led to their use as a means of increasing meat production. This, in turn, led to the application of synthetic estrogen such as DES as a more cost effective growth enhancement tool for meat production (Schmidely, 1993). The growth promoting capabilities of estrogens have been known for some time (McMartin, 1978) but the mechanism by which growth promotion is accomplished and the accompanying risks to organism health are still not completely understood. Competing with the possibility of growth enhancement in fish is the observation that estrogen can significantly inhibit the hypoosmoregulatory mechanisms important to smoltification in fish (Madsen et al., 1997). Other researchers have observed the suppressive action of estrogen on plasma growth hormone (Bjornsson & Persson, 1992), somatotroph function (Miwa & Inui, 1986) and Cortisol production by interrenal cells of salmonids (Young, 1996). In my experiment, during the fresh water phase, estrogen (E2) treatment appeared to promote growth as evidenced by increased wet weight gain, specific growth rate, and feed efficiency (Table 4.5). The increase in protein efficiency ratio (statistically significantly higher value) and in percent protein deposited (not statistically significant) for fish given the estrogen treatment during the freshwater phase supports the observation of Davies and Danzo (1981) that there 96 likely is connection between plasma estrogen levels, growth and synthesis of proteins in vertebrates. During the sea water growth phase of the present study, the apparent inhibition of some of the performance parameters in the estrogen treatment groups (wet weight gain, specific growth rate, feed efficiency, protein efficiency ratio and percent protein deposited) appeared to support the inhibitory effects observed by Madsen et al. (1997). Measurement of plasma vitellogenin concentrations of the fish in this study during the sea water growth phase is required to verify the apparent inhibition and i f this is found to be the case, the present finding would agree with those of Madsen etal. (1997). 5.4 T 3 and T 4 Determination Figure 5.1 graphically represents how different hormone concentrations change temporally during the period that freshwater salmon undergo the process of smoltification. From the diagram, it is apparent that during the parr-smolt transformation phase a considerable rise in T 4 just precedes an elevation of plasma E2 concentration. The observation of statistically a higher plasma T 3 concentration (metabolic derivative of T 4 ) in the E2 treatment groups at the end of the freshwater phase may have been a consequence of the comparatively higher plasma E2 concentration in the fish fed the diet with estrogen (E2) relative to the concentrations of E2 present in the fish given the 4-NP and control diets. ,0 METABOLIC CHANGES, INITIATION OF SMOLTIFICATION AND GROWTH ACCELERATION SEAWATER TOLERANCE MOROPHOLOGICAL CHANGES , SILVERING , OUTMIGRATION (HOMING IMPRINTING?) PARR-* -> -4 -> -> -> -> SMOLT -> PARR - REVERTANT O I C A T E C H O L A M I N E S J F A M J J A Month Figure 5.1: Hormone changes during parr-smolt transformation (Hoar & Randall, 1988) 97 5.5 Determination of 4-NP in Samples Quantitative analytical determination of 4-NP in the treatment diets indicated that a broad dose range of 4-nonylphenol had been successfully incorporated into the feed pellets. The lowest dose diets (control, 0.002 mg/kg and 0.2 mg/kg 4-NP) contained between 1-2 mg 4-NP/kg. The estrogen (E2) treatment diet contained about 5 mg of 4-NP/kg in addition to the 31 mg of 17-(3 estradiol/kg with which it was dosed. The second experimental preparation of the 20 mg 4-NP/kg was found to contain approximately twice the desired dose. This latter diet was not treated twice in the course of preparation although the higher weight of 4-NP weighed into the marine oil which was used to incorporate it into the diet may not have been sufficiently distributed in the oil by sonication since the time was about the same for all treatment preparations (three minutes). Comparisons of the results between the theoretical doses of 4-NP and the analytically determined concentrations indicated that 4-NP was present in the concentration range of approximately 1 mg/kg, 2 mg/kg, 40 mg/kg and 1233 mg/kg of 4-NP diet on an air dry basis. It is possible that the higher doses of 4-NP may not have been homogeneously distributed into the mash during mixing of the basal diet ingredients prior to steam pelleting. It is also possible that the steam used in the pelleting process may have resulted in some degradation of 4-NP although this is unlikely since steam distillation is used in the analytical method that is employed for determination of alkylphenols including 4-NP. The analyses revealed that the krill preparation that was used to coat the diet pellets contained a measurable concentration of 4-NP. Also, the marine anchovy oil that was used to incorporate the 4-NP doses into the diet pellets and to administer the krill coating to the pellets contained measurable concentrations of 4-NP. Since measurable quantities of 4-NP were not expected in the diet components, there was effectively no untreated control in this experiment although the concentration of 4-NP in the control diet was low (~1.1 mg/kg) relative to the diets supplemented with the highest doses of 4-NP. There are many possible sources of 4-NP in both the marine oil and the krill including contamination from plastic shipping containers, cleaning products used in the industrial processing of the products and of the biological sources themselves. Based on calculations using 27 g/kg krill containing an average of 4.91 mg/kg 4-NP and using 129.7 g of marine oil containing an average of 0.49 mg/kg 4-NP, the basal diet contained a baseline concentration of 0.196 mg/kg of 4-NP from the preceding components before the addition of any 4-NP or E2 treatment doses. The levels of 4-NP in the marine protein sources (anchovy meal, 98 squid meal) were not assessed in this study but it is possible that these sources contributed some 4-NP to the basal diet as well. Decylphenol and octylphenol concentrations were measured during the analyses of all water, diet and tissue samples. Heptylphenol was not detected in any of the samples. Octylphenol was often found when decylphenol was also detected and since decylphenol cannot be a degradation product of 4-NP, it is possible that all three compounds were contaminants of the 4-NP mixture. At the end of the sea water phase octylphenol was measured in most of the tissues, livers and gall bladders of the 4-NP treatment samples even though only trace amounts of 4-NP could be detected in two of the liver samples of the 4-NP treatment groups. It is impossible to determine if the octylphenol that was detected was in fact a degradation product or a contaminant without further MS-MS analyses for octylphenol. 5.6 Concluding Remarks for Experimental Section The findings of this experiment suggest that dietary exposure of coho salmon to 4-NP for one month before their entry into sea water did not obviously compromise their osmoregulatory performance or their subsequent growth performance in sea water under the present experimental conditions. Significant differences in some growth and performance parameters were, however, seen in the estrogen positive control groups during the experiment phase when fish were being fed diets containing 17-(3 estradiol. For the estrogen (E2) treatment groups, wet weight gain, specific growth rate, feed efficiency and protein deposition rate were significantly higher than noted for fish given the other treatments during the fresh water exposure period although this trend appeared to be reversed (not significantly) in the sea water growth phase when the fish were fed the control diet. In considering the risk posed by exposure of juvenile coho salmon to 4-nonylphenol, information was learned regarding the importance of considering the physiological changes that accompany smoltification of the fish in risk characterization of exposure effects. The nature of the risk of exposure to 4-NP may be different for fish exposed in a freshwater environment than for those exposed to 4-NP in sea water. Juvenile salmonids do not drink water but excrete a considerable amount of dilute urine in fresh water whereas they drink water and excrete smaller volumes of concentrated urine when living in sea water. The process of smoltification may involve hypoosmoregulatory changes that permit elimination of contaminants from the fish following sea water transfer. It is clear that characterization of an environmental risk such as exposure of coho 99 salmon to 4-NP requires systematic examination of effects that take into account natural physiological processes of the fish, physical properties of the contaminant, background exposure levels, species-specific biological differences and cumulative effects of contaminants. The key to understanding the value of information gained is to seek new knowledge from a micro and a macro perspective. Scientific investigation is one framework for understanding new knowledge, however, a more holistic examination provides a different framework from which to assess risks. The next sections of this thesis wil l take a different perspective to look at the broader issues of the endocrine disruption controversy. 100 EXAMINING RISK-Within an Ecosystem Context Enlightenment thinkers know a little about everything. Specialists know a lot about a little. Post-Modernists doubt that we know anything E.O. Wilson. The concept of an ecosystem grew from the ecologists understanding of the natural world to reflect their view of human society and its relation to nature. Originally seen as a basic unit of nature, the metaphor of an ecosystem was described as a series of interconnected processes that could follow a predictable ecological succession to eventually reach a steady state. This complex systems approach was further refined by including consideration of feedback mechanisms competing to achieve dynamic equilibrium of processes in the model. The largely anthropocentric view of ecosystems prevails now as the tension between defining ecosystems as units that are free to be manipulated by humans or as understanding ecosystems as a basis for protection rather than control and these divergent viewpoints are seen in everyday life. Intimately connected to this debate is the discourse about the social construction of the natural world and the degree to which humans are detached from it. The social construct of an ecosystem places considerable emphasis on structures and energy flows rather than discreet geographic boundaries (Francis, 1995). The structures and processes operate over different spatial and temporal scales - some processes are extremely slow. The metaphor of ecosystem health finds its roots in medicine and public health focusing on the action and reaction to stress on the system. The various metaphors for describing ecosystems fundamentally assume that complex systems are best understood by disaggregating them into component parts to be addressed by the appropriate experts. This approach denies the interconnectedness of a more biocentric view of the world. Thus hierarchical perspectives are integral to understanding ecosystems as functional sub-systems that are but part of a larger 101 system. Ecosystems can be interpreted as holistic and hierarchical but at the same time operating under mechanical principles that humans can manipulate. Faced with an ever-increasing number of competing interactions, a more holistic ecosystem worldview allows for a degree of flexibility in collecting and interpreting the knowledge that gives humans insight into their relationship with the surrounding world and associated risk. Figure 6.1 describes a knowledge framework based on an ecosystem model to consider risk in which human beings are interconnected to biological and social aspects of their world. Precautionary Principle Action Figure 6.1: Ecosystem approach to considering risk. This model of influence interactions could be applied to many environmental or social issues but I would like to focus specifically on its application to examining risk associated with endocrine disrupting compounds. The diagram is intended to reflect the interconnectedness of all of the variables but at the same time highlights the differences in how the influences between these elements can be determined. Using a metaphor of arrows similar to those describing competing equilibria in chemical reactions allows the depiction of a system in which all influence factors are in constant interconnected flux. The model presupposes acceptance of a worldview that perceives humans as an interdependent part of the ecosystem. The darkest arrows between 102 natural ecosystem risk and human health risk are intended to reflect a similar approach to social construction of risk shared by individuals who accept this worldview. The sciencific method can provide measurable evidence to gain an understanding of the nature of a risk. The moderately dark arrows describe links between science or elements of the system for which quantitative information can be obtained and interpreted. Operating simultaneously, however, are social factors such as values, gender, culture, politics and socio-demographic factors that, in turn, affect perception and response to risk but are more difficult to quantitatively measure and trace interconnected influence. The lightest arrows illustrate only some of these interactions - in realty each of the social factors are connected to all elements of the framework depending on each individual's biographical situation that determines their perception of risk. The diagram is intended to illustrate the difficulty in incorporating social variables such as the influence of spirituality or power relationships into a framework for understanding risk. A crucial element for consideration in this framework is economics. Economics are intimately linked to all elements within this diagram, although the strength of the role in determining an action or policy can be interpreted in a multitude of ways. Cultural and political variables can have direct impact on the influence of economics on scientific assessment of risk and perception of risk. In some ways, science can operate as a model of a market economy. The commodity traded is knowledge and the metric of valuation can be money. Issues of public trust in science enter the equation for determination of perception of risk when research is funded by the private sector rather than the public sector. The stronger voice of the third sector or civil society in recent decades has had considerable influence in the recognition of social variables in managing environmental risk. Non-governmental organizations are now capable of funding research and transmitting knowledge about environmental risks. Cultural, political and economic interests can have direct influence on how non-governmental organizations portray environmental risks and this, in turn, has direct implications on the perception of risk in both the lay community and the political sphere. The consequences of manufactured risks accompanying technological development can have a pervasive influence on the environment. When faced with large-scale uncertainty of some risks, the precautionary principle forms a basis for erring on the side of caution when considering technologically related hazards. However, for many manufactured risks, the precautionary 103 principle plays more of a role in attenuating the perception of risk than it does in understanding the best management options of the risk. Understanding risk within the framework of an ecosystem depends on recognition that making technologies risk-free may be an unattainable objective although determining a scientifically and socially acceptable alternate that is "safe enough" to protect human and environmental health may be viable. The flow of knowledge towards "understanding of risk" on the left side of Figure 6.1 stresses the role of science in risk assessment. Distinct prescriptive steps to ecological and human health risk assessment build a substantial body of knowledge from which probabilistic estimates of risk can be assigned. The right side of the diagram suggests that perception of risk can drive policy or response at a personal level to an environmental risk with a degree of independence of the scientific understanding of the risk. Social constructs deeply rooted in knowledge that is connected to perception of risk and economics plays are significant role in all aspects of this theoretical framework. Risk perception, is, in fact diffused throughout the various identified elements within the system although the spectrum of influence on decision-making and response is variable. Implicit to the argument supporting this ecosystem-based approach to the examination of risk is that communication based on scientific facts flows freely throughout the system but for the most part, this is not the case. Selecting the risk management philosophies that are most compatible with society depends on developing the strongest scientific basis for estimating risk and simultaneously including perception of risk in the process. The first half of this thesis has been dedicated to demonstration of application of the scientific method to gain a better understanding of the risk associated with dietary exposure of salmon to an endocrine disrupting contaminant. With the underlying goal to augment existing information that supports the risk management strategies for 4-nonylphenol, the experiment makes a small contribution to the store of scientific knowledge. It is equally important to gain information regarding i f and how perceptions of risk linked to endocrine disruption are unique and what implication of this perception of risk within influence groups may have on policy directions. The endocrine disruption hypothesis provides an interesting opportunity to examine some of the social constructs born from values. Unique to this particular class of environmental contaminants is a direct inference that observed exposure effects could be interpreted as posing risk to human fertility and sexuality. The media repeatedly exploits this factor despite the lack of legitimizing scientific evidence to suggest this is so. Scientific and political interest concerning 104 the risk of endocrine disrupting contaminants in the environment has grown at an unprecedented rate over the past five years although public concern about this risk appears to be limited. In Canada, as in several European Union nations, the United States and Japan, policy has been implemented to manage the perceived risk of endocrine disrupting compounds including the declaration of 4 -nonylphenol as being toxic. For these reasons, examining some of the social considerations and the role of risk communication linked to exposure to endocrine disrupting contaminants within the risk society framework is valuable. 105 7 THE EDC CONTROVERSY Literature and life are both in equal measure instruments of research and therefore truth. Carlo Bo We live in the modern world. The evolutionary process of modernization influences our perceptions, our fundamental philosophical truths and our happiness. This process depends on social change and technological developments, some of which have occurred at an unprecedented rate over the past three decades. Yet critically examining our process of becoming modern reveals both seriously negative consequences and highly beneficial results. At the same time, social and political changes that result from the process of modernization have reduced public trust in institutions of government as well as bureaucratic and scientific expertise in an increasingly uncertain world. The ecosystem approach to examining risk as illustrated in Figure 6.1 can be further refined to describe the influences of key interest groups related to the endocrine disruption controversy. The interconnectedness of the various groups in this diagram relies heavily on risk communication for the transmission of knowledge that shapes perceptions of risk within each individual group. The perceptions of risk both between and within these groups are, in some cases, vastly different and thus the strength of their influence in adopting a policy to address the risk is variable. Since these perceptions are also shaped by less obvious influences of gender, culture, values and politics, it is difficult to assess the effect that each key group has on policy decisions. Human health experts are an example of a heterogeneous interest group where the discernible difference between physicians who deal with symptoms of illnesses and physicians who are possibly more connected to knowledge linked to causes and effects of illness may lead to very different perceptions of risk within one group. Examining the perception of risk that is held within one of these influence groups would be useful to our understanding of how policy and voluntary action address the potential hazard posed by exposure of humans to endocrine disrupting compounds. 106 Figure 7.1: Influence diagram, of key interest groups in the endocrine disruption controversy For the purpose of this analysis of the EDC controversy, the history of one particular event can serve as a useful point of reference for understanding the nuances of sociological examination of risk, scientific assessment of the risk, the perception of risk, stigmatization of risk and the role of risk communication. The influence of the associations of actors representing the interest groups described in figure 7.1 can be traced in the context of this event. In the second week of December of 1998, the technical superintendent, the director of sales and the environmental coordinator of a Vancouver Island, BC pulp mill met to discuss a potentially major crisis. The director of sales had received a fax with their weekly sales documents from a large Tokyo newspaper that was a long established client and purchased large quantities of newsprint on a weekly basis. In the fax, the newsprint customer expressed concern that their company was planning to adopt a policy of not using products for which environmental effects 107 could be linked to endocrine disruption. Since company management wanted to publicize that the production of their newspaper had no connection to pollutants that could be regarded as endocrine disrupters, the purchasing department regretted to inform the BC newsprint supplier that i f "serious concentrations" of EDC compounds were involved in the production of the newsprint, they would be required to switch to a newsprint supplier who guaranteed their production process did not pose this risk. The fax identified the EDC compounds that were of concern to the company management in an attached list. The attached list contained approximately 67 compounds among which were dioxin, nonylphenol, bisphenol A and several pesticides. There was no definition of what was meant by serious concentrations. The director of sales at the BC mill called a meeting with the two key technical personnel to learn what was so serious about endocrine disrupting compounds that would drive a long-term client to consider replacing their newsprint supplier. Following the meeting, the environmental coordinator telephoned me at my former employer to seek information from me, as a toxicologist, to provide information on the preceding topic to senior mill sales and technical staff in response to their potential crisis. I told the environmental coordinator that I could provide some background information to him since I had followed the rise of the endocrine disruption concern for nearly two years and I had gathered many references linked to environmental contaminants that could act as endocrine disrupters. Past technical meetings of various pulp and paper industry associations had introduced the concern of EDCs related to pulp and paper effluents and the environmental coordinator had known of EDCs for several years but had never been faced with the pressure of a major client proposing to switch suppliers. The environmental coordinator faxed to me the list of contaminants of concern that their customer had forwarded to the mill. I recognized this list as being a summary or extraction of the US EPA general list of chemicals that would be subjected to Tier II screening for endocrine disruption potential. The compounds listed were not specific to pulp and paper processing although many of the compounds listed could potentially be found somewhere in pulp and paper processing. I provided the information to the mill's environmental coordinator and emphasized that much uncertainty was attached to the current state of the science examining pollutants that acted as endocrine disruptor and considerable research had been undertaken world-wide. I provided the environmental coordinator with the information that the compounds in the faxed list were currently being screened by the US EPA. Also I mentioned that Canadian research was in 108 progress to look at priority substances that were also listed on the US EPA list and indicated that no policies to regulate contaminants on the basis of being endocrine disruptors had been adopted by the US or Canada. Further, I mentioned that was aware of voluntary initiatives on the part of some European countries to reduce or eventually eliminate some compounds that were suspected as being endocrine disruptors (I did not know of the policy position of Japan), and that endocrine disrupting sublethal effects related to reproduction of fish were included in the Canadian policy to eliminate dioxin in pulp mill effluents by the year 2000. The environmental coordinator stressed the lack of transmission of current knowledge from scientific experts to the pulp and paper industry in order to provide enough background information to deal with environmental concerns of clients with respect to endocrine disrupting compounds. I proposed to hold a workshop designed specifically to provide Canadian expert knowledge to pulp and paper industry representatives who could be faced with the same challenge that this mill faced. The environmental coordinator supplied the background information to key mill staff involved with sales of paper products from which a statement could be prepared that defined the company's approach to addressing the risk of EDC exposure. This statement was sent to the Japanese newsprint client whose company had expressed serious concerns. The Japanese client agreed to continue their relationship with the BC newsprint supplier under the provision that updates regarding developments regarding endocrine disruption be provided to the Japanese newspaper as the B C mill addressed the particular environmental concern. During the planning process of the workshop that I proposed to hold in May 1999, I contacted scientific authorities from both the US and Canadian pulp and paper associations to invite them to speak about current understanding of the risk posed by endocrine disrupting contaminants found in mill effluent. In January, the representative from the US pulp and paper association declined my invitation to speak at the workshop since his superiors preferred not to have a national industry representative speaking publicly about EDCs and therefore no perceived stand on the part of industry could be implied. In January, the scientific representative of the Canadian pulp and paper industry association accepted the invitation to speak on the subject of current research investigating the risk associated to EDCs linked to pulp and paper processing. Several weeks later, this scientific representative rescinded his acceptance of the invitation to speak at the workshop since the endocrine disruption issues were described being as "too controversial" to address at this point. 109 The workshop was held in May 1999 and had an attendance of 87 delegates. Twenty of the twenty-two British Columbia pulp mills and two Alberta pulp mills were represented. Two scientific experts from the Canadian federal government, who chair or participate in some of the global research initiatives to investigate the effects of endocrine disruptors, presented papers at the meeting. Three university professors presented current research issues related to EDCs. Two presentations were given by members of the federal government conducting research in local British Columbia watersheds related to investigations concerning endocrine disrupting contaminants. Dr. Louis Guillette, Jr. gave the plenary presentation in which he outlined the past decade's development of the endocrine disruption hypothesis since the observations that he had first observed in Florida alligators. Following the workshop, when Dr. Guillette described his hectic speaking engagement schedule of the months prior, he said he had been asked to speak at the meeting sponsored by the Environment Agency, Government of Japan. Dr Guillette was among the list of speakers at the meeting held in Kyoto on December 11-13, 1998 that included most of the key US and Japanese researchers whose investigations were linked to endocrine disruption. Entitled "International Symposium on Environmental Endocrine Disruptors '98", among the nearly 1000 delegates were Theo Colburn, many of the key scientists from Britain and the European Union, representatives of several Japanese universities, representatives from industry and several members of the Japanese government from ministries linked to human health and the environment. The greeting from the opening of the meeting by Kenji Manabe, Minister of State, Director-General of the Environment Agency stated: "....The problem of endocrine disruptors (EDs), more commonly known in Japan as environmental hormones, is one of the most important issues confronting the world today. National governments, industry and researchers are now busy gathering scientific information on a problem so serious it could well threaten the very survival of mankind itself. "2 This international symposium held in Kyoto attracted the attention of most of the Japanese press. The national news broadcast both information from the meeting presentations and events linked to the meeting. A demonstration against the US plastics industry regarding the risks posed by bisphenol A and phthalate components of plastic products was publicized. In the two weeks of 2 Kenji Manabe. 1998. International Symposium on Environmental Endocrine Disruptors '98. Agency, Government of Japan. 110 Environment December 1998 immediately following the symposium, several articles were reported in the Japanese press about the risks associated with endocrine disruption.3 The response to this meeting is very likely the signal event that spread the concern regarding endocrine disrupting compounds of the Japanese newspaper to a BC pulp mill. The event described above provides a historical example of how the influences of key interest groups illustrated in figure 6.2 played out. Our modern technology-driven world allows experts from all parts of the globe to assemble easily to share information, technology facilitates rapid flow of information and the communication of risks that accompany this modernization linked to technology evolution cause considerable concern to society. In this case, the private sector, as illustrated by the Japanese newspaper and the BC pulp mill lacked information. As a toxicologist but not an expert, I had fundamental understanding of a small part of the science supporting the endocrine disrupting hypothesis but no information regarding policy positions of the Canadian government, let alone other world governments. Several governments were conducting considerable scientific investigation and Canadian experts were, indeed, part of the global endocrine disruption picture but communication of this fact was limited. The lay community, NGOs or both were aware of the risks associated with endocrine disrupting compounds present in plastics. Media response in the press may have created or promoted an existing controversy linked to EDCs. The statement on the part of Mr. Manabe, as a senior government official in a decision-making role was definitely perceived as alarming. The understanding of risk of each of the groups described in Figure 6.2 is influenced in different fashions by worldview, notions of technological stigma, perception of risk and risk communication. Assessing the interaction of these influences in an integrated manner can shed light on the understanding of the action and policy that results from response to a perceived risk. Before the elements of a controversy can be analyzed with any rigor, it is necessary to develop an understanding of the sociological framework used to make such an examination. This leads to some very fundamental philosophical questions whose answers are deeply rooted in individual perceptions and probably often unanswerable. How do we describe what is modern? In the process of becoming modern, how do we deal with the negativities and uncertainties that accompany social and technological development? How do we make sense of the "facts" and what, indeed, is a "fact"? How do we respond to situations where uncertainty is great and our 3 Dr. Louis Guillette. 1998. Professor. University of Florida, Gainesville. Personal Communication. I l l knowledge of potential impacts is limited? How do we marry the subjectivity of individual perceptions with the positivist contribution to knowledge? Challenging questions such as these build the stage for examining the social experience of risk and the dynamic processing of risk of various participants in modern society. Anthony Giddens (1991) and Ulrich Beck (1992) share the premise that in the process of modernization, society has transformed from what could be described as class society to "risk society". Their concept that describes the modern world is referred to as "modernity". Intimately connected to this notion of modernity is accompanying risk and, in a pluralistic society, identifying, evaluating and managing risk is shaped by our individual perceptions. The social change and technological developments that advance knowledge are paralleled by amplification of the associated uncertainty in this knowledge, thus, risk could be thought of as being "manufactured". Risk Society Social constructivists such as Giddens and Beck use "modernity" as the shorthand term to describe modern society or industrialized civilization. A central premise of modernity is that modernization by development of technology could be thought of as bringing the end of nature and the end of tradition. There is no part of the planet that has completely escaped any impact of technological advance. The end of nature does not mean that the world it describes has disappeared but it implies that a transformation has taken place. Similarly, traditions persist throughout our many cultures but globalization and reflexivity have contributed to the modification or, in some cases, destruction of some traditions. Modern society has seen an unstoppable convergent or globalizing trend in the way countries manage their political affairs, education systems or family lives. The reflexive nature of modernity simply means that nearly all aspects of social activities and structure are subject to "chronic revision in the light of new information or knowledge" (Giddens, 1991). Bruno Latour argued that we live in a hybrid world in which clear distinction between nature and culture has been lost. (Latour, 1993) The persistent idea that nature and culture can somehow exist without each other fails to recognize that we build, act and live in an artificial civilization whose characteristics change the porosity of the boundary between nature and culture. Risks could be thought of as man-made hybrids as well. This type of hybridization could be justified within a postmodern framework. In order to gain some sort of understanding of the culture and 112 politics of a risk society, it is important to recognize that politics, ethics, science, technology, communication and perceptions have shifting senses of realities. That is not to say that there is no reality, as a true postmodernist would hasten to point out, but these realities are perceived differently. Two key trigger events of technological development of the twentieth century accelerated the evolutionary process of modernization. The Soviet Union ushered in a new era with the launch of the first artificial earth satellite, Sputnik, on October 4, 1957. This grand technological leap created a sense of paranoia and self-doubt, especially in the United States, as the Soviets released information about the project in limited bursts leaving much room for speculation but simultaneously conveying a sense of mystery and wonder. Paralleling the birth of the race towards "the end of nature" was the genesis of uncertainty and recognition of risk accompanying technological developments as the human mind began an unspoken competition with nature. Human preoccupation with the future and the process of becoming modern drove the quest for "facts" and the processing of knowledge into technology development at a rate never before seen in history. The marriage between science and technology bred a public perception that scientists could elucidate the facts to address concerns associated with the overwhelming exponential growth of technical capability given enough time. However, the rate at which some of the accompanying concerns evolved grew exponentially as well and recognition of uncertainty began a gradual process of undermining absolute trusts in science. The pivotal event that signaled the figurative end of tradition and caused dramatic social change was, in fact, a scientific accident. Russell E. Marker was an organic chemist keen to study the new family of medicines called steroids in the late 1950's. His research provided the clues that the organic chemist, Carl Djerassi, used to create the next essential step to synthesize a chemical that could prevent pregnancy. A l l of the events that led to the discovery of the Pill were a series of coincidences in which the goals were relatively small but accompanied by considerable uncertainty. Never in the beginning of his research, did Djerassi dream of contraception as a goal of his work. The 1960's marked the beginning of social revolution that resulted from the discovery of the Pil l and effectively changed the course of history as many traditions coupled with class, sex, gender, family, institutions and governments altered or dissolved. The notion of risk society can be used to better understand why modernity has exerted such a powerful paradigm shift in society. Ulrich Beck argues that risk society is a stage of modernity -113 what he calls second reflexive modernity (Beck, 1992). Anthony Giddens describes risk society as the world that lives after the end of nature and the end of tradition (Giddens, 1991). Risk society is marked by the transition from external risks to "manufactured risks" or those that are created from the progression of human development, especially by the development of science and technology. A new moral climate of politics has arisen from the transition to risk society in which risks that did not originate in the political sphere must be politically managed. There is a constructed nature inherent to risk. When threats or hazards are immaterial or invisible all knowledge about them is mediated and as such subject to interpretation. In risk society, when immaterial or invisible technologically induced hazards develop, scientists, social theorists, journalists, politicians, business managers and the lay public can adopt similar structural positions with respect to truth, objectivity and certainty of knowledge. In reality some people have greater access to information and research facilities for acquisition of knowledge but the ontology of risk does not favour any specific form of knowledge. The ways in which individuals interpret information, apprehend knowledge and the communication of this knowledge can influence the societal response to a risk. The notion of risk society is a concept of ambivalence in that distinctions are destroyed and antitheses are reconnected (Beck, 2000). A postmodernist framework supports this notion that the peculiar construction of a constructed reality forms a public frame of reference based on possible outcomes with destruction or disasters in some cases, not yet seen. Risks could be viewed as a sort of virtual reality. Perception of a threatening future influences an individual's current action to address risk. The distinction between knowledge, latent impact and symptomatic effects is a critical component of the postmodern interpretation of risk society. Postmodernist plurality recognizes that risks can be simultaneously factual statements and value statements or something in between. Risk and the perception of risk are the unintended consequences of the logic of control predominant in modernity. Originally, Talcott Parsons suggested that modernization was the means for constructing order and control in a nation state. The by-products of these modernization processes, or risks, challenge the assertion of control by the nation state especially when consequences are global in nature and suffused with indeterminacy and uncertainty. Manufactured risk or uncertainty expresses the control or, perhaps more importantly, the lack of control over certain risks. A postmodernist approach helps to come to terms with the duality of 114 globality and locality. Environmental risks know no boundaries and globalization drives the creation of international institutions. Globalization influences uncertainty in risk in that the point of impact is sometimes removed from the point of origin and at the same time, the transmission or evolution of hazards may be invisible and untrackable to everyday perceptions. This innate invisibility of some risks means that, unlike many other political issues, they must be actively brought to the attention of society and only then can they become actual threats. These threats have a dualistic nature based on synthesis of cultural values, symbolic interpretation and scientific fact. Impacts grow, on the other hand, because either knowledge is insufficient to understand them or there is no demand to seek this knowledge. Knowledge about risks is connected to history and to the symbols of one's culture such as individual interpretation of nature. Since these perceptions are individual and dynamic, response varies throughout the world and this is a key reason why risks are perceived and managed politically differently in different states. Adding complexity to this issue is the spatial disjuncture between knowledge and impact - perceptions are necessarily contextual and constituted locally. The technological advances in communication offered by the information age have dramatically altered this local contextuality. Perceived Risk In a society enamored with the human and technological development that leads to the self-defined "better quality of life", the production of and unequal distribution of unprecedented and socially indiscriminate hazards is inevitable. The dark side of progress clothed in the term "risk" is both difficult to understand and difficult to measure. If we accept a definition of risk as being a systematic way of dealing with hazards and insecurities that accompany the process of modernization, then the elastic concept becomes intimately linked to interpretation. Since risk is inherently subjective and constructed by human minds and culture to help them cope with dangers and uncertainties of life, there can be no such thing as "real" risk or "objective" risk (Slovic 1992). There is clearly a distinction between "defined" and "perceived" risks but the distinction between these two is often blurred. The intuitive judgments that people make when asked to evaluate and characterize certain risks may be irrational and thus, without an understanding of public perception of risk, even the most well-intended policies may be essentially ineffective. 115 Perceptions of risk are tied to understanding what constitutes danger, hazards and threat and for whom. There is a constructed nature associated with risk perception that is based on people's ability to follow a logic or reason for its revelation. This is highly variable among individuals and is not born from the basis of voluntary imagination. The process of differentiating between defined risks that can be identified, characterized and statistically measured and perceived risks is especially challenging as a growing number of technologically induced hazards are inaccessible to human senses. Using psychometric analyses to study perceived risk, Slovic et al. (1987) made a vast contribution to our understanding of the distinction between defined and perceived risks by recognizing that cognitive processes of all individuals guide their intuitive risk judgments. The psychometric paradigm that evolved from psycho-scaling and multivariate analyses is a useful quantitative representation of risk perceptions in that cognitive processes and societal risk taking attitudes can be examined together. It is critically important to recognize that "risks" mean many things to different people both within the scientific community and the lay public. When scientific experts were asked to evaluate specific risks, their responses correlated highly with estimates of annual fatalities. However, the lay public could provide estimates of annual fatalities but they formed judgments of risk based on hazard characteristics such as catastrophic potential and threat to future generations. These perceptions of risk differed from both their own and scientific experts estimates of annual fatalities. The genesis of public perception of risk and the framing of the public understanding of the science associated with a risk can never be a purely intellectual process (Wynne 1992). The characterization, comparison and regulation of risk must be sensitive to a broader conception of risk. For many people first hand experience with some hazards is not apparent so exposure to information about these hazards rests with the media. The media plays a critical role in the risk communication process and often, public perception of risk depends on the transmission of knowledge delivered via the media. Language and imagery used by the media to portray events influences public perception of risk in different ways. 116 Scientific experts define risk in a narrow, technical way whereas the public perceives risk in a much broader, complex way that includes value-laden considerations such as equity, catastrophic potential and controllability (Slovic, 1992). When environmental risks are accompanied by considerable uncertainty, attempts to confine and control associated risks that result in conflicts influence public perception of risk and undermine trust in the institutions and scientific experts responsible for managing these risks. The public's perception of risk is not a purely intellectual process and as such, intuitive judgment shapes public attitude and willingness to tolerate some forms of risk while absolutely rejecting other forms of risk. The notion of being involuntarily subjected to many forms of environmental risk influences public perception of risk and in a world where human and technological development is coupled with considerable uncertainty, the immateriality and invisibility of many risks cause great concern. People incorporate perception of risk into their everyday lives so it is important that dimensions of both defined risk and perceived risk be incorporated into risk management and policy-making. The public perception of risk associated with the endocrine disruption hypothesis has not yet been analysed. There is a considerable concern among the informed public about the implications of exposure to xenoestrogens on human and ecosystem health although the publicized uncertainty regarding the associated scientific knowledge makes public perception about this risk somewhat malleable. Despite periodic but unsustained media attention to the endocrine disruption issues, the lay public is not especially well informed about the debate. The public perception may be that endocrine disruption is just another "manufactured risk" and as such individuals become somewhat inured to it. Living in risk society implies that individuals claim a degree of responsibility for their own exposure to risk so perhaps a visible generalized perception is less detectable. Because some endocrine disruption effects have been observed in human males, the perception that hormonally active chemicals pose a threat to human male sexual and reproductive health is very real. Extrapolating effects seen in wildlife and fish to potential risk to human health is challenged both in the scientific community and the lay public. Since reproductive responses are exceptionally species specific, the validity of animal models is questioned and thus public 117 perception does not readily associate environmental exposure of animals to estrogenic substances with observed effects in the human population (Djerassi, 2001)4. Stigma Stigmatization is very closely related to perceptions of risk. The word itself is taken as meaning a victim "marked" as flawed, deviant, limited, spoiled or undesirable in some way in the view of an observer (Krimsky & Golding, 1992). When any stigmatizing characteristic is apparent, the perception of the observer is changed in a negative fashion. When individuals or conditions are perceived to have attributes that exist outside the realm of prevailing standards of good or social norm, they are denigrated or avoided as being unduly dangerous. Jones et al. (1984) characterized six essential aspects of social stigma as being: concealable, changing over time, disruptive, derogative aesthetic qualities, uncertain of origin and perilous. Increasing concern that human and ecosystem health can be compromised by the development and application of technology has become a prominent driving force in the evolution of stigma and perception of risks. Technological stigmatization plays a crucial role in the public's perception of the benefits or risks of technological innovations and willingness to accept scientific evidence to make these decisions. The concept of stigma transcends the notion of something being a hazard. It involves the perception that something should be avoided not only because it may be considered dangerous but also because it undermines what is accepted as a positive condition. For this reason, stigma introduces a particular challenge to policymakers. Stigma is intimately connected to the visibility or ability to conceal the effect of a particular hazard. Dread consequences or involuntary exposure contribute to the development of high perceptions of risk and stigma. Often a critical event, accident or hazardous condition provides the imagery that can become associated with a particular risk. Negative imagery and negative emotional responses become linked to a perceived risk in an almost irreversible way. A very important component of stigma is that a condition that is accepted as normal or right is in some way violated and its impact may be seen as unequally distributed across groups or regional areas, uncertain in consequences and potentially catastrophic (Gregory et al., 1995) 4 Djerassi, Carl, Professor, Stanford University, Department of Chemistry. Personal communication. 2001. 118 Stigma often brings into question scientific competency in applying proper values and precautions in managing perceived hazards. Stigmatization is born with or without support of scientific knowledge and as such public fear of technological or environmental hazards cannot be alleviated strictly on the basis of supplying more scientific information. Social construction of stigma can rest solely on the individual's innate conservativism and the romantic notion of retaining "the good old days" that are perceived as being free of the negativities associated with technological development. (Djerassi, 2001)5 Uncertainties linked to visibly negative outcomes of a hazard amplify the stigma associated with technological risk. Public perception of abnormally great risk, distrust of management and evidence of technological failures significantly contribute to associate stigma. Public acceptance of tolerable risks can be influenced by communication and the imagery chosen by media to portray such issues. There is a complex interaction between psychological, social and political factors that influences an individual's willingness tb face difficult trade-offs when faced by high consequence risks. For this reason, policy and risk management processes are especially challenging when decisions about future economic activities and social responses are inextricably connected to stigma and public reticence to accept reasonable risks. Stigmatization of the endocrine disruption hypothesis is complex. Considerable uncertainty within the scientific community has been publicized and thus stigma, bearing resemblance to general chemophobia is apparent. The stigma associated with environmental exposure risk is present despite any direct evidence for risk to human health but human health observations based on accidental exposure or occupational exposure to endocrine disrupting substances seems to be enough for the public to form negative perceptions of the risk. The added complication to the stigmatization of the endocrine disruption issue is exacerbated by the media focus on human male reproductive health and, indirectly, sexuality. This stigma, in turn, challenges human male power and virility by implying a threat to male sexual health through uncontrolled exposure to some substances that have been demonstrated to cause physically observable reproductive abnormalities in animals. The sperm count debate strikes a particularly sensitive chord with most males despite conflicting evidence that global decline of sperm counts may or may not be 5 Djerassi, Carl, Professor, Stanford, University Department of Chemistry. Personal communication. 2001. 119 occurring. It is impossible to resolve the sperm count controversy in view of the multitude of complicating factors such as regional differences in population lifestyle and climate, male diet preferences for vegetarian foods containing phytoestrogens, dietary fat intake or even potential to be exposed to female urine in recycled water (Djerassi, 2001).6 However, stigma associated with endocrine disruption and decline of human male sperm counts is further complicated by an aura of taboo involving the connection between male virility with power. Risk Communication Risk communication is emerging as one of the most important aspects of contemporary risk management. It was originally seen as a means of bridging the gulf between expert views and public perception of risk. Risk communication has captured the attention of the public and the private sector as a component of the risk management process that shapes risk selection. People's selection of risk is dependant on a number of variables and the interconnection of cognitive processes and sociological approaches must be considered when examining why people make certain choices over others about matters of risk. The role of journalists and the media as interpreters of science and disseminators of scientific knowledge is extremely important in influencing public perception of risk and connecting this to economic, social and political response to perceived hazards. Risk communication involves a series of processes not all of which are connected to media but the major pathway for transmitting information to the lay public in a fashion that enables decisions regarding perceived risks to be made lies with the media. Language, text and imagery are, in turn, connected to the genesis of the public perception of risk and stigma can be influenced directly by media presentation of a hazard or conflict. As a conduit between scientific knowledge and the public's interpretation of it, risk communication has profound ramifications on the perceived seriousness of a risk. The media effectively plays a significant role in influencing public perception of risk and much of society's awareness of environmental risks is due, in part, to popularizing of the science that links human health and mortality to ecosystem health (Matthews, 2000)7. The media can nurture the 6 Djerassi, Carl, Professor, Stanford University, Department of Chemistry. Personal corxuriunication. 2001. 7 Matthews, Ralph. UBC Department of Anthropology and Sociology, Personal communication. 2000. 120 evolution of a controversy by serving as a channel for debate between the key actors in a conflict or by strictly providing information to the public during times of conflict. Often, large segments of the public face no first hand experience with a perceived risk or conflict and thus their decisions are based on their interpretation of the risk as it is communicated to them. This allows large latitude in perceptions born from varied structure and synthesis of risk messages and as a result, controversy can be amplified or attenuated on the basis of the presentation of information. The objective of risk communication is the transmission of information with the goal of educating the recipient of this information. This process involves passing on information about an event or hazard from an information bearer, or transmitter, to a receiver by means of a channel with a minimum of distortion of this information. Information transfer, however, is only one part of the risk communication process that also includes developing shared meanings, communities and institutions and establishing relationships of trust. (Kasperson et al., 1988). Risk communication is complicated and is not strictly a function of accurate and objective transmission of information via media. Rumour theory plays a significant part in formation of public perception of risk. When an information "vacuum" is thought to exist between the scientific knowledge of a risk and the perception of the risk, information can be synthesized to fill this vacuum and transmitted via rumour (Leiss & Powell, 1997). Rumour theory was derived from social network analyses and graph theory in which transmission of values and ideas is linked by complex social structures. These are not necessarily balanced social networks but the thread of commonality is emergent "wholes" of thoughts or perceptions that are distinct from their "parts" which, in turn, determine the nature of these parts. This is the central premise of gestalt tradition in psychology that suggests that individual perceptions are seen in a particular way because they are literally preconceived within complex and organized frameworks of the human mind. Perceptions are not formed independently of mental schemes but rather, in a fundamental sense, they are constituted by them. Rumour influences risk perception in that transmission of information can be convoluted as it is passes between actors or groups within a social structure. Information can be lost along the way or completely reinterpreted within the context of the values of a dominant actor or group within a social network. Cohesive sub-groups within a social network may perceive risk in a certain way based on relatively little accurate information but rumour transmission may overwhelm transmission of accurate knowledge and irreversibly construct public perception of risk. 121 Rumour and the communication of risk throughout the social network have a place within the framework of the public sphere as a mediating space between the state and society that was proposed by Habermas (19 ). The public sphere is the forum for public discussion in which both sides recognize the power of reason and the benefit derived from exchange of arguments between individuals, the confrontation of ideas and enlightened opinion. A commercial model of "manufacture of opinion" increasingly inspires publicity and public communication that fuels evolution of risk perception (Habermas, 1987). An implied assumption in the process of risk communication within the public sphere is that all perceptions are rational but such is not always the case. The risk communication associated with the endocrine disrupting hypothesis has some characteristics that are unique. Although concern and controversy existed in the scientific community for several decades, publication of one book conceived by one woman considerably influenced the perception of risk within the realm of policy makers and the lay public and bred a sense of urgency to address this potentially catastrophic hazard. The popular press adopted a distinctly gender biased position in describing the basis for the controversy. Endocrine disruption by natural and/or man-made hormones is the first environmental hazard accompanying the overwhelming technological advancement of the latter half of the twentieth century perceived to pose a direct threat to human male reproduction and sexuality. Arguably, for most of the past century, communication of women's health issues received considerable media attention. The risk of breast cancer to women is only one of the major issues related to women's reproductive health that has benefited from considerable media attention. Communication of science provided a basis for the understanding of and coping with health risk and contributed to women adopting self-stewardship practices relating to their own health. Publication of information related to human female health has been widespread since Victorian times. In contrast, male health related issues were rarely seen in the popular press. Slovic et al. (1992) have, in fact, documented what they describe as "white male effect" in which perception of risk was considerably less for white males than other groups studied including women as independent groups. "I believe (although I have no data at my fingertips) that women are more sensitized to chemical risks - pregnancy, children. More women seem to have "multiple chemical sensitivity" or at least let it be known. Unlike breast cancer for example, I don't see 122 men's groups organized around the possible chemical causes of prostate cancer." (Krimsky, 2001)8 The 1991 Wingspread Conference (Bern et al., 1992) was pivotal in the history of issues related to environmental hormones and formed the basis from which Theo Colburn would later construct her book describing the issues thought to be associated with endocrine disruption. The statement that was born from this meeting included an urgent warning that humans in many parts of the world were being exposed to contaminants that had disrupted development and reproduction of wildlife and laboratory animals, and that unless these chemicals are controlled, humans face the danger of widespread disruption in embryonic development and the prospect of damage that would last a lifetime. The conference sought to reach a global consensus on how to approach the issue of environmental exposure to hormones and establishing a correlation between effects of endocrine disrupting substances on human health. Since many of the endpoints seen in fish, birds, reptiles, amphibians, mammals and humans had effects that specifically compromised the health of male organisms, the media represented the information with emphasis on the potentially serious consequences on the health of male organisms (Colburn, 1995). Titles such as "Gender Benders Present in the Environment" and "Sperm Counts Effected by Environmental Chemicals" began to appear in the popular press after the publication of Theo Colburn's book. The book publicized some of the issues raised at the Wingspread conference and at the same time found its way to near the top of several bestsellers lists in developed nations. Television documentaries such as Sex Under Siege (CBC 1996) based on the B B C production Assault on the Male (1996) portrayed the issue in a fashion that emphasized magnitudes of uncertainties which could invoke a sense of fear in the programs audience. Following the broadcasts of documentaries in Canada, U K , US, Japan and Europe, non-scientific journals regularly contained media coverage, usually no more than one page, which described the potential risk of exposure of humans and animals to environmental hormones. 8 Krimsky, Sheldon. Tufts University, Personal Communication. 2001. 123 "It is not easy to bring a risk issue on the front burner of the general public's mind. It can happen in many ways: • A catastrophe • A sustained campaign backed by government action • A dramatic media event • Specific and dramatic human or environmental hazards (eg. Ozone hole) The issues surrounding endocrine disrupters are very diffuse. For instance, there is no smoking gun of human risk, there is no catastrophe (e.g. Love Canal), the media attention has not been sustained, and the scientific community is still in a wait and see mode." (Krimsky, 2001)9 Marshall McLuhan was right: the media is the message - the successful connection between information and a target audience depends intimately on the clothing of the message (Bresnick, 2000)10. In the case of media representation of science that deals with environmental risk of exposure to endocrine disrupting substances, the information is wearing a distinctly masculine fashion. Communication of information about environmental risk in this manner is deliberate and accompanied by discernible gender bias. Gender bias is not apparent in the communication of issues related to environmental or human health risks to exposure to environmental hormones within the scientific community but gender bias in media is necessary to target male members of society with information regarding the implications of serious risk to male health, human reproductive health in general and human species survival. This may have advantageous consequences. The mass media has provided the avenue for recognition of the urgency to address this issue and since many of the policy makers are men, information targeting a male audience has an impact on changing the relative importance and thus agenda-setting for dealing with the environmental/human health risks. Gender bias in media coverage of risks associated with environmental hormones is demonstrated by the overt references to male sexuality that often accompanies the information. Very often the science is accurately described but bold print statements such as "the state of men's sperm is ghastly all clues seem to point to a bizarre culprit: estrogen" (Gentlemen's Quarterly March 9 Krimsky, Sheldon. Tufts University. Personal communication. 2001. 1 0 Bresnick, Martin. Yale University School of Music. Cecil and Ida Green Visiting Scholar Lecture Series University of British Columbia. November 16, 2000. 124 1999 pp. 191-197) are designed to sensationalize. Semiotics with sexual undertones effectively transmits information to male audiences. In developing a strategy to manage an environmental or human health risk, it is important to understand the balance between the contribution of society and scientific experts to the formulation of the perceptions of risks and the tangential evolution of prescriptive policy to deal with these risks. Effective management of risk can therefore be simultaneously driven by the public perception of risk and the perception of risk held by those individuals charged with the responsibility for establishing prescriptive policy. Specifically targeting men within both of these circles has an advantage for accelerating the risk evaluation/management process when threats to human male health are possible. Sometimes, for very many reasons, members of rural society are reluctant to deal with issues that may be directly linked to human health risks, but when the risks are translated into terms that describe a risk to male virility, the issues become serious concerns (Veiga, 2000)1 1. Men sometimes resist human health issues and view them as a sign of weakness (Giddens, 2000)1 2 "...there is more of a bias the other way around - linked to masculinity because men see illness as a sign of weakness and therefore repress it or avoid it as related to themselves. Women see doctors more..." (Giddens, 2000)13 The possibility of environmental exposure of human males to compounds that could compromise sexual or reproductive health has substantial hegemonic implications. Although promotion of gender equity continues, men are frequently in positions of power. As an unexpected consequence of our technological and social evolution, environmental exposure to substances that could feminize, alter reproduction or have adverse effects on male health in general could be viewed as an uncontrolled challenge to the appearance of power in men. Sometimes, scientists prove to be poor communicators outside their research realm. This can have extremely detrimental consequences on the formation of perception of risk in the public sphere. It is becoming more common that scientists demonstrate talent for writing with vast 1 1 Veiga, Marcello. UBC Department of Mining and Mineral Processing. Personal communication 2000. 1 2 Giddens, Anthony, Director, London School of Economics, Personal communication 2000. 1 3 Giddens, Anthony, Director London School of Economics, Personal communication 2000. 125 general public appeal although Rachel Carson can be credited as the first woman to be recognized as a scientist and a writer. Theo Colburn collaborated with professional writers to produce a book with appeal to popular culture. In risk society where individual perceptions rule trust in science, risk tolerance and most aspects of social life, translating knowledge to the greatest number individuals as possible is beneficial. Carl Djerassi has pursued a newer genre of writing for the lay public with the overall goal to expose as many individuals as possible to accurate scientific knowledge disguised as entertainment. He has coined the phrase "science as fiction" to describe this style of communication which can influence public perception of risk. "...What I am trying to do is to use fiction as really a method of smuggling serious, accurate scientific ideas, concepts, behavioural aspects, even facts into a public's mind that is either ascientific or anti-scientific or not interested in science, or not aware of the issues. Everyone likes to hear a story. And then, in fact, after they've heard the story, and hopefully it was an interesting one to keep their attention, they actually learned something. And in the process, of course, of my trying to write this, to do that, which is not easy, I have learned something in the process too." 1 4 The social amplification of the risk of the endocrine disrupting hypothesis is simultaneously attenuated and amplified. The perception of endocrine disrupting substances as being just one more environmental contaminant to be inured limits the controversy to concerned individuals, interest groups, and members of the scientific community. There is still a large segment of the global population who are either unaware of or do not care about the potential consequences of the risk. At the same time, there is a particular aura of harm linked to this risk since effects are concealed and accompanied by considerable uncertainty. This amplification can be further extended since the endocrine disrupting hypothesis has unsubstantiated yet perceived connections to issues of reproduction, gender and sexuality. (ABC Radio National Broadcast, Interview: Carl Djerassi and Norman Swan. November 20, 2000, Sydney Australia) 126 EXAMINATION OF PERCEPTION OF RISK How odd it is that anyone should not see that all observation must be for or against some view if it is to be of any service. Charles Darwin Making sense of the physical, historical, cultural and moral knowledge that forms the basis of what people understand to be the risks posed by endocrine disruption requires an understanding of the nature of the knowledge itself and its transmission between various actors. To test the public perception of the endocrine disruption controversy within a heterogeneous interest group, a survey of ten questions was prepared to determine i f there was general knowledge about the existence of environmental contaminants that can behave as hormones. This questionnaire is found in Appendix C. The exercise was not intended to provide sufficient survey data from which statistically sound conclusions could be drawn but rather to practice one qualitative method. Within a six-week period, I developed and submitted a document for ethical approval and then distributed the small survey. It was my hypothesis that intermediary experts such as the medical community would be good indicators of the public perception of the risks of endocrine disrupting compounds to human health since these experts have some training as scientists and they have extensive occupational contact with patients. The limited evidence collected did not support such a hypothesis. The intent of the survey was to provide information in order to make recommendations for future study aimed at the investigation of the effect of EDC risk communication on the perception of the controversy and also to explore the hypothesis that the "white male effect" observed by Slovic et al. (1997) was also seen in the perception of risk associated with exposure to contaminants that can act as hormones in bodies of organisms. Of the fifty-seven questionnaires that were provided to a diverse spectrum of the medical profession, only fourteen were returned completed. Physicians who were asked to respond were general practioners, gynecologists, urologists, surgeons, dentists, medical students, endocrinologists, cancer research physicians, pediatricians and research cardiologists. The 127 questionnaires were mostly sent by e-mail or left at office reception desks with the request to fax the response. Four questionnaires were given directly to the physicians who expressed interest that they would be willing to complete them. Table 8.1 provides a profile of questionnaire the respondents. Years of Medical experience: Under 2 years 6 Between 2-10 years 2 Over 10 years 6 Males 8 Females 6 Identified Ethnicity Caucasian 12 Chinese 1 Hispanic 1 Table 8.1: Profile of Respondents Although this survey exercise was not particularly successful, some interesting information was learned from the responses. Of the individuals who chose to respond to my survey questions, eight were women and six were men. A l l individuals who chose to return the completed survey identified their ethnic origin as Caucasian, Hispanic or Chinese. Two people did not know where exposure to endocrine disrupting compounds could come from. Other responses for sources of EDCs were identified as water supply, pollution, ingestion of hormones from food (meat: specifically chicken was identified as a source), hormones used as growth enhancers in animals, plastics, contaminants passed through the food chain, hormone replacement therapy, illegal use of steroids, birth control pills and white dental filling materials. 128 Interestingly, one respondent, who is a dentist with 26 years experience in the field, had considerable knowledge of the EDC issue and particularly noted the risk of exposure of Bisphenol A from plastic-based white dental filling materials to children under age six. Although the need to fill permanent teeth of children under six years of age is less common than filling the teeth of adults, the tendency of children to swallow more often than adults implies that greater concentrations of Bisphenol A leached into saliva would pose greater exposure risk. In the dental community, scientific communication regarding the possible risk associated with white plastic-based dental fillings has been evident since the early 1990s (US National Research Council, 1999). Table 8.2: Questionnaire responses summary Question Response no yes same effect undecided 2 - greater effect on humans than on other organisms? 8 2 4 3 - direct effect on humans? 11 3 4 - effect different from other pollutants? 1 9 4 5 - does a controversy exist? 9 5 6 - ever been asked for information about EDCs? 12 2 7 - ever noticed gender bias in the media? 4 10 8 - do you know if a policy exists concerned with EDCs? 14 129 Table 8.2 provides the responses of the fourteen respondents to the survey questionnaire. Eight of the respondents believed that the exposure of humans to endocrine disrupting compounds would have a greater effect than on other organisms, two respondents believed that the effects would be less than those on other organisms and four respondents were undecided. Eleven out of fourteen respondents believed that human health could be affected directly by exposure to endocrine disrupting compounds and three respondents were undecided. In the opinion of nine individuals, the effects of exposure of humans to endocrine disrupting compounds were different than exposure to other pollutants. Three reasons were suggested for this difference: 1) effects on human sexuality, 2) elected use of anabolic steroids in humans and 3) mediation of the effects. One respondent stated that effects of exposure of humans to EDCs were not different than those for other pollutants (this respondent used the example that many pesticides are EDCs) and four respondents were undecided. With the exception of two people, who had never heard of any concerns regarding effects of EDCs and two undecided respondents, nine respondents agreed that a controversy exists and of these people, four noticed discernible gender bias in the popular press that reported information about EDC exposure. One person identified the effects on males regarding fertility and sperm counts as one indication of gender bias in the press dealing with this issue. Only two individuals (understandably since their medical specialties were endocrinology and urology) had been asked for information about EDC exposure with neither men nor women requesting information more often. It was interesting to learn from an individual from the BC Prostate Cancer Support Group (although no questionnaire was returned) that she would estimate that of people requesting information about prostate cancer, 60% were women. In fact, women comprised a significant portion of the membership of their support group. Of the individuals asked to respond to the survey, none were aware of any public policy dealing with EDCs. From this exercise I learned some important lessons for using surveys to collect information about perception of risk to endocrine disrupting compounds: • At least six months are needed to conduct a survey that would allow a statistically sound number of respondents to reply. • Experimental design requires that a relatively large number of survey questionnaires would be needed to achieve enough representative responses to provide sample information about a population upon which statistical analyses could be performed. 130 • Any response ratio greater than ten per cent is good. • November and December are perhaps not the best months to ask members of the medical community to respond to a small questionnaire. • Individuals, who have an interest in a particular controversy, have considerable understanding of it and seek more information. • In examining perception o f risk with professionals connected to human health issues, there appears to be a difference between individuals whose situation sees them more in contact with symptom-related issues than for individuals whose situation may see them faced with information related to causes of illness. • To seek information about the importance of EDC exposure as a risk to human and/or ecosystem health, it may be preferable to survey biologists rather than physicians. • The lay community may be the best influence group to survey (with statistically sound response numbers) in order to examine a possible white male effect related to environmental exposure to contaminants such as EDCs. • The art of designing a survey questionnaire requires special care to encourage responses that do not reflect bias on the part of the individual who constructs the survey. • To properly construct a questionnaire to examine perception of risk of the effects due to exposure to environmental contaminants, questions should be carefully selected to shed light on the ways in which culture, gender, politics, values and socio-demographic factors may influence an individual's tolerance or rejection of risk. • Qualitative methods such as survey questionnaires can uncover issues of concern that may have been completely overlooked by the researcher at the time of survey design, (e.g. concern on the part of the dental community regarding the risk associated with Bisphenol A leached into saliva) The Endocrine Disruption Controversy Revisited Within the past ten years, the body of scientific evidence that supports the concern about exposure of organisms to environmental hormones has exponentially grown. In Canada, 131 significant portions of the research budget of Environment Canada has been allocated for such programs as the Toxic Substance Research Initiative (TSRI). The latter has been dedicated to examining issues connected with endocrine disrupting substances. Since 1995, the OECD in Europe has sponsored an international task force which coordinates many research activities originally suggested at the Wingspread conference. Most developed nations are in some way involved in research associated with examination of the environmental and human health risks of exposure to endocrine disrupting substances. The controversy surrounding the endocrine disruption hypothesis evident in the scientific community is not readily visible in the lay public. In the evolution of the controversy, since there have been no trigger events and media attention is sporadic, the public perception of the risk is difficult to assess. In the absence of events that raise public awareness, knowledge transmission in the form of popular fiction or a "synthetic trigger event" even in the form of science-as-fiction stories is a viable option in the risk communication process. With vast amounts of knowledge easily accessible on the internet, the general public is exposed to the same scientific knowledge and popular press as that available to scientists and policy makers. Overall trust in science as a safeguard for the health of society is waning. The evolution to risk society is seeing a shift from trust in experts and the nation state to preserve health and security of the public to decision making on the part of individual members of society in management of their own exposure to risk. Since individuals have the tools and information to shape their own perceptions of risk, the controversy that would be otherwise very visible in the public sphere is essentially invisible. There is considerable evidence that the public is aware of the seriousness of the endocrine disrupting hypothesis. As one example, in the US and Europe, market demand exists for private companies to act as certifying agencies that can test the estrogenicity of materials used in the manufacture of baby products (Nicar, 2001). A significant proportion of consumers elect not to expose infants to plastic baby products that are perceived to contain plasticizers having endocrine disrupting potential. Private laboratories are responding to this demand by conducting the few screening tests available to date that can determine potential estrogenicity but they cannot conclusively determine human health impacts. "Most people do not have any idea about the massive amount of information regarding endocrine disrupters. People have a worry budget and it takes a while before some of the issues 132 get on to it. For example, the New York Times has been very sceptical about endocrine disrupters. The word doesn't appear very much. For a general public response, these terms have to enter into the vocabulary of popular culture. Ironically, the scientific community has incorporated the term - there is funding etc. But there remains lots of skepticism about the role the hormone-modulating chemicals play in disease." (Krimsky, 2001)1 5 "While it is still the case that endocrine disrupters are not on the radar screen of the general public, there are pockets of interest groups, such as breast cancer activists and pesticide activists who have embraced the framework for understanding risks. For example, I am on a committee in my City and State to look at pesticide risks from spraying pesticides for West Nile Virus. The people on the committee are aware that some of the chemicals in use are endocrine disrupters and use that understanding to inform their decision about the use of chemicals." (Krimsky, 2001)16 Modernity has evolved from a social system that is based on a figuratively fixed structure to one that recognizes the dynamic nature and plurality of the social network. Faced with competing perceptions of values and flexible interpretation of knowledge, defining and managing risk in a pluralist society is challenging. When "facts", "scientific truths" and "values" can have relative meanings that vary with individuals, gender, social groups, cultures, or institutional status, a postmodernist approach to the construction of perceived risk is applicable. In the construction of the controversy involving the endocrine disruption hypothesis, uncertainty and freedom of interpretation of information on the part of individuals may drive the perception of risk of endocrine disrupting substances as threatening human and ecosystem health. When the lay public is inundated with vast amounts of useful and useless information, perhaps synthetic trigger events or risk communication disguised as entertainment have the greatest probability of conveying the best available knowledge upon which to make risk judgments. The content and mechanism of risk communication is thus supremely important in this process. As individuals rely more on their own perceptions of controllability of risk exposure and less on the information provided by scientists, collective controversy may be more diffuse and less apparent. Nonetheless, individual tolerance of risk to exposure of compounds that may affect normal 1 5 Krimsky, Sheldon. 1 6 Krimsky, Sheldon. Tufts University. Personal communication. 2001. Tufts University, Personal communication, 2001. 133 endocrine system function is influenced by scientific uncertainty some of which is impossible to overcome. When the effects of perceived risk to human population of endocrine disrupting substances are multigenerational or in any way connected to sexual taboo, visibility of the controversy diminishes. The validity of animal models for understanding toxicology of substances will always be questioned and as such it is unlikely that extrapolation from observations of wildlife and fish will be accepted as a viable model of human health effects. However perceptions born of individual ecological worldviews challenge the resistance to accept animal models of health and behaviour. With competing worldviews, competing ethics, competing philosophies of what defines facts and values comes competing perceptions of risk. A l l of these processes are dynamic and thus by framing the controversy in somewhat postmodernist terms, it is understandable that risk is manufactured because we do not and cannot have all the answers. Science does not have infinite intellectual, temporal or financial capital to find all the answers. If the questions that are posed are laced with inherent plurality of meaning, science cannot define answers. Karl Popper (1959) suggests that science is built upon shifting sands that does not find "facts" but instead finds best approximations to truths. In the case of risk associated with the endocrine disruption hypothesis, like most risks, individual perception of best approximation of truth guides personal cognitive response to hazards. Rhetoric clouds risk communication and excessive information noise desensitizes the public to the best approximations to truths that science can provide. When we live in a society whose transformation can be figuratively described by the end of nature and the end of tradition, we cannot construct an understanding of the manufactured risk that is inseparable from scientific and technological development from past history because this does not exist. The social and political climate that emerges from this transformation presumes a reorientation of values and adaptable strategies for managing them. This may mean that personal choices or political decision-making would be based on information that is incomplete or less than perfect. Erring on the side of caution, the basic premise of the Precautionary Principle, has been suggested as the best management strategy for the ever-increasing risks we face in contemporary life. Application of the Precautionary Principle is risk specific and, in view of limited knowledge providing best approximations of truths, probably allows for conservative margin of error in management decisions. The temporal aspect of risks such as exposure to endocrine disrupting substances weakens the application of Precautionary Principle since 134 damage, i f indeed it exists, has already been done by the time the effects are noticed and cumulative effects of other environmental or health hazards cannot be examined in isolation. Theo Colburn (1996) asked for application of the Precautionary Principle in dealing with the issues associated with the endocrine disrupting hypothesis since her fundamental truth written in Our Stolen Future is that stakes are potentially very high and information very limited so acting with caution is prudent. Controversy exists whether the Precautionary Principle can be applied to the endocrine disruption hypothesis in the same fashion it is proposed to address global warming, climate change and development of viable AIDS vaccines. Many have accepted Colburn's fundamental truth and the synthetic trigger event that sparked the controversy, publication of Our Stolen Future, is probably a logical product of evolutionary processes of controversy in risk society. 135 CONCLUSIONS "By this knowledge we are loosenedfrom the chains of a most narrow dungeon, and set at liberty to rove in a most august empire; we are removed from presumptuous boundaries and poverty to the innumerable riches of an infinite space, of such beautiful worlds. Giordano Bruno The nature of inquiry is such that by seeking answers to questions, one discovers more questions to ask. Answers to questions provide us with information from which we can synthesize knowledge. The secret to gaining the most useful knowledge is by asking the right questions. Any knowledge is a valuable contribution when questions are many and answers are few. This experiment has provided new knowledge that in future provides a background for asking new questions about the effects of dietary exposure of juvenile coho salmon to 4-nonylphenol at the time of smoltification. In this study, juvenile coho salmon were exposed to a wide range of 4-NP concentration range via the diet (1 - 1233 mg/kg) for 28 days before their transfer to sea water. None of the treatments compromised the osmoregulatory ability of the fish or the utilization of feed or protein for growth during early sea water residency. The effect of the estrogen (E2) positive control actually represents the response of the fish to 31 mg/kg 17-p estradiol combined with 5 mg/kg 4-NP in the treatment diet. During the fresh water exposure phase, wet weight gain, condition factor, hepatosomatic index, specific growth rate, feed efficiency and protein efficiency ratio were higher than noted for the control and 4-NP treatment groups. Only the condition factors and hepatosomatic indices were not statistically significantly higher. This trend was qualitatively reversed during the sea water growth phase when all the fish were fed the basal diet with respect to wet weight gain, specific growth rate, protein efficiency ratio and percent protein deposited but these responses were not significantly lower relative to the other groups in the study. The findings of this experiment may indicate that the four-week period selected for dietary exposure of juvenile coho salmon to 4-nonylphenol was not the critical exposure window in the early life history of this species or that this species responds differently to 4-NP than other salmonids. Certainly, the endpoints observed by other researchers were not replicated in this 136 study. However, exposure of juvenile coho salmon at various times in the early life stages of the fish may provide some clues about the critical exposure window. In this research both dietary and waterborne routes of 4-NP administration should be explored just in case the dissimilar responses of the fish between studies were due to route of 4-NP administration. This latter possibility is unlikely since the fish in this study were exposed to 4-NP via the diet and water (skin and gills). Despite dose response body burdens of 4-NP in the experimental fish as determined by chemical analyses, the contaminant was effectively eliminated from the fish after twelve weeks post seawater transfer. Chemical analyses of the water in the experimental tanks confirmed the presence of 4-NP in the water column thus as mentioned, the fish received 4-NP by waterborne and dietary routes and this contributed to the observed results. The concentration of 4-NP in the water column varied with water depth and varied with increasing suspended organic matter in the water. Important knowledge was gained in this study about the ubiquitous nature of 4-nonylphenol contamination. Since small concentrations of 4-NP were determined in both the marine oil and the hydrolyzed krill that were used to prepare the diets, there was no control diet for the experiment that was free of 4-NP contamination. The technique for preparation of samples for analytical determination was refined through observation of tissue contaminant concentrations which varied with depth from the undersurface of the fish skin. Some information was collected that suggested that careful observation of coloration changes of the fish at parr-smolt transformation might be indicative of subtle effects of 4-NP on the endocrine function of the fish at this phase of its life cycle. Future examination of colour changes linked to parr-smolt transformation should be photographically documented. Although the performance and survival of the fish through the fresh water exposure phase and sea water growth phase of the experiment did not demonstrate significant differences between groups under the specific experimental conditions, it cannot be concluded categorically that dietary exposure of juvenile coho salmon to 4-NP did not effect the fish at this life stage although the evidence to date suggests this. Inhibitory effects of 4-NP treatment in the diets were not observed during the fresh water or sea water growth phase of the experiment. However, the observations of other researchers regarding inhibitory effects of estrogen treatment before transfer of the fish to sea water were observed in my experiment during the sea water growth phase when all fish were fed the basal diet. Vitellogenin analyses would be required to validate these observations in view of the previous research results. 137 In addition to this experimental evidence, a subjective examination of perception of risk to exposure to endocrine disrupting compounds has constructed the beginning of a framework from which the broader issues could be examined in a more comprehensive fashion. From the limited number of responses to the questionnaire designed to gain information about how members of the medical community view the connection of ecosystem health and human health, these individuals provided some support for the idea that human health issues may be linked to environmental contamination of pollutants. Knowledge of endocrine disruption as a class of pollutants was not well known but it was not unknown either. The information learned from the questionnaire suggests some gaps in risk communication related to endocrine disruption. Since there appears to be a difference between the perception of risk held by members of the medical community whose professional focus is linked to symptoms and the perception of risk shared by members of the medical community in particularly specialized fields, future qualitative surveys must consider intra group differences between influence groups in addition to the differences between various influence groups when examining the perception of risk associated with effects of environmental contamination. 138 FURTHER STUDY The truth is that life is marvelous for every second that we consider it worth living -that is to say, whenever we feel at one with reality. Seen in this light, the beauty of the Earth would be nothing but the permanent flame of our instinct for life - an agent of conservation. Jean-Louis Dumas-Hermes A number of specific research questions arise from results of this experiment. Filling the knowledge gaps in understanding the process of endocrine disruption using a fish model seems to be an infinite challenge. If exposure of coho salmon to 4-nonylphenol immediately prior to parr-smolt transformation does not have a significantly deleterious effect on growth and performance of this species, is there another sensitive developmental phase when exposure of the fish to the contaminant adversely affect the fish? The bilary-fecal route of elimination seems to effectively remove low levels of 4-NP that appear to be present in all formulated diets. Is this process the primary means that the fish uses for coping with environmental exposure to 4-nonylphenol and is the efficacy of this process life stage dependant? Does the bilary-fecal pathway allow 4-nonylphenol to form conjugates with glucuronides or sulphates that render the complex less bioavailable to the fish? Since 4-nonylphenol was found in low concentrations in all of the formulated diets, is it possible to construct a truly uncontaminated control diet? Moreover, does the presence of natural phytoestrogens (e.g. those present in soybean meal) in formulated diets contaminated with 4-NP affect the growth and performance of salmon? Since the exposure pathway to the endocrine system of the fish was not only the dietary route but also waterborne exposure and possibly dermal exposure, is it possible to isolate critical exposure pathways that may, indeed, have a deleterious effect on overall fish health? Do these observations have implications of ecological significance? The determination of octylphenol in many of the samples at the conclusion of the sea water phase suggests that this compound may not only be a contaminant. Further study is required to determine i f this octylphenol was a contaminant or a degradation product and this would involve 139 LC-MS analyses and more comprehensive examinations of octylphenol concentrations in samples relative to 4-nonylphenol concentrations. A number of questions arise when observing the endocrine disrupting compound controversy from the qualitative perspective. What effect has risk communication played in driving policy response to this controversy? Has interest in the endocrine disruption controversy diminished since the later 1990's and is endocrine disruption still perceived in the scientific community as a very important mechanism of action in environmental contamination of pollutants? Is the controversy, in fact, present in the general public or is the scientific community largely responsible for the concern for the risk to human and ecosystem health? Is the controversy invisible due to the influence that the information age has had on public perceptions of risk? Is there, indeed, a "white male effect" in the perception of risk of EDC exposure and does this have any observable connection to the direction that policy is taking to manage these risks? What weight of evidence is needed to implement a policy to manage the risk of exposure to endocrine disruption compounds? Do cultural differences or economics play a role in arriving at a sufficient weight of evidence to support action to be taken? In order to fully elucidate the magnitude and the severity of the risk of exposure of organisms to endocrine disrupting compounds, many future years of scientific research will need to explore the issue from disciplines such as toxicology, ecology, biochemistry, risk assessment and decision research. The momentum of interest in EDC research spawned by Theo Colburn's book Our Stolen Future seems to continue despite shift in public attention towards global security issues rather than risks posed by exposure to environmental pollutants. As new knowledge is gained, we move that much closer to understanding the implications of our technological genius on future generations of all species in contemporary risk society. 140 REFERENCES CITED Ahel, M . , E. Molnar, S. Ibric & W. Giger. 2000. Estrogenic metabolites of alkylphenol polyethoxylates in secondary sewage effluents and rivers. Water Science & Tech. 42: (7-8)15-22. Bailey, J.K., R.L. Saunders & M.I. Buzeta. 1980. Influence of parental smolt age and sea age on growth and smolting of hatchery-reared Atlantic salmon (Salmo salar). Can. J. Fish. Aquat Sci. 37:1379-1386. Beck, Ulrich. 1992. World Risk Society Polity Press, Oxford, U K . Bern, W.H., P. Blair, & S. Brasseur. 1992. Consensus statement from the work session on "Chemically-Induced Alterations in sexual development: the human/wildlife connection. In Chemically-induced Alterations in Sexual and Functional Development: The Human/Wildlife Connection. Report on the Wingspread Conference, Racine WI, July 26-28,1991. Princeton Scientific Press. Princeton NJ, US A. Bjornsson, B.T. & P. Persson. 1992. Plasma growth hormone levels in rainbow trout are decreased by both prolactin estradiol-170 treatment. In Proceedings of International Symposium of Fish Endocrinology. INRA, Paris. Blackburn, J. & W.C. Clarke. 1987. Revised procedure for the 24 hour seawater challenge test to measure seawater adaptability of juvenile salmonids. Can. Tech. Report Fish. Aquatic Sci. No. 1515. Blackburn, M . A . & M.J . Waldock. 1995. Concentrations of alkylphenols in rivers and estuaries in Endland and Wales. Water Research 29: (7)1623-1629. Bligh, E.B. & W.J. Dyer, 1959. A rapid method for total lipid extractions and purification. Can J. Biochem. & Physiol. 37, 911-947. Bradley, T .M. & A.W. Rourke. 1984. A n electrophoretic analysis of plasma proteins from juvenile Oncorhynchus tshawytscha. J. Fish. Biol. 24: 703-709. Brix, R., S. Huidt & L. Carlsen. 2001. Solubility of nonylphenol and nonylphenol ethoxylates: on the possible role of micelles. Chemosphere. 44(4): 759-763. Brown, S.B. & J.G. Eales. 1977. Measurement of L-thyroxine and 3,5,3'-triodo-L-thyronine levels in fish plasma by radioimmunoassay. Can.J.Zoo. 55:293-299. Brown, S.B., K. Haya, L.E. Burridge, D. Bennie, J.T. Arsenault, R.E. Evans, K. Burnison, J. Sherry, J.G.Eales, D. MacLatchy & W.L. Fairchild. 2001. The Effects of Alkylphenols 141 on Growth of Atlantic Salmon Smolts. Presented at the 28 Annual Aquatic Toxicity Workshop. Winnipeg, Manitoba. Brzozowski, A . M . et al. 1997. Molecular basis for agonism and antagonism in the estrogen receptor. Nature. 389: 753-758. Carson, Rachel. 1962. Silent Spring, Houghton Mifflin, Boston. Christiansen, L. B., K . L . Pedersen, S.L. Pedersen, B. Korsgaard & P. Bjerregaard. 2000. In vivo comparison of xenoestrogens using rainbow trout vitellogenin induction as a screening system. Environ. Toxicol. Chern. 19(7): 1867-1874. Cho, C.Y. & S.J. Kaushik. 1985. Effects of protein intake on metabolizable and net energy values of fish diets, p. 95-117. In. C.B. Cowey, A . M . Mackie & J.G. Bell (editors) Nutrition and feeding in fish. Academic Press. London, U K . Colburn, T. 1996. Our Stolen Future, Penguin Books, New York. Comber, M.H.I., T.D. Williams & K . M . Stewart. 1995. The effects of nonylphenol on Daphnia magna. Water Research 27: 273-276. Davies, J & B.J. Danzo. 1981. Hormonally responsive areas of the reproductive system of the male guinea pig. II. Effects of estrogens. Biological Reproduction 25: 1149-1158. Donaldson, E .M. , U . H . M . Fagerlund, D.A. Higgs & J.R. McBride. 1979. Hormonal enhancement of growth. In Fish Physiology VIII. Bioenergetics and growth. Academic Press. New York. 456-597. England, D.E. 1995. Chronic toxicity of nonylphenol to Ceriodaphnia dubia. Report prepared for the Chemical Manufacturers Association by A B C Laboratories Inc. Report No. 41756. an Publications. London, U K . Fairchild, W., S. Brown, E. Swansburg, & J. Aresenault. 1999. Does an association between pesticides use and subsequent decline in catch of Atlantic salmon (Salmo salar) represent a case of endocrine disruption? Environ. Health Per sped. 107:349-358. Folmar, L .C. & W.W. Dickhoff. 1980. The parr-smolt transformation (smoltification) and seawater adaptation in salmonids. A Review of selected literature. Aquaculture. 21:1-37. Francis, George. 1995. in Louis Quesnel (editor) Social Sciences and the Environment University of Ottawa Press, Ottawa, On. 145-171. Geisy, J.P. S.L. Pierens, E . M . Snyder, S. Miles-Richardson, V.J . Kramer, S.A. Snyder, K . M . Nichols & D.A. Villenueve. 2000. Effects of 4-nonylphenol on fecundity and biomarkers of estrogenicity in fathead minnows (Pimephales promelas) Environ. Toxicol. Chern. 19 No. 5: 1368-1377. Giddens, A . 1991. Modernity and Self-Identity Polity Press. Cambridge U K . 142 Giles, M . A . & W.E. Vanstone. 1976. Ontogenetic variation in the multiple hemoglobins of coho salmon (Oncorhynchus kisutch) and the effect of environmental factors on their expression. J. Fish. Res. Board. Can. 33: 1144-1149. Gimeno, S., H . Komen, A . G . M . Gerritsen & T. Bowmer. 1998. Feminisation of young males of the common carp, Cyprinus carpio, exposed to 4-rerr-pentylphenol during sexual differentiation. Aquat. Toxicol. 43: 77-92. Government of Japan - Environment Agency, Japan Society of Endocrine Disrupter Research, 1998. Proceedings of the International Symposium on Environmental Endocrine Disruptors. Kyoto, Japan. Government of Japan - Environment Agency, Japan Society of Endocrine Disrupter Research, 1999. Proceedings of the International Symposium on Environmental Endocrine Disruptors. Kobe, Japan. Gregory, R., J. Flynn & P. Slovic. 1995. "Technological Stigma" in American Scientist. 83: 220-224. Guillette, E. 1999. Persistent Organic Pollutants: Present and Future Implications for Human Health. Presented at Endrocrine Disrupting Compounds: A n Informal Workshop. BC Research Inc. Vancouver, BC Canada. Guilllette, L.J. Jr., T. S. Gross, G.R. Masson, J. M . Matter, H.F. Percival and A.R. Woodward. 1994. Developmental abnormalities of the reproductive system of alligators (Alligator mississippiensis) from contaminated and control lakes in Florida. Environ Health Perspect 102: 680-688. Guillette, L.J. Jr., T. S. Gross, D.A. Gross, A . A . Rooney, and H.F. Percival. 1995. Gonadal steroidogenesis in vitro from juvenile alligators obtained from contaminated and control lakes. Environ. Health Persp. 103 (Suppl. 4): 31-36. Guillette, L.J. Jr. and D.A. Crain. 1996a. Endocrine-disrupting contaminants and reproductive abnormalities in reptiles. Comments on Toxicology 5 (4-5):381-399. Guillette, L.J. Jr., D.B. Pickford, D.A. Crain, A . A . Rooney, and H.F. Percival. 1996b. Reduction in penis size and plasma testosterone concentrations in juvenile alligators living in a contaminated environment. Gen. Comp. Endocrinol. 101 (1): 32-42. Guillette, L.J. Jr., D.A. Crain, A . A . Rooney and D.B. Pickford. 1995. Organization versus activation: The role of endocrine disrupting contaminants (EDCs) during embryonic development in wildlife. Environ Health Perspect 103 (suppl. 7): 157-164. Guillette, L.J. Jr., S.F. Arnold and J.A. McLachlan. 1996c. Ecoestrogens and Embroys - Is there a scientific basis for concern? Animal Reprod. Sci. 42 (1-4): 13-24. Gunnes, K. & T. Gjedrem. 1978. Selection experiments with salmon. IV. Growth of Atlantic Salmon during two years in the sea. Aquaculture. 15:19-33. 143 Graumann, K. , A . Breithofer & A. Jungbauer. 1999. Monitoring of estrogen mimics by a recombinant yeast assay: synergy between natural and synthetic compounds? The Science of the Total Environment 2 2 5 : 69-79. Groot, C , L. Margolis & W.C. Clarke. 1995. Physiological Ecology of Pacific Salmon. University of British Columbia Press. Vancouver, BC, Canada. Guidotti, T. L. & P. Gosselin. 1999. The Canadian Guide to Health and the Environment. Duval House Publishing and The University of Alberta Press. Edmonton, A l . Canada. Habermas, J. 1987. The Theory of Communicative Action Beacon Press, Boston USA. Hale, R.C., C. Smith, P. de Fur, E. Harvey, E.Bush, M . Las Guardia & G.Vadas. 2000. Nonylphenols in sediments and effluents associated with diverse wastewater outfalls. Environ. Toxicol. Chern. 19 No 4: 946-952. Harries, J.E., D.A. Sheahan, S. Jobling, P. Mattiessen, P. Neall, E.J. Routledge, R. Rycroft, J.P. Sumpter and T.Tylor. 1996. A survery of estrogenic activity in United Kingdom Inland Waters. Environ. Toxicol. Chern. 15(11): 1993-2002. Haya, K. & L.E. Burridge. 1997. The effect of Cortisol and nonylphenol on growth and ornithine decarboxylase activity of juvenile Atlantic Salmon (Salmo salar). Presented at 24 t h Annual Aquatic Toxicity Workshop. Niagara Falls, Ontario. Hayes, T.B., T.H. Wu, & T.N. Gi l l . 1997. DDT-like effects as a result of cortiocsterone treatment in an anuran amphibian: Is DDT a corticoid mimic or a stressor? Environ. Toxicol. Chern. 1 6 : 1948-1953. Healey, M.C. 1991. Diets and feeding rates of juvenile pink, chum, and sockeye salmon in Hecate Strait, British Columbia. Trans. Am. Fish. Soc. 1 2 0 : 303-318. Henderson, M . A . & C C . Graham. 1998. History and status of Pacific Salmon in British Columbia. In Assessment and Status of Pacific Rim Salmonid Stocks - Bulletin Number 1 North Pacific Anadromous Fish Commission. Vancouver, Canada. Pp: 13-22 Higgs, D.A., J.S. MacDonald, C D . Levings and B.S. Dosanjh. 1995. Nutrition and Feeding Habits in Relation to Life History Stage. In Physiological Ecology of Pacific Salmon C. Groot, L. Margolis & W.C. Clarke, editors. University of British Columbia Press. Vancouver, Canada. Pp: 161-315. Higgs, D.A., U . H . M . Fagerlund, J.R. McBride & J.G. Eales. 1979. Influence of orally administered L-thyroxine or 3,5,3'-triiodothyronine on growth, food consumption, and food conversion of underyearling coho salmon (Oncorhynchus kisutch). Can. J. Zool. 5 7 : 1974-1979. Hirano, T., T. Ogasawara, S. Hasegawa, M . Iwata, & Y . Nagahama. 1990. Changes in plasma hormone levels during loss of hypoosmoregulatory capacity in mature chum salmon (Oncorhynchus keta) kept in seawater. Gen. Comp. Endocrinol. 78: 254-262. Hoar, W.S. & D.J. Randall. 1988. Fish Physiology. Volume XI . The Physiology of Developing Fish. Part A : Eggs and Larvae. Academic Press, Inc. London, U K . 144 Hoar, W.S. & D.J. Randall. 1988. Fish Physiology. Volume XI. The Physiology of Developing Fish. Part B: Viviparity and Posthatching Juveniles. Academic Press, Inc. London, U K . Holcombe, G.W., G.L. Phipps, M . L . Knuth & T. Felhaber. 1984. The acute toxicity of selected substituted phenols, benzenes and benzoic acid esters to fathead minnows (Pimephales promelas). Environ. Pollut. A35: 367-381. Huls, A . G . 1996. Determination of effects of nonylphenol on the growth of Scenedesmus subspicatus. 86.81 S A G Report AW-185. Ishibashi, H. , K. Tachibana, M . Tsuchimoto, M . Kobayashi, S. Watabe, F. Shiraishi & K. Arizona. 2001. Estrogenic activity of dietarty commercial diet with vitellogenin assay of ovariectomized goldfish Carassius auratus and yeast two-hybrid assay. Presented at SET A C Europe 11 t h Annual Meeting, Madrid, Spain. Janssen, P.A.H. , J.G.D. Lambert, A .D. Vethaak & H.J.T. Goos. 1997. Environmental pollution caused elevated concentrations of oestradiol and vitellogenin in female flounder, Platichthysflesus. Aquat. Toxicol. 39: 195-214. Jobling, S., M . Nolan, C R . Tyler, G. Brighty & J.P. Sumpter. 1998. Widespread sexual disruption in wild fish. Environ Sci. Technol. 32:2498-2506. Kasperson, R.E., O. Renn & P. Slovic. 1988. The social amplification of risk: a conceptual framework. Risk Analysis 8:(2) 177-187. Kelly, S.A. & R.T. Di Guilio. 2000. Developmental toxicity of estrogenic alkylphenols in killifish (Fundulus heteroclitis). Environ. Toxicol. Chem. 19 (10): 2564-2570. Keller-Byrne, J.E., S.A. Khuder, E.A. Schaub. 1997. Health effects of Vietnam veterans of exposure to herbicides. J. Am. Ind. Med. 31(5): 580-586. Kime, D.E. 1999. A strategy for assessing the effects of xenobiotics on fish reproduction. The Science of the Total Environment. 225: 3-11. Korner, W.V., Hanf, W. Schuller, C. Kempter, J.W. Metzger & H . Hagenmaier. 1999. Development of a sensitive E-screen for quantitative analysis of estrogenic activity in municipal sewage plant effluents. The Science of the Total Environment. 225: 33-48. Korner, Wolfgang, P. Spengler, U . Bolz, W. Schuller, V . Hanf & J.W. Metzger. 2001. Substances with estrogenic activity in effluents of sewage treatment plants in Southwestern Germany 1. Chemical Analysis. Environ. Toxicol. Chem. 20 (10): 2142-2151. Krimsky, Sheldon & Dominic Golding. 1992. Social Theories of Risk. Praeger Publishers Connecticut, USA. Krimsky, S. 2000. Hormonal chaos: the scientific and social origins of the environmental endocrine hypothesis. The Johns Hopkins University Press, Baltimore, M D . 145 Kristinsson, J.B., R.L. Saunders, & A.J . Wiggs. 1985. Growth dynamics during the development of bimodal length-frequency distribution in juvenile Atlantic salmon (Salmo salar). Aquaculture. 45: 1-20. Latour, B. 1993. We Have Never Been Modern Hemel Hempstead. London.UK. Leiss. W. & D. Powell. 1997. Mad Cows and Mother Milk McGill-Queens University Press. Montreal, Canada. Liber, K. , M . Knuth & F. Stay. 1999. An integrated evaluation of the persistence and effects of 4-nonylphenol in an experimental littoral ecosystem. Environ. Toxicol. Chem. 18 No.3:357-362. Lister, A . L . & Glen J Van Der Kraak. 2001. Endocrine disruption: Why is it so complicated? Water Qual. Res. J. Canada 36 (2): 175-190. Lussier, S.M. D Champlin, J. Livolsi, S. Poucher, R. Pruell. 2000.Acute toxicity of para-nonylphenol to saltwater animals. Environ. Toxicol. Chem. 19(5): 358-364. Maack, G. & H. Segner. 2001. Is there a period in zebrafish development particularly sensitive to estrogen exposure? Presented at SETAC Europe 11 t h Annual Meeting, Madrid, Spain. McDonald, J.S., I.K. Birtwell & G.M. Kruzynski. 1987. Food and habitat utilization by juvenile salmonids in the Campbell River estuary. Can. J. Fish. Aquat. Sci. 44: 1233-1246. McMartin, K . E . 1978. Diethylstilbestrol: a review of its toxicity and use as a growth promotant in food producing animals. J. Environ. Pathol. Toxicol. 1:279-313. McMaster, M.E . & K.R. Munkittrick. 2001. The ecological relevance of changes in the reproductive endocrine performance of fish at Canadian pulp mill sites over the period of mill moderization (1989-1999). Presented at SETAC Europe 11 t h Annual Meeting, Madrid, Spain. Madsen, S.S., A . B . Mathiesen & B. Korsgaard. 1997. Effects of 17 p-estradiol and 4-nonylphenol on smoltification and vitellogenesis in Atlantic salmon (Salmo salar). Fish Physiology and Biochemistry 17: 303-312. Metcalfe, T.L., C D . Metcalfe, Y . Kiparissis, A . Nimi, C M . Foran & W.H. Benson. 2000. Gonadal development and endocrine responses in Japanese Medaka (Oryzias latipes) exposed to o,/?'-DDT in water or through maternal transfer. Environ. Toxicol. Chem. 19(7): 1893-1900. Michalek, Joel D. & N . Ketchum. 2000. Serum dioxin and cancer in veterans of Operation Ranch Hand. Organohalogen Compounds 48: 99-102. Milne, D.J. 1950. The Differences in the Growth of Coho Salmon on the East and West Coast of Vancouver Island in 1950. Fish. Res. Board. Can. Prog. Rep. Pac. Coast Stn. 85: 80-82. Milne, D.J. 1964. The Chinook and Coho Salmon Fisheries in British Columbia. With appendix by H. Godfrey. Fish. Res. Board. Can. 142:46p. 146 Miwa, S. & Y . Inui. 1986. Inhibitory effects of 17-cc methyl testosterone and estradiol-17p on smoltification of sterilized amago salmon (Oncorhynchus rhodurus) Aquaculture 53: 21-39. Mommsen, T.P. and P.J. Walsh. 1988. Vitellogenesis and oocyte assembly. In Hoar, W.S. and D.J. Randall, eds. Fish Physiology Vol XIA. Academic Press, London, U K . pp 347-406. Munkittrick, K.R., G.J. Van Der Kraak, M.E . McMaster, C.B. Portt, M.R. Van den Heuval & M.R. Servos. 1994. Survey of receiving water environmental impacts associated with discharges from pulp mills. 2. Gonad Size, liver size, hepatic EROD activity and plasma sex steroid levels in white sucker. Environ. Toxicol. Chern. 13: 1089-1101. Nicar, M . 2001. Director of Research, Diagnostic Systems Laboratories Inc. Houston, Texas. Personal communication. Nicolas, J -M. 1999. Vitellogenesis in fish and the effects of polycyclic aromatic hydrocarbon contaminants. Aquat. Toxicol. 45: 77-90. Panter, G.H., R.S. Thompson, & J.P. Sumpter. 1998. Adverse reproductive effects in male fathead minnow (Pimephales promelas) exposed to environmentally relevant concentrations of oestrogens, oestradiol and oestrone. Aquat. Toxicol. 40: 335-360. Phare, M - A . , J .M. McKernan, R.D. Breu & P. Larcombe. 2001. EDCs, First Nations and the Role of Toxicology. Presented at 28 t h Annual Aquatic Toxicity Workshop, Winnipeg, Manitoba. Popper, K. 1959. The Logic of Scientific Discovery Harper & Row, New York. Prasad, R. 1989. Effect of nonylphenol adjuvant on macrophytes. Adjuvants Agrochem. 1:51-61. Purdom, C.E., P.A. Hardiman, V.J . Bye, N.C. Eno, C.R. Tyler & J.P. Sumpter. 1994. Estrogenic effects of effluents from sewage treatment works. Chern. Ecol. 8:275-285. Royce, W.F., L.S. Smith, & A.C . Hartt. 1968. Models of oceanic migrations of pacific salmon and comments on guidance mechanisms. Fish. Bull (US). 66:441-462. Sandercock, F.K. 1991. Life History of Coho Salmon (Oncorhynchus kisutch) in Pacific Salmon Life Histories. University of British Columbia Press. Vancouver, BC, Canada. Schmidt, C.W. 1997. Amphibian deformities continue to puzzle researchers. Environ. Sci. Tech. 31(7): 324A-326A. Scholz, S. & H.O. Gutzeit. 2000. 17-a-ethinylestradiol affects reproduction, sexual differentiation and aromatase gene expression of the medaka (Oryzias latipes). Aquatic Toxicol. 50:363-373. Schreck, C.B. & P.B. Moyle. Editors. 1990. Methods for Fish Biology. American Fisheries Society. Bethesda, Ma. USA. 147 Servos, M . , S. Luce, J. Toito, M . McMaster, K. Munkittrick, S. Huestis, M . Hagen & A . Colodey. 1997. The rapid decline of polycholorinated dibenzo-p-dioxins and furans in fish exposed to pulp and paper mill effluents in Canada. Presented at the 3 r d International Conference on the Environmental Fate and Effects of Pulp and Paper M i l l Effluents. Rotorua, NZ. Servos, M . 2000. National Water Research Institute, Burlington. Personal communication. Servos, M . , D.T. Bennie, B.K. Burnison, P. Seto. & A . Schnell. 2001. Occurrence, Fate and Release of Alkylphenol Polyethoxylates in Municipal Effluents and Industrial Effluents in Canada. Presented at 28 t h Annual Aquatic Toxicity Workshop, Winnipeg, Manitoba. Servos, M . , D. Bennie, K . Burnison, P. Cureton, N . Davidson & T. Rawn. 2001. Uncertainties associated with assessing the risk of an endocrine active substance in the Canadian environment. 2001. Endocrine disruption: Why is it so complicated? Water Qual. Res. J. Canada. 36(2): 319-330. Shang, D.Y., R.W. MacDonald & M.G. Ikonomou. 1999. Persistence of nonylphenol ethoxylate surfactants and their primary degradation products in sediments from near a municipal outfall in the Strait of Georgia, British Columbia, Canada. Environ. Sci. Technol. 33: 1366-1372. Shang, D.Y. , R.W. MacDonald & M.G. Ikonomou. 1999. Quantitative determination of nonylphenol polyethoxylate surfactants in marine sediment using normal-phase liquid chromatography-electrospray mass spectrometry. J. Chromotogr.A 849: 467-482. Sharara, F., D.B. Seifer & J.A. Flaws. 1998. Environmental toxicants and female reproduction. Fertility and Sterility 70(4): 613-622. Sherry, J., A . Gamble, M . Fielden, P. Hodson, K. Burnison & K. Solomon. 1999. An ELISA for brown trout (Salmo truttd) vitellogenin and its use in bioassays for environmental estrogens. The Science of the Total Environment. 225:13-31. Slovic, P. 1992. Perceptions of risk: reflections on the psychometric paradigm" in Social Theories of Risk Krimsky, S & D. Golding (eds). Praeger Publishers. Westport, CT, USA. Slovic, P. 1987. Perception of risk. Science 236: 280-285. Slovic, P. 1997. Trust, emotion, sex, politics, and Science: Surveying the Risk Assessment Battlefield. In Environment, Ethics, and Behaviour, M . H . Bazerman et al (Editors) New Lexington, San Francisco. Soto, A . , C. Sonnenschein, K . L . Chung, M.F. Fernandez, N . Olea, F. & Olea-Serrano. 1995. The E-Screen assay as a toll to identify estrogens: An update on estrogenic environmental pollutants. Environ. Health Persp. 103: Suppl. 7: 173-178. Sumpter, J.P. and Jobling, S. 1995. Vitellogenesis as a biomarker for estrogenic contamination of the aquatic environment. Environ. Health Persp. 103: 129-140. 148 Sumpter, J.P. C. Tyler, A . Sherazi, J.E. Caunter, T.D. Williams, M.J . Hetheridge, M.R. Evans, & U . Friederich. 2001. Bisphenol A : Multigenerational study with the fathead minnow (Pimephales promelas). Presented at SET A C Europe 11 t h Annual Meeting, Madrid, Spain. Talmage, S.S. 1994. Environmental and Human Safety of Major Surfactants, Alcohol Ethoxylates and Alkylphenol ethoxylates. Lewis, Boca Raton, FL. Tanabe, S. 1998. Contamination by Endocrine Disrupting Chemicals in Marine Mammals. Presented at 2 n d International Symposium on Environmental Endocrine Disruptors. Kyoto, Japan. Teskeredzic, Z., D.A. Higgs, B.S. Dosanjh, J.R. McBride, R.W. Hardy, R . M . Beames, J.D. Jones, M . Simell, T. Vaara & R.B. Bridges. 1995. Assessment of undephytinized and dephytinized rapeseed protein concentrate as sources of dietary protein for juvenile rainbow trout (Oncorhynchus mykiss). Aquaculture 131:261-277. Thorpe, J.E. 1977. Bimodal distribution of length of juvenile Atlantic salmon (Salmo salar) under artificial rearing conditions. J. Fish. Biol. 11: 175-184. Thorpe, J.E. & R.I.G. Morgan. 1980. Growth rate and smolting rate of progeny of male Atlantic salmon parr (Salmo salar). J. Fish. Biol. 17: 451-459. Thorpe, J.E., C. Talbot & C. Villarreal. 1982. Bimodality of growth and smolting in Atlantic salmon (Salmo salar). Aquaculture. 28: 123-132. Tyler, C.R. & E.J. Routledge. 1998. Natural and anthropogenic environmental oestrogens: the scientific basis for risk assessment - oestrogenic effects in English rivers with evidence of their causation. Pure and Applied Chern. 70:1795-1804. U.S. Nation Research Council. 1999. Hormonally Active Agents in the Environment National Academy Press. Washington DC, USA. van den Briel, V . 1999. Coho Salmon, (Oncorhynchus kisutch), Trial Experiment: Endocrine Disrupting Compound. Unpublished report prepared for Department of Fish Culture & Fisheries. Wageningen Institute of Animal Sciences. Wageningen Agricultural University. The Netherlands. van den Heuval, M.R., R.J. Ellis, E. Bandelj, L .H . McCarthy & T.R. Stuthridge. 2001. A Summary of reproductive-endocrine effects of a New Zealand pulp mill effluent. 2001. Seminar presented September 28 U B C Department of Chemical Engineering. van der Kraak, G., K.R. Munkittrick, M.E . McMaster & D.L. MacLatchy. 1998. A comparison of bleached kraft mill effluent, 17p-estradiol and P-sitosterol effects on reproductive function in fish. In Kendall, R.J., R.L. Dickerson, J.P. Geisy & W.P. Suk (eds) Principles and Processes for Evaluating Endocrine Disruption in Wildlife. SETAC Press, Pensacola FL. Ward, T.J. & R.L. Boeri, 1990a. Acute static toxicity of nonylphenol to the freshwater algae (Selenastrum capricornutum). Report prepared for Chemical Manufacturers Association by Resource Analysts. Study No.8969-CMA. 149 Ward, T.J. & R.L. Boeri, 1990b. Acute flow through toxicity of nonylphenol to sheepshead minnow (Cyprinodon variegatus). Report prepared for Chemical Manufacturers Association by Resource Analysts. Study No.8972-CMA. Ward, T.J. & R.L. Boeri, 1990c. Acute flow through toxicity of nonylphenol to the mysid shrimp (Mysidopsis bahid). Report prepared for Chemical Manufacturers Association by Resource Analysts. Study No.8974-CMA. Ward, T.J. & R.L. Boeri, 1990d. Acute static toxicity of nonylphenol to the marine algae (Skeletonema costatum). Report prepared for Chemical Manufacturers Association by Resource Analysts. Study No.8970-CMA. Ward, T.J. & R.L. Boeri, 1991a. Early life stage toxicity of nonylphenol to the fathead minnow (Pimephales promelas). Report prepared for Chemical Manufacturers Association by Resource Analysts. Study No.8979-CMA. Ward, T.J. & R.L. Boeri, 1991b. Chronic toxicity of nonylphenol to the mysid shrimp (Mysidopsis bahid). Report prepared for Chemical Manufacturers Association by Resource Analysts. Study No.8977-CMA. Wilson, E.O. 1998. Consilience - The Unity of Knowledge Alfred A . Knopf Publisher. New York, USA. Wynne, B. 1992. "Misunderstood misunderstandings: social identities and public uptake of science" Public Understanding of Science 1(3): 281-304. Young, G. 1996. Androgens and estradiol-170 inhibit Cortisol production of interregnal cells of salmonids. In Proceedings of Illth International Symposium of Fish Endocrinology Hakodate. 150 ADDITIONAL BIBLIOGRAPHY Adams, B., U . Beck & J. van Loon, editors. 2000.The Risk Society and Beyond: Critical Issues for Social Theory. Sage Publications. London, U K . Ahel, M . & W. Giger. 1985. Determination of alkylphenols and alkylphenol mono and diethoxylates in environmental samples by high performance liquid chromatography. Anal. Chern. 57: 1577-1583. Ahel, M . & W. Giger. 1993a. Aqueous solubility of alkylphenols and alkylphenol polyethoxylates. Chemosphere 26: 1461-1470 Ahel, M . & W. Giger. 1993b. Partitioning of alkylphenols and alkylphenol polyethoxylates between water and organic solvents. Chemosphere 26: 1471-1478. Ahel, M . , W. Giger & M . Koch. 1994a. Behaviour of alkylphenol polyethoxylate surfactants in the aquatic environment I: Occurrence and transformation in sewage treatment. Water Research 28:1131-1142. Ahel, M . , W. Giger & C. Schaffner. 1994b. Behaviour of alkylphenol polyethoxylate surfactants in the aquatic environment II: Occurrence and transformation in rivers. Water Research 28: 1143-1152. Ahel, M . , J. McEvoy & W. Giger. 1993. Bioaccumulation of the lipophilic metabolites of nonionic surfactants in freshwater organisms. Environmental Pollution 79: 243-248. Anderson, J.S., D.A. Higgs, R . M . Beames & M . Rowshandeli. 1996. The effect of varying the dietary digestible protein to digestible lipid ratio on the growth and whole body composition of Atlantic salmon (Salmo salar) (0.5-1.2 kg) reared in sea water. Can. Tech. Report of Fisheries & Aquatic Sci. 2104. Afonso, L.O. B., P .M. Campbell, G.K. Iwama, R.H. Devlin & E . M . Donaldson. 1996. The effect of the aromatase fradozole and fradozole aromatic hydrocarbons on sex steroid secretion by ovarian follicles of coho salmon. Gen. Comp. Endocrinol. 106: 169-174. Arvai, Joseph L. 2001. Linking Prescriptive and Descriptive Risk Communication Approaches for Risk Management and Decision Making. PhD. Thesis. The University of British Columbia. Asbell, Bernard. 1995. The Pill: A biography of the Drug that Changed the World. Random House. New York, USA. Beck, U . 2000. Risk Society and Beyond Polity Press, Oxford, U K . 151 Birnbaum, L . S. 1995. Developmental effects of dioxins and related endocrine disrupting chemicals. Toxicology Letters 82/83:743-750. Blaxhall, P.C. & K.W. Daisley. 1973. Routine haematological methods for use with fish blood. J. Fish. Biol. 5: 771-781. Bon, E., U . Barbe, J. Nunez Rodriguez, B. Cuisset, C. Pelissero, J.P. Sumpter & F. Le Menn. 1997. Plasma vitellogenin levels during the annual reproductive cycle of the female rainbow trout (Oncorhynchus mykiss): establishment and validation of an ELISA. Comp. Biochem. Physiol. 117B (1): 75-84. Bortone, Stephan A . and W.P. Davis. 1994. Fish intersexuality as indicator of environmetntal stress. Bioscience 44 (10): 165-172. Clark, W.C. 1982. Evaluation of the seawater challenge test as an index of marine survival. Aquaculture. 28: 177-183. Crisp, T. P., E.D. Clegg, R.L. Cooper, W.P. Wood, D.G. Anderson, K.P. Baetcke, J.L. Hoffmann, M.S. Morrow, D.J. Rodier, J.E. Schaeffer, L.W. Touart, M.G. Zeeman & Y . M . Patel. 1999. Environmental endocrine disruption: an effects assessment and analysis. Environ. Health. Persp 106: supplement 1: 11-52. Crisp, T. P., E.D. Clegg, & R.L. Cooper. 1997. Special Report on Environmental Endocrine Disruption: An effects Assessment and Analysis. Prepared for Risk Assessment Forum. US Environmental Protection Agency. Washington DC, USA. Dechaud, H. , C. Ravard, F. Claustrat, A . B . de la Perriere & M . Pugeat. 1999. Xenoestrogen interaction with human sex hormone-binding globulin (hSHBG). Steroids 64: 328-334. Del Junco, D., F. Kadlubar, S. Vernon, G. Stancel, A . Sweeney, X . Wu, N . Lang, A . Schecter, A . Garzon & T. Wheeler. 2000. Detecting an association between prostate cancer occurrence and TCDD exposure in the US Vietnam veteran population. Organohalogen Compounds 48: 95-98. Dempsey, S. M . and M . J. Costello. 1998. A Review of Oestrogen Mimicking Chemicals in Relation to Water Quality in Ireland. Published by Environmental Protection Agency, County Wexford, Ireland. Desaulniers, D., K . Leingartner, T. Zacharewski & W.G. Foster. 1998. Optimization of an MCF7-E3 cell proliferation assay and effects of environmental pollutants and industrial chemicals. Toxicol. In Vitro. 12: 409-422. De Voogt, P., K. De Beer & F. Van Der Weilen. 1997. Determination of alkylphenol ethoxylates in industrial and environmental samples. Trends in Anal. Chern. 16 (10): 584-595. Di Guilio, R.T. & E. Monosson. 1996. Interconnections Between Human and Ecosystem Health Chapman & Hall. New York, USA. Dosanjh, B.S., D.A. Higgs, D.J. McKenzie, D.J. Randall, J.G. Eales, N . Rowshandeli, M . Rowshandeli. & G. Deacon. 1998. Influence of dietary blends of menhaden oil and 152 canola oil on growth, muscle lipid composition and thyroidal status of Atlantic salmon (Salmo salar) in sea water. Fish Physiology and Biochemistry 19: 123-134. Figurovskii, N . A . & Y.I Solov'ev. 1988. Aleksandr Porfir'evich Borodin: A Chemist's Biography. Springer-Verlag Press. Berlin Heidelberg. Germany. Finkelstein, J., W. McCully, D. MacLaughlin & J. Godine. 1988. The morticians mystery. Gynecomastia and reversible hypogonadotropic hypogonadism in an embalmer. N. Engl. J. Med. 318: 961-965. Flynn, J., P. Slovic & C.K. Mertz. 1994. "Gender, Race and Perception of Environmental Health Risks" in Risk Analysis Vol . 14 No. 6. pp 1101-1108. Flynn, J., P. Slovic & H. Kunreuther. Editors. 2001. Risk, Media and Stigma - Understanding Public Challenges to Modern Science and Technology Earthscan Publications. London, U K . Foster, Warren G. 2001 Endocrine disruption and human reproductive effects: an overview. 2001. Endocrine disruption: Why is it so complicated? Water Qual. Res. J. Canada 36 No.2: 253-271. Fox, Glen A. 2001. Effects of endocrine disrupting chemicals on wildlife in Canada: Past, Present and Future. 2001. Endocrine disruption: Why is it so complicated? Water Qual. Res. J. Canada 36 No.2: 233-251. Frassinettei, Stefania, A . Isoppo, A . Corti & G. Vallini. 1996. Bacteial attack of non-ionic aromatic surfactants: comparison of degradative capabilities of new isolates from nonylphenol polyethoxylate polluted wastewaters. Environ. Tech. 17: 199-213. Giddens, A . 1990. The Consequences of Modernity Polity Press. Cambridge U K . Giddens, A . & C. Pierson. 1998. Conversations with Anthony Giddens Stanford University Press. USA. Glickman, T. S. & M . Gough. Editors. 1990. Readings in Risk Resources for the Future. Washington DC, USA. Groot, C. & L. Margolis. Editors. 1991. Pacific Salmon Life Histories. University of British Columbia Press. Vancouver, BC, Canada. Hannigan, J. A . 1995. Environmental Sociology - A Social Constructionist Perspective Routledge Press. London, U K . Hansen, P. D-., H . Dizer, B. Hock, A . Marx, J. Sherry, M . McMaster & C. Blaise. 1998. Vitellogenin - a biomarker for endocrine disruptors. Trends in Anal. Chem. 17 No. 7: 448-451. Haya, K. & L.E. Burridge. 1997. The effect of Cortisol and nonylphenol on growth and ornithine decarboxylase activity of juvenile Atlantic Salmon (Salmo salar). Presented at 24 t h Annual Aquatic Toxicity Workshop. Niagara Falls, Ontario. 153 Heemskerk, Vincent H. , M.A.R.C . Daemen & W.A. Buurman. 1999. Insulin-like growth factor-1 (IGF-1) and growth hormone (GH) in immunity and inflammation. Cytokine & Growth Factor Reviews 10: 5-14. Hesser, E.F. 1960. Methods for routine fish hematology. The Progressive Fish-Culturist. October: 164-171. Hewitt, M . & M . Servos. 2001. An overview of substance present in Canadian aquatic environments associated with endocrine disruption. 2001. Endocrine disruption: Why is it so complicated? Water Qual. Res. J. Canada 36 No.2: 191-213. Hoffman, D.J., B .A. Rattner, G.A. Burton, Jr., & J. Cairns, Jr. 1995. Handbook of Ecotoxicology Lewis Publishers. Boca Roton, FL. USA. Hussard, T. H. 1991. Understanding Biostatistics Mosby-Yearbook Publishers. St Louis, MO. USA. Hylland, K . & C. Haux. 1997. Effects of environmental oestrogens on marine fishes. Trends in Anal. Chern. 16 No. 10:606-612. Jones, E.E. et al. 1984. Social Stigma: The Psychology of Marked Relationships W.H. Freeman. New York. Jones, A . 1992. The Globe and Males. Gender Issues Educational Foundation Occasional Papers Series 1. McGi l l University Press. Montreal. Kime, D.E. 1998. Endocrine Disruption in Fish. Kluwer Press. Boston, M A USA. Kime, D.E., J.P. Nash & A.P. Scott. 1999. Vitellogenesis as a biomarker or reproductive disruption by Xenobiotics. Aquaculture 177: 345-352. Kloas, W. I. Lutz & R. Enspanier. 1999. Amphibians as a model to study endocrine disruptors. Science of the Total Environment 225 (l):42-59. Korach, Kenneth S. & John A McLachan. 1995. Techniques for detection of estrogenicity. Environ. Health. Persp. 103 (suppl.7): 5-8. Korner, W., V . Hanf, W. Schuller, C. Kempter, J.W. Metzger & H. Hagenmaier. 1999. Development of a sensitive E-screen for quantitative analysis of estrogenic activity in municipal sewage plant effluents. The Science of the Total Environment. 225: 33-48. Kraus, Nancy., T. Malmfors & P. Slovic. 1992. "Intuitive Toxicology: Expert and Lay Judgements of Chemical Risks" in Risk Analyses 12: Lash, S., B. Szersznski & B. Wynne. (Editors) 1996. Risk, Environment and Modernity -Towards a New Ecology Sage Publishers. Thousand Oaks, CA. USA. Lashley, B.L. & J.W. Overstreet. 1998. Biomarkers for assessing human female reproductive health, an interdisciplinary approach. Environ. Health Persp. Suppl. 106(4): 955-961. 154 Lee, H-B. 1999. Review of analytical methods for the determination of nonylphenol and related compounds in environmental samples. Water Qual. Res. J. Canada. 34(1): 3-35. Legler, J. 2001. Development and application of in vitro and in vivo reporter gene assays for the assessment of (xeno-)estrogenic compounds in the aquatic environment. PhD. Thesis. Wageningen Agricultural University. Wageningen. The Netherlands. Legler, J., C.E. van den Brink, A . Brouwer, A.J . Murk, P.T. van der Saag, A.D. Vethaak & B. van der Burg. 1999. Development of a stably transfected estrogen receptor-mediated luciferase reporter gene assay in the human T47D breast cancer cell line. Toxicological Sciences 48: 55-66. McCallum, L M . & D.A. Higgs. 1989. An assessment of processing effects on the nutritive value of marine protein sources for juvenile Chinook salmon (Oncorhynchus tswawytscha). Aquaculture 11: 181-200. Macnaughton, P. & J. Urry. 1998. "Rethinking Nature and Society" 1-31 and "Inventing Nature", 32-74 in Contested Natures. Sage Publishers, Thousand Oaks, CA, USA. Maguire, J.R. 1999. Review of the persistence of nonylphenol and nonylphenol ethoxylates in aquatic environments. Water Qual. Res. J.Canada 35(1): 37-78. Martin, B. , and Richards, E. 1995. "Scientific knowledge, controversy and public decision making." In Handbook of Science and Technological Studies Thousand Oaks, Ca. Sage Publications. Munkittrick, K.R. 2001. Assessment of the effects of endocrine disrupting substances in the Canadian environment. 2001. Endocrine disruption: Why is it so complicated? Water Qual. Res. J. Canada 36(2): 293-302. Nicar, M . 1998. Immunoassay Techniques in the Study of Estrogen Mimics in Health and Disease. Presented at SETAC 19 t h Annual Meeting. Charlotte, North Carolina. Nilsen, B . M . , K. Berg, A . Arukwe & A . Gorsoyr. 1998. Monoclonal and polyclonal antibodies against fish vitellogenin for use in pollution monitoring. Marine Environmental Research 46: (1-5): 153-157. Norberg, B. & C. Haux. 1985. Induction, isolation and a characterization of the lipid content of plasma vitellogenin from two Salmo species: rainbow trout (Salmo gairdneri) and sea trout (Salmo truttd). Comp. Biochem. Physiol. 81B: 869-876. Oberdorster, E. & A . Oliver Cheek. 2000. Gender benders at the beach: endocrine disruption in marine and estuarine organisms. Environ. Toxicol. Chem. 20(1): 23-36. Olea, N . A . Rivas & F. Olea-Serrano. 1997. Human exposure to endocrine-disrupting chemicals: assessing the total estrogenic xenobiotic burden. Trends in Analytical Chemistry. 16 (10): 613-619. Olmstead, A. & G.A. LeBlanc. 2000. Effects of endocrine-active chemicals on the development of sex characteristics of Daphnia magna. Environ. Toxicol. Chem. 19(8): 2107-2113. 155 Parrott, J., M.Wade, G. Timrn & S. Brown. 2001. An overview or testing procedures and approaches for identifying endocrine disrupting substances. 2001. Endocrine disruption: Why is it so complicated? Water Qual. Res. J. Canada 36(2): 273-291. Report of Proceeding: European Workshop on the Impact of Endocrine Disrupters on Human Health and Wildlife. December 2-4, 1996 Weybridge U K . Routledge, E.J. & J.P. Sumpter. 1996. Estrogenic activity of surfactants and some of their degradation products assessed using a recombinant yeast screen. Environ. Toxicol. Chern. 15:241-248. Scott, J. 2000. Social Network Analyses: A Handbook Sage Publications. London U K . Shearer, K . D . 1994. Factors affecting the proximate composition of cultured fishes with emphasis on salmonids. Aquaculture. 119:63-88. Sheridan, M . A . 1986. Effects of thyroxin, Cortisol, growth hormone and prolactin on lipid metabolism of coho salmon, Oncorhynchus kisutch, during smoltification. Gen. Comp. Endocrinol. 64: 220-238. Simons, S. 1996. Environmental estrogens: can two "alrights" make a wrong? Science 272: 1451. Slovic, P. Editor. 2000. Perception of Risk Earthscan Publishing. Sterling V A , USA. Snow, C P . 1959. The Two Cultures and the Scientific Revolution. Cambridge University Press. Cambridge, M A USA. Snyder, S.A., T.L. Keith, D.A. Verbrugge, E . M . Snyder, T.S. Gross, K . Kannan and J.P. Geisy. 1999. Analytical methods for detection of selected estrogenin compounds in aqueous mixtures. Environ. Sci. Tech. 33: 2814-2820. Spengler, P., W. Korner & J.W. Metzger. 2001. Substances with estrogenic activity in effluents of sewage treatment plants in Southwestern Germany 1. Chemical Analysis. Environ. Toxicol. Chern. 20(10): 2133-2141. Sonnenschein, C. & A . M . Soto. 1998. An updated review of environmental estrogen and androgen mimics and antagonists. J. Steroid Biochem. Molec. Biol. 65(1-6): 143-150. Staples, C.A., D.R. Peterson, T.F. Parkerton, & W.J. Adams. 1997. The environmental fate of phthalate esters: a literature review. Chemosphere. 35(4): 667-749. Staples, Charles A. , P.B. Dorn, G .M. Klecka, D.R. Branson, S.T. O'Block & L.R. Harris. 1998. A review of the environmental fate, effects and exposures of bisphenol A . Chemosphere. 36: 2149-2173. Straube, E., W. Straube, E. Kruger, M . Bradatsch, M . Jacob-Meisel & H.-J. Rose. 1999. Disruption of male sex hormones with regard to pesticide: pathophysiological and regulatory aspects. Toxicology Letters 107: 225-231. 156 Sutcliffe, R. 2001. Endocrine disrupting substances and ecological risk assessment of commercial chemicals in Canada. 2001. Endocrine disruption: Why is it so complicated? Water Qual. Res. J. Canada 36(2): 303-317. Suto, A . 2000. Effects of Nonylphenol on Stress Response in Rainbow Trout (Oncorhynchus mykiss). M.Sc. Thesis. The University of British Columbia Tattersfield, L. , P. Mattiessen, P. Campbell, N . Grandy and R. Lange (editors); SETAC Europe/OECD/EC Expert Workshop on Endocrine Modulators and Wildlife: Assessment and Testing, April 10-13, 1997 Veldhoven, Netherlands. Ternes, T.A. M . Stumpf, J. Mueller, K. Haberer, R,-D. Wilken & M . Servos. 1999. Behaviour and occurrence of estrogens in municipal sewage treatment plants - 1. Investigations in Germany, Canada and Brazil. Science of the Total Environment 225: 81-90. Ternes, T.A., P. Kreckel and J. Mueller. 1999. Behaviour and occurrence of estrogens in municipal sewage treatment plants - II. Aerobic batch experiments with activated sludge. Science of the Total Environment 225: 91-99. Tremblay, L. , & G. Van Der Kraak. 1998. Use of a series of homologous in vitro and in vivo tests to evaluate the endocrine modulating of P-sitosterol in rainbow trout. Aquatic Toxicology 43: 149-162. U.K. Environment Agency. 1997. Draft comprehensive risk assessment report for nonylphenol. National Centre for Ecotoxicology and Hazardous Substances. London, U.K. Wallace, R.A. & A Wolf. 1999.Contemporary Social Theory: Expanding the Classical Tradition Fifth edition. Prentice Hall. New Jersey, USA. Yassi, A. , T. Kjelstrom, T. de Kok & T.L. Guidotti. 2001. Basic Environmental Health. World Health Organization. Oxford University Press. Oxford, U K . Zacharewski, T. 1997. In Vitro Bioassays for assessing estrogenic substances. Environ. Sci. Tech. 31(3): 613-623. Zar, J.H. 1984. Biostatistical Analyses Prentice Hall. Englewood Cliffs, NJ , USA. Zou, E. & M . Fingerman. 1999. Effects of exposure to diethyl phthalate, 4-(terr)-octylphenol, and 2,4,5-trichlorobiphenyl on activity of chitobiase in the epidermis and hepatopancreas of the fiddler crab, Uca pugilator. Comparative Biochemistry and Physiology Part C 122:115-120. Zuckerman, S. 1940. The histogenesis of tissues sensitive to estrogens. Biological Review 15: 231-271 157 APPENDIX A Summaries of Proximate Analyses of fish and diet sample 158 (0 CO Q s C (0 rj) c </) iZ «+-o E E (0 X 0) in CO CO CO CN CO CO o CN 73 T— T— T— T— T— T— T— T— CO CO C o o o o o o o o o o o o X d d d d d d d d d d d d LL o i^ - T— CO 00 a> o 00 CD CO o CD CD I-- CD r- CD CD 00 00 o o _i o > CO CD cr> T— CD CN a> 00 O) o o a> CN tf) CN CN O) tf) a> 00 00 o a m tf) CD tf> tf) CM T3 o d d d d d d d d d d d •*-» tf) +~* CO T" CN| _i CO 00 CN tf) •sr ^~ in O) At c CO CD tf) tf) co CD c ns 0) CN csi csi CN CN CN csi csj csi csi csi csi S & > •<* tf) CO oo tf> a> •<? CN i^ T— 00 CO CN CO T— o tf) o Q CD CO o CT> CN CO CO i^ - O o "D CN CN CO CN CO csi CN CN csi csi CO CO *J (0 Wt CN CN oo co 1^ o O) J^- tf) c CN CD 00 CD 00 CO O) o CSI o ra d T— d d d CO CO a> CN CN CN CN CN CN CSI CN CN CN CN S o o o CM ro c Ci E ro c ro o i m o i m t o N m o i o D N o D N c o c o c o c o c o c o c o c o c o c o c o c o O CO CO T - T -•<* "tf M-d tf) o d d CD d o oo co d oo tf> co -fl-ea co co •<3-T CD co •ST CO CO o d d •^ r co oo in o o o i - o r ^ T - r ^ a ) T - T - i ^ -T - O O O O O O O oo in csi co •<r co tf) T CO "fl" tf) o CN tf) o oo CO 1^ CD CO CN o tf) r«- •«- r*. CD tf) CO O) tf) tf) ^ CO CD tf) co cn tf) •^ r o tf) oo o o csi csi tfi ih tfj tf) i^ CN tf) tf) •<3- CN T - ^ 00 cq T - CN tfj tf) tfj tf) c o CD CD 0_ 0_ CD CD E E CN CN O O O O O O o CD CD CD ±c ^ Q_ CL CL -z.-z.~z. CD CD CD E E E CSI csi 0 O O CN CD CD CD ^: ~- ~~ — 0_ 0. CL z z z ^3" CD CD CD E E C O O c o o O O O ^ ^ CN CN CN LU LU 00 ^ N (N CN CO CSI CO CN CN CN CN tf) T - CD CN CO CSI CN CN CN O CO CD "fl" CO CO CN CN CSI CO CM CSI CN CN CN 159 (0 •4-1 ro 0 S £ ro D) C Q) -1 </> i l i t -CD E E 3 (0 X a> • t f CO m CD in CD m CO TJ T— T— T— T— T— ^ — ^ — ^ — ^ — J C o o o o O o O o o o o o X o o o o O d O o d d d d ul 00 CD o O o ^ — CN CT) m CN CT) TJ • t f m 00 in o co m • t f m CO C o CO CO co 00 co co co CO CO co CO CO o o o o > CD 00 CO CO CD o T— • t f CO o a> 00 CD CT) CO c n CT) CN T— CO m CO r- " a in CO • t f in CD CO CN r»- CD CD CM TJ 4-1 CO L -Q) _i •st CN CN o CD • t f CT) CT) in m CO c T— CN CT) CO CD CT) CT) co E (S h-' CD r-^  CD te £ Q . <D (0 co 1^  T- -^ CD m CT) CO • t f > CD in o O CT) CN CO CT) CT) CO • t f CO h~ c n CN CN CO o CO CO o o in • t f cb CO CT) d CN 1^  in JJ CN T— CN CN T— CO Wt -tf CT) CD CN o CT) CN • t f m CD CD c • t f *~ O O h~ CN in CD in in CO CO A 00 O CT) ^ — CT) CT) CO CT) T — a) CD CD CD CD CD CD CD CD CD in CD in £ c 0) E CD ul 73 CD CO m CO CD co CN CD T— C O m 00 CT) CD CO CD CN CO CO CD in C o CN CN CN CN CN CN CN CN CN CN CN CN o J o > 0) Q CT) in CT) CD T— CO T— 00 CO CD CD CD •tf in co in •tf co in CN o CN CD •tf o CT5 m m CO oo CO h-o o CM T J CO d d d d d d d d d T J " C CM J c o CO in co CN CM co •tf • t f oo in CO o oo oo CT) CN oo CO ro 0) iri • t f in in in in in in • t f • t f in • t f |USt 1 E |USt 1 Aug > a> Q CT) m •tf in r— "tf in CD co CT) in in CT) CN 00 • t f CO CO m o CN CD o in in o co •tf CO T J 00 cb cb cb c d r-' c d CT) cb r--' in CO c CN CD CO CO CT) co o • t f CD o CN oo 00 CD CN in o CN CO o ro a> E CN •tf CO d co •tf d •tf • t f • t f • t f • t f d •tf CO o i CO 5 CO C as O ) O ) m XL a. z •tf 0. • t f O ) XL D ) O ) O ) XL XL XL C D D ) XL XL 0_ 0_ D_ CL 0_ CL Z Z c c o o O O cn a> E E CN CN O O O d o d • t f O J E CN d • t f - t f - t f CT) D ) TO E E E ^ o o O CN CN " t f TO TO E E o o o o O O CM N CN CN LU LU CO -tf CN CO CN CN N C M in CN CO CN CN CN CN T- CD O CO CT) -tf CO CO CN CO CN CN CN CO CN CN CN CN CN CN CN 160 June 21st 2000 Condition Hepatosomatic Weight Length Sex Factor Liver Wt Index Tank No. 223 1 (2000mg 4NP/kg) 20.2 12.3 M 0.0058 0.30 0.015 2 24.8 13.2 M 0.0057 0.30 0.012 3 22.7 12.9 M 0.0056 0.29 0.013 4 19.1 12.3 F 0.0055 0.23 0.012 5 21.4 12.4 F 0.0060 0.20 0.009 6 23.5 12.8 M 0.0059 0.22 0.009 7 19.4 12.4 F 0.0054 0.21 0.011 8 16.3 11.5 F 0.0058 0.20 0.012 9 23.4 12.6 M 0.0062 0.25 0.011 10 18.4 12.3 M 0.0053 0.28 0.015 11 23.6 12.8 M 0.0059 0.28 0.012 12 21.9 12.2 M 0.0065 0.41 0.019 Tank No. 224 1 (E 2 31 mg/kg) 25.1 13.0 F 0.0060 0.74 0.029 2 23.8 12.8 F 0.0060 0.74 0.031 3 24.0 12.5 F 0.0065 0.69 0.029 4 26.6 13.4 F 0.0058 0.94 0.035 5 23.9 12.8 M 0.0060 0.78 0.033 6 26.1 13.3 F 0.0058 0.62 0.024 7 21.6 12.3 F 0.0062 0.79 0.037 8 25.3 13.0 M 0.0061 0.82 0.032 9 21.7 12.3 M 0.0062 0.60 0.028 10 28.1 13.1 M 0.0066 0.85 0.030 11 20.8 12.0 F 0.0065 0.63 0.030 12 21.7 12.5 F 0.0059 0.58 0.027 Tank No. 225 1 (0.2mg 4NP/kg) 20.0 12.4 M 0.0056 0.38 0.019 2 22.9 13.1 M 0.0054 0.29 0.013 3 21.2 12.6 F 0.0056 0.27 0.013 4 14.2 11.2 M 0.0055 0.28 0.020 5 25.6 13.6 F 0.0053 0.34 0.013 6 23.4 13.0 F 0.0056 0.35 0.015 7 22.3 12.6 F 0.0059 0.23 0.010 8 18.0 11.9 M 0.0058 0.21 0.012 9 18.4 11.8 M 0.0060 0.26 0.014 10 20.7 12.6 M 0.0055 0.27 0.013 11 19.8 12.2 F 0.0058 0.28 0.014 12 21.5 12.3 M 0.0062 0.34 0.016 161 June 21st 2000 Condition Hepatosomatic Weight Length Sex Factor Liver Wt Index Tank No. 226 1 (20mg 4NP/kg) 22.9 12.9 F 0.0056 0.27 0.012 2 20.8 12.3 M 0.0060 0.29 0.014 3 20.6 12.1 M 0.0062 0.23 0.011 4 20.4 12.3 M 0.0059 0.17 0.008 5 21.5 12.7 M 0.0056 0.26 0.012 6 18.0 11.8 F 0.0059 0.22 0.012 7 19.3 12.3 F 0.0055 0.20 0.010 8 21.9 12.8 M 0.0055 0.28 0.013 9 24.5 13.3 F 0.0055 0.29 0.012 10 19.6 12.6 F 0.0052 0.24 0.012 11 22.9 12.6 M 0.0061 0.29 0.013 12 20.5 12.4 M 0.0057 0.24 0.012 Tank No. 227 1 (0.002mg 4NP/kg) 24.8 13.2 F 0.0057 0.29 0.012 2 21.6 12.6 M 0.0057 0.28 0.013 3 18.6 11.8 F 0.0061 0.22 0.012 4 20.5 12.3 M 0.0059 0.27 0.013 5 19.5 12.1 M 0.0059 0.31 0.016 6 24.8 13.2 F 0.0057 0.37 0.015 7 19.9 12.4 F 0.0056 0.24 0.012 8 21.3 12.3 F 0.0061 0.24 0.011 9 25.8 13.4 F 0.0056 0.32 0.012 10 21.1 12.3 M 0.0061 0.23 0.011 11 25.5 13.1 M 0.0060 0.33 0.013 12 20.0 12.2 M 0.0059 0.26 0.013 Tank No. 228 1 (Control) 20.3 11.9 F 0.0065 0.71 0.035 2 22.4 12.9 F 0.0055 0.28 0.013 3 20.8 12.2 F 0.0061 0.23 0.011 4 23.4 12.6 M 0.0062 0.26 0.011 5 17.6 11.7 F 0.0059 0.18 0.010 6 24.6 12.9 M 0.0060 0.37 0.015 7 23.6 12.7 F 0.0061 0.41 0.017 8 19.2 12.0 F 0.0060 0.29 0.015 9 22.9 12.6 M 0.0061 0.31 0.014 10 20.2 12.3 F 0.0058 0.27 0.013 11 19.4 12.1 M 0.0059 0.30 0.015 12 18.5 11.8 F 0.0061 \ 0.25 0.014 162 June 21st 2000 Condition Hepatosomatic Weight Length Sex Factor Liver Wt Index Tank No. 229 1 (2000mg 4NP/kg) 20.9 12.3 M 0.0060 0.27 0.013 2 19.8 12.1 F 0.0060 0.23 0.012 3 22.8 13.0 M 0.0055 0.24 0.011 4 21.7 12.3 M 0.0062 0.30 0.014 5 17.2 11.5 F 0.0061 0.28 0.016 6 22.6 12.9 F 0.0056 0.26 0.012 7 19.8 12.3 F 0.0057 0.15 0.008 8 24.6 12.8 F 0.0062 0.33 0.013 9 20.1 12.2 M 0.0059 0.29 0.014 10 17.0 11.4 F 0.0062 0.35 0.021 11 24.5 11.3 M 0.0093 0.25 0.010 12 21.1 12.4 F 0.0059 0.25 0.012 Tank No. 230 1 (20mg 4NP/kg) 18.7 11.9 F 0.0060 0.16 0.009 2 24.8 13.1 F 0.0058 0.27 0.011 3 21.3 12.5 F 0.0058 0.19 0.009 4 26.0 13.3 M 0.0058 0.32 0.012 5 24.8 13.0 F 0.0059 0.31 0.013 6 18.9 12.0 F 0.0059 0.26 0.014 7 18.8 11.6 F 0.0065 0.27 0.014 8 22.9 12.3 M 0.0066 0.25 0.011 9 19.1 12.1 M 0.0058 0.16 0.008 10 23.0 12.7 F 0.0059 0.28 0.012 11 20.2 12.3 F 0.0058 0.25 0.012 12 20.6 12.4 F 0.0058 0.26 0.013 Tank No. 231 1 (0.2mg 4NP/kg) 22.0 12.5 F 0.0060 0.31 0.014 2 25.6 13.2 F 0.0058 0.36 0.014 3 19.1 11.6 F 0.0066 0.23 0.012 4 20.2 12.3 F 0.0058 0.23 0.011 5 24.1 13.0 F 0.0058 0.33 0.014 6 19.7 12.3 F 0.0057 0.22 0.011 7 22.6 12.4 F 0.0063 0.30 0.013 8 20.9 12.4 M 0.0058 0.25 0.012 9 18.3 11.9 M 0.0058 0.18 0.010 10 20.4 12.4 M 0.0057 0.26 0.013 11 21.1 12.7 F 0.0055 0.25 0.012 12 25.6 13.4 F 0.0056 0.36 0.014 163 June 21st 2000 Condition Hepatosomatic Weight Length Sex Factor Liver Wt Index Tank No. 232 1 (0.002mg 4NP/kg) 20.5 12.2 F 0.0060 0.22 0.011 2 21.5 12.3 M 0.0062 0.27 0.013 3 20.2 12.4 F 0.0056 0.28 0.014 4 19.4 11.9 M 0.0062 0.26 0.013 5 21.0 12.3 M 0.0060 0.29 0.014 6 20.3 12.3 F 0.0058 0.30 0.015 7 26.5 13.2 M 0.0060 0.36 0.014 8 20.9 12.5 M 0.0057 0.25 0.012 9 19.5 12.0 F 0.0061 0.29 0.015 10 21.1 12.5 M 0.0057 0.21 0.010 11 25.2 13.0 M 0.0060 0.36 0.014 12 27.7 13.5 F 0.0059 0.44 0.016 Tank No. 233 1 (E 2 31 mg/kg) 23.2 12.9 F 0.0057 0.60 0.026 2 27.2 13.4 F 0.0059 0.95 0.035 3 26.7 13.0 F 0.0064 1.00 0.037 4 28.5 13.8 F 0.0056 0.74 0.026 5 18.9 12.0 M 0.0059 0.69 0.037 6 29.5 13.8 F 0.0058 0.80 0.027 7 23.2 12.9 F 0.0057 0.72 0.031 8 26.4 13.2 M 0.0060 0.80 0.030 9 22.0 12.5 M 0.0060 0.81 0.037 10 19.6 13.9 M 0.0038 0.78 0.040 11 26.3 13.4 F 0.0057 0.81 0.031 12 22.5 12.8 F 0.0057 0.67 0.030 Tank No. 234 1 (Control) 18.2 11.6 M 0.0063 0.29 0.016 2 22.6 12.7 M 0.0058 0.31 0.014 3 23.1 12.6 F 0.0061 0.33 0.014 4 18.1 12.0 M 0.0056 0.35 0.019 5 17.8 11.8 M 0.0058 0.17 0.010 6 19.2 12.0 F 0.0060 0.21 0.011 7 20.1 12.1 F 0.0061 0.30 0.015 8 19.8 11.8 M 0.0065 0.32 0.016 9 21.2 12.8 F 0.0053 0.32 0.015 10 20.8 12.5 F 0.0057 0.23 0.011 11 21.5 11.9 F 0.0069 0.36 0.017 12 21.3 12.5 F 0.0058 0.30 0.014 164 September 12th 2000 Condition Hepatosomatic Weight Length Sex Factor Liver Wt Index Tank No. 223 1 (2000mg 4NP/kg) 72.9 18.9 F 0.0052 1.09 0.015 2 77.4 18.9 F 0.0055 1.02 0.013 3 78.9 18.9 F 0.0056 1.21 0.015 4 62.2 17.9 F 0.0053 0.98 0.016 5 48.5 15.9 F 0.0060 0.6 0.012 6 86.2 19.9 F 0.0052 1.26 0.015 7 52.7 16.4 F 0.0059 1.06 0.020 8 73.7 18.6 M 0.0055 1.17 0.016 9 70.5 18.3 F 0.0056 0.98 0.014 10 74.9 18.6 M 0.0056 0.98 0.013 11 75.1 18.4 F 0.0058 1.22 0.016 12 93.8 19.9 M 0.0056 1.48 0.016 Tank No. 224 1 (E 2 31 mg/kg) 70.0 17.7 M 0.0062 1.70 0.024 2 107.1 21.1 M 0.0053 1.27 0.012 3 68.7 17.6 F 0.0062 1.91 0.028 4 55.1 16.4 F 0.0062 0.94 0.017 5 46.8 15.0 M 0.0070 1.05 0.022 6 72.6 17.8 M 0.0063 1.34 0.018 7 64.3 17.8 M 0.0056 0.98 0.015 8 80.0 18.7 F 0.0059 1.22 0.015 9 47.9 15.0 M 0.0072 0.75 0.016 10 79.5 18.9 M 0.0056 1.33 0.017 11 72.7 18.3 F 0.0057 1.11 0.015 12 60.6 17.2 F 0.0058 0.94 0.016 Tank No. 225 1 (0.2mg 4NP/kg) 70.1 18.8 M 0.0051 1.21 0.017 2 79.7 18.3 F 0.0063 1.14 0.014 3 67.3 17.7 F 0.0059 1.01 0.015 4 69.7 18.1 M 0.0057 0.80 0.011 5 38.8 18.6 F 0.0029 1.00 0.026 6 84.3 19.2 F 0.0057 1.42 0.017 7 51.0 16.0 M 0.0062 0.90 0.018 8 66.9 17.4 F 0.0062 0.91 0.014 9 69.3 18.1 M 0.0057 1.12 0.016 10 71.3 17.9 M 0.0060 0.84 0.012 11 85.7 19.1 M 0.0059 1.36 0.016 12 80.9 18.9 F 0.0057 1.13 0.014 165 September 12 t h 2000 Condition Hepatosomatic Weight Length Sex Factor Liver Wt Index Tank No. 226 1 (20mg 4NP/kg) 67.1 18.0 F 0.0056 1.01 0.015 2 64.5 17.9 M 0.0055 0.92 0.014 3 94.1 19.7 M 0.0058 1.55 0.016 4 66.5 17.9 F 0.0056 0.98 0.015 5 64.5 18.4 F 0.0050 1.10 0.017 6 92.8 19.9 F 0.0056 1.15 0.012 7 78.4 18.9 F 0.0056 1.35 0.017 8 74.9 18.5 F 0.0057 1.16 0.015 9 83.3 19.4 F 0.0054 1.30 0.016 10 114.0 20.8 M 0.0059 1.48 0.013 11 93.5 19.7 M 0.0058 1.25 0.013 12 100.4 20.6 F 0.0054 1.32 0.013 Tank No. 227 1 (0.002mg 4NP/kg) 69.6 18.3 M 0.0055 1.17 0.017 2 70.3 18.0 M 0.0059 1.12 0.016 3 57.4 16.9 F 0.0059 0.88 0.015 4 99.1 19.9 F 0.0060 2.12 0.021 5 72.3 18.4 F 0.0056 0.98 0.014 6 57.8 17.4 F 0.0054 0.87 0.015 7 86.2 19.3 F 0.0057 1.25 0.015 8 62.3 17.2 F 0.0060 0.97 0.016 9 56.9 17.0 M 0.0057 1.18 0.021 10 92.7 19.8 M 0.0057 1.54 0.017 11 66.5 17.8 F 0.0057 1.20 0.018 12 53.9 17.5 M 0.0049 0.77 0.014 Tank No. 228 1 (Control) 80.4 18.3 M 0.0063 1.62 0.020 2 63.4 17.5 F 0.0058 0.91 0.014 3 57.7 16.9 M 0.0059 1.03 0.018 4 72.5 18.5 M 0.0055 1.11 0.015 5 96.4 20.3 M 0.0054 1.31 0.014 6 100.1 20.7 F 0.0053 1.52 0.015 7 78.1 18.8 M 0.0056 1.09 0.014 8 74.5 18.2 F 0.0060 1.07 0.014 9 117.3 21.2 M 0.0057 1.82 0.016 10 70.0 18.2 F 0.0056 0.96 0.014 11 63.3 17.6 M 0.0057 0.96 • 0.015 12 104.1 20.5 M 0.0057 1.42 0.014 166 September 12th 2000 Condition Hepatosomatic Weight Length Sex Factor Liver Wt Index Tank No. 229 1 (2000mg 4NP/kg) 54.4 16.6 F 0.0059 0.74 0.014 2 51.1 16.8 M 0.0053 0.81 0.016 3 102.0 19.4 M 0.0067 1.82 0.018 4 90.4 19.7 F 0.0056 1.43 0.016 5 70.6 17.7 F 0.0062 1.19 0.017 6 76.6 18.7 F 0.0056 1.14 0.015 7 64.1 17.5 M 0.0058 1.22 0.019 8 64.7 18.2 M 0.0052 0.95 0.015 9 66.1 17.7 F 0.0058 1.35 0.020 10 76.6 18.2 F 0.0062 1.28 0.017 11 93.1 19.7 M 0.0058 1.64 0.018 12 72.6 17.9 M 0.0062 1.30 0.018 Tank No. 230 1 (20mg 4NP/kg) 65.0 17.7 F 0.0057 1.33 0.020 2 82.5 18.9 M 0.0059 1.55 0.019 3 71.1 18.2 F 0.0057 1.08 0.015 4 79.2 18.7 M 0.0058 1.12 0.014 5 104.3 20.5 M 0.0057 1.87 0.018 6 65.1 18.1 F 0.0053 0.92 0.014 7 74.4 19.3 M 0.0049 1.07 0.014 8 54.9 16.9 F 0.0056 0.93 0.017 9 118.0 21.0 M 0.0060 1.82 0.015 10 82.0 19.1 F 0.0056 1.50 0.018 11 79.7 19.2 F 0.0054 1.42 0.018 12 78.2 18.7 M 0.0058 1.12 0.014 Tank No. 231 1 (0.2mg 4NP/kg) 81.0 18.7 F 0.0060 1.37 0.017 2 67.5 18.4 M 0.0052 0.95 0.014 3 61.4 17.3 M 0.0058 0.72 0.012 4 62.9 17.2 M 0.0061 1.04 0.017 5 68.6 18.3 M 0.0054 0.90 0.013 6 84.2 19.7 M 0.0052 1.20 0.014 7 81.5 19.4 M 0.0053 1.20 0.015 8 107.1 20.9 F 0.0055 1.59 0.015 9 86.9 19.3 M 0.0058 1.12 0.013 10 85.3 19.6 M 0.0054 1.25 0.015 11 98.2 20.4 M 0.0054 1.74 0.018 12 64.3 19.1 F 0.0044 0.66 0.010 167 September 12th 2000 Condition Hepatosomatic Weight Length Sex Factor Liver Wt Index Tank No. 232 1 (0.002mg 4NP/kg) 74.7 18.4 M 0.0058 1.26 0.017 2 69.4 18.1 M 0.0057 1.16 0.017 3 90.0 19.7 F 0.0056 1.27 0.014 4 91.6 19.8 M 0.0056 1.37 0.015 5 81.5 19.1 M 0.0056 1.13 0.014 6 71.0 18.9 M 0.0050 1.24 0.017 7 75.8 18.7 M 0.0056 1.13 0.015 8 74.1 19.1 M 0.0051 0.89 0.012 9 80.6 18.7 M 0.0059 1.22 0.015 10 84.1 18.8 F 0.0061 1.15 0.014 11 75.1 18.3 F 0.0059 0.97 0.013 12 80.7 18.6 M 0.0060 1.38 0.017 Tank No. 233 1 (E 2 31 mg/kg) 80.6 18.9 M 0.0057 1.13 0.014 2 76.6 19.3 M 0.0051 1.13 0.015 3 60.4 17.6 M 0.0054 0.98 0.016 4 68.7 18.3 M 0.0054 0.96 0.014 5 54.0 17.1 F 0.0053 0.79 0.015 6 57.8 16.5 F 0.0064 1.41 0.024 7 57.6 17.2 M 0.0056 0.87 0.015 8 74.5 18.6 F 0.0056 1.32 0.018 9 70.2 17.9 F 0.0060 1.16 0.017 10 73.4 18.4 M 0.0057 1.12 0.015 11 83.5 18.5 F 0.0064 1.76 0.021 12 72.9 18.6 F 0.0055 1.23 0.017 Tank No. 234 1 (Control) 78.8 18.6 F 0.0059 1.10 0.014 2 77.7 18.8 M 0.0056 0.87 0.011 3 59.1 17.4 F 0.0055 0.75 0.013 4 63.8 17.5 F 0.0058 0.85 0.013 5 65.5 18.2 M 0.0053 0.84 0.013 6 73.4 18.2 F 0.0059 1.16 0.016 7 77.6 18.4 F 0.0060 1.06 0.014 8 69.4 18.1 M 0.0057 0.94 0.014 9 75.6 18.6 M 0.0057 1.10 0.015 10 58.2 17.0 F 0.0058 0.88 0.015 11 89.4 19.7 M 0.0056 1.17 0.013 12 95.4 19.8 M 0.0058 1.24 0.013 168 CO LU r -< X o ft CL > a: < co (fl "o c (D -C _c O. (fl L_ o £ <F § Q O ~ Q. E ro CO CD O c n co CD CD CD oo • t f CN CD O) 1^ CO h~ CO CO CO -tf o 1^ CN CD CD IS-Is- 00 CO CN ID CD CN CN CN O • t f IS-CO CN CO Is-•tf CO r~- Is-Q. E o O . ro Q CN Is-CN CN CD CO CD CD ID IS-"<- 00 CD • t f CD co c n CN in •tf co CN o -tf co co co CD CO CN CN CO O CO • t f CO CN CN co Is-CD T -iri cb r^ - Is-CN Cl. E o o . CO Q 00 CO co -tf I s-' CT) CD Is- c o in CT) cb •tf oo CT) -tf co in m o CN CN in CN in o in in CN CN C D oo •tf IS-co m 00 CT) •tf - t f IS- IS-co Q. E o ro Q o co T - Is-od ob CT) CD o O CN CN CO I s-' cb IS-CD CO Is- Is-o CN CO CD CO CN O in CN CN O T -Is- in CO -tf" CN CN o CT) in IS-O CT) CO •tf c b in Is- Is-• t f Q. E o o . ro Q CD CO o in i^ cd CD CT) CO CO CT) CO in cb CO CT) CN i - in m co CN co o •tf CN CN CO co •tf CN CN CO CT) CO CN CO CO IS- T -c d -tf CN CN Is-o cb Is-Is- Is-CN 00 cb in Is- Is-in Q_ E o o . ro Q 0 0 in 3 °> ~ CT) CO O CN CN CO CO CN CD CN O CO co Is-CO CN 00 00 CN CO •tf co m cci CN CN Is- co •<- •tf CN Is- Is-D) xi CL z •* o o -o CM co CM CN XL C n H co g CD ° CD IS-IS-cb I s -m o o in co CN m co CN CN CD CD IS-cb CN CN in CD CN Is-oo T -o CN CN co co CM CN •tf •<-co CT) CD CT) CT) CD CT) in O IS-CN IS-cb in o cb CT) CD CN in cb m CN CN co CN CM CN m CM I s - co in co •tf in CN CN co CN •tf co in -tf i— Is-CO ; CM CM •<- CM o o d o o •tf CN CD - t f IS- CO 00 CO cb cb 00 CO CO in cb CN o CD CD - t f cb cb CN CN Is-•tf CO Is-o -tf •tf in CO CO Is- r*-co = CM CM CT) CN CN CO d d o o in CT) 00 O CO ob CD CD CD CT) CN CN m CM CM CO IS- - t f O CN cb cb CN CN •tf co CO Is-CO CD CD IS-CO CO IS- IS-in I CO CN CN o OO CD c n c o o i CT) CT) CM CN CO t-O -tf IS-' IS-' 00 CD in CD O m oo in in • t f CM CN CT) CD •>- CN CN CN CT) CN o m cb iri CN CN o CN • t f IS-CO O) -tf CO -tf IS- IS-O ) XI E CM CM x: c ro IS- -tf m CM CO h-' CD O) in CT> CM CD i^ CM CN CD cb -tf o o o Is-•tf c o CM o ) r>-cb r-^ CN CN c o co CN O) CO O CN CO CM • t f CD 00 CN CO CD CD O) CO cb •>- -tf r^- co cb cb •tf co cb •tf •tf m cb CD CT) CD CN CN co •tf cb CN in T -m -tf cb cb CN CN CN in co in CD •tf m c d co CO o o co cb •<- -tf CO CD cb cb CO CN in CN col CM CO co CM CT) CD cb iri CN CN o co co oo CN co CO -tf • t f = CN CM 169 111 X o CL > < o c 0) - C JZ _Q_ (0 >» LL o £ T " I -3 Q _Q> — Q, E CO C/> — 1 0 T * re Ln S 1° § s o co in CD co CN i^ co CD CD o cb cci T - o T f CO CN CO CD 00 CD CN CN CD 00 ib CN CN co 1^ co oo O v -T f CN z Tf G) E CN : o LO CM CM C re CN T f CN CD O t-O O CO CO T f oo CO CD 1^ 1^ T t CD co o co cn 1 -oo to co CN CD LO CD CO CD LO CN CN LO oo co 1^ T f lO co co CO Tf' CN • LO = CN CN 1^ LO CD LO d t-o o CO CD O CO T f CD oo h-CN CD cb cb CD CN CN r- CN CD CD •<- csi O T f CO T f CD 1^ CSI CN co O CO h- LO CO Csi CO I LO = CN CN CSI T f CD 00 _ CN oo ° CD r» LO 1^ T f LO CD CN CO LO CD CN T -1^ CN CO 00 CD CO i^ CSI CSI CD T f csi h- CN co co csi csi LO CD CD CN 00 CD CD CD oo oo CN O T f 00 LO LO CSI T f csi oo CD csi csi CD O T f 00 i^ cb CN CN T f O LO CN csi co LO I LO CSI CN 1 - h-T - T f CD CD CD CD T f CO cb T f lO T f CN cb cb CO i -CD h-to ib T f — T f 00 csi csi CD i -CO CO ib ib CN CSI CD CD CD T f T f O) iC CL z Tf cn E o ; to CSI CN c re H co 0 0 £ CD CD °. CD CN CN CO cb cb LO 1^ cb T f h -LO CD cb cb co T f csi LO CO lO csi csi CO CO T f iq ib ib CSI CSI T t LO T f T f T f Csi I CO CN CN £ 1 -^ 00 o °> °- CD LO CD T - CD LO i^ CO CO 00 CSI cb cb CD 1^ -00 CN CN CN 00 1^ cb CN t- LO T - T f 1^ CD CN CN CN CN CO 1^ CD LO 00 LO CN CO co I CD = CN CN CD co CD '<-> CD h~ co CD CO cb cb T f CO T - T f i^ cb 00 CD CD CO CN csi CO T f r-~ CD O ib cb CN CN CO CO O CD T f CO CD r CN CN T f CD 00 00 CD CD CD CD o cb T f 00 CD O ib cb T f CD cb O h-r- LO cb cb oo co 00 00 CN T - ' LO CN 00 CSI CO CD ib T f CN csi LO 00 T f 1^ CSI 00 CO CO T f ib CO = CN CS| 1^ T f LO T f d d o o o o LO CO cb cb LO CN CN CN O T f oo r-CO CD 00 1^ CN Csi CSI CO CD O csi D) Tf cn E CM O : o CN CM iC c re H LO T f cb CN CSI cb 00 CD CO o cb cb 00 Csi T f T -00 CD csi csi h- CO h- CD T f T f CSI CN LO co CSI o ib ib LO m . CD ° . CSI T - CO CD LO CN CD ib T f CD LO CN ib cb LO CO CN OO CN T - LO csi csi CSI CO T - o cb cb CN CN oo i^~ 00 CD co co i^ = CSI csi 00 CSI CD CD CN CO CN h-T f T f O LO CD O csi csi o CSI ib CS| CD CD o T f ib T f CN CN o oo T f T - o O CD T f ib r~- i^ z CN CN 170 CO LU r-< X o a: Q_ >•' < CO to o c 0 _Q. CO >» LL <F § Q a> — a. E ro CO 1^  CSI cb oi CSI CD o o cb cb co co csi CO CO co CN ib CN LO LO CO TJ CN CN O) > CN 1^  < T f h~ CO CD LO o CD T f S CO ib 1-^  1^  c o o CO CM CM iC c re I -CD CD CD co CD CD CD CD CO O 00 cb CD CO O CD ib co CN T-LO CO csi CN T f CD CD f ib ib CN CSI CD T f CO LO T f ' T f 1^- h~ CO CN CO CD CD CO CO CN CO LO T- Tf 1^-' 1^  00 CO •<- o ib ib CS| Tf csi CS| csi CO CN CN Csi CO CO ib CN 1^  o CO o ib ib CN CN CD T f o o T f ib 1^-co = CN CN P T ? CN o 1^  T f I--' CN CN CN CO CN CO LO CN CO CO CO CO CN csi CO ib CM LO CO CN CN T - CO h- oo T f CO 00 CM CN CO °2 CN ^ CO CO co cb CN CO T f CO cb Tf cb 00 CO CD co cb CO CO CO o CN csi T f CO ib CN 1^  r~- co cb co CN CN CO CO Tf co co CN O co cb io i oo CN CN O 1^-i ^ r-~ cb LO i ^ cb Tf co CD oo cb cb LO co h~ CD LO CO Tf r-cb cb CN CN LO CN CO CO D) Tt O) E o O : O CM, cn CM CM JC c re oo CD ^ IO LO CO CO CO h-ib ib •<- CM oo ib ib csi oo T- CN 00 00 csi csi T f CM LO r^ -LO CSI ib TT CM CSI CO o ib io co T f i ^ T f ib h- i ^ CO : CN CSI CD 00 CO 00 i ^ cb CD CN cb CN CO CM cb cb m Tf csi o co CO csi csi Tf o oo cn cb ib CN CN co o CO T f 00 CD 00 Tf CO CO CO CD cb CD LO O CM cb cb co o cb CO o CO CO ib cb o Tf •>- csi o i ^ T f CN T f CD CO T f cd ib CN CSI o co ib CD Tf o LO cb Tf i ^ r-1" CO = CM CN CO CN CO CO T f CO CO CO T f CN T f CD CD CO CM CD CD CN CO csi csi LO CO ib CN 00 CO CSI o cb ib CN CN LO CO T f CSI CO CO T f LO I CO ; CN CN LO CO LO T f CO CO CD T f CD ib cb CO oo T f T f csi csi co CN oo csi CO CD CN CN CO 1^-CO CN CD oo CO CO o = o CO CM X. c re CD CO 00 LO CD CD CD CD CO 00 CD CM CD K O T f oo r-~ ib ib co csi csi CD r^ . T f co csi i^. CO ib csi CN CN cq -TT ib ib CN CN co co T f oo oo CD LO T f T f i^- h-CN I o = CO CM CO 00 CD T f 00 T f 00 00 CO CD O CN r-^  ib CN CN O T f co csi csi 1^  IO CS| CD T f 00 CN cb oo CN CN CO T f csi 1^-CD T- h-cb 1^- 1^-CO I o CO r- T f I CD CN| d o o I LO o oo I cb CO CD CD lO T f 001 cb cb T f T f co LO CM csi T f o ib CM co CD LO T f CN CN CD CD T f T f CD 00 O T t ib O ; CO CN 171 CO LU X o ft Q_ > ft < CO sz o c OJ sz sz >> Li_ o ® <F § Q © ~ a. E CO CO CD O o o d o o CO CO CO o CD ? CD CD CM co m CD LO CN in in co CN CN CD co in •tf CD iri •tf CN CN c o • t f IS-I s- in m o •tf iri I s- I s-2C •tf £ CN : CO CN XI c <9 3 d o CD co co CO CD • t f CN CO 1^  m in -tf CO CD o CN CN in m CN T -CN CN CO m iri CN CN m I s - co iri iri CN CN oo in CN CO • t f -tf" IS- IS-CN m CD CO IS-d d o o m co co •tf co t- in CO CD I s-co CO CD CO IS-CO CD CD in CN • t f CO * - o CN CO CD o iri CN I s- CN m CD •tf iri CN CN co oo •tf co iri •tf IS- IS-00 CO CN CO IS-•tf d d o o I s-CD CN CN CD I s-1^  Is-' CD CO CO T- 00 • t f CO CD CD CO CO CD "tf CD • t f CO CN CO o 00 1-iri i^ CN CN in CO I s-CN O CD • t f CN IS- IS-C0 CO IS- in d d o o in c o in CO IS-m I s-00 CD -^' 1^  CN o CN O •tf CN CO CN CO - t f CO iri co CN CN CN CD 00 • t f CO IS- IS-co CN I s- O O CO d d o o •tf o CN CD CO CO CD in co iri iri co co CN I s- o CD I s-CN CN CD CD CD CO CD CD CN CN •tf - t f CO CD CO CO IS- IS-O) cu z •tf O) CN O : CN CO CM XL c co CD in I s- CT) CO CT) CT) CD OO CN in •tf co co oo in •tf o CD iri 00 • t f CD co CN CN CD CD iri CN m "tf •tf in co iri CN CN • t f IS-m co in •tf CO • t f IS- IS-CN i CN CO •tf O CT) CD CD CO co in in •<-cci IS-CN m CD CD IS-co CN o in co in CN CN IS-•tf IS- iri CN CN CD • t f CO IS-m co •tf in CN • t f IS- IS-CO CD co T-o o CD IS-O CT) 1^  CD CD CN CD CD CD CO CN CD CD CN CN CN • t f • t f CN T- CO •tf - t f •tf "tf CN CN CT) I s-m m iri iri I s- I s-t CN : CO CN • t f CO CO in CD CD CN CO CD d f—] O) oi CT) CD CT) CD CD 00 CN •tf CO CD CT) IS-' CD CT) CN iri CO CT) CT) in iri IS- oo i - oo c o co • t f CN o IS-• t f CN •tf iri CN CN in IS-o co CD I s-iri •tf I s- I s-"53 xi E T— in co • CM CN = CO CN CO CO CM c re H CD CD • t f CO CN iri o I s-CN •tf m •<- CN CN CO in CN I s- CD CO • t f iri CD CN CN CT) o • t f IS-C0 - t f CD m • t f CO I s- I s-c o CO co IS- 00 • t f IS-CO CD o CN CD CD -tf CN CD CD • t f CN CO CO CN CN • t f CO CT) CD iri CN CN CN IS-00 IS-00 CD CO O CO -tf' I s- I s-00 o m oo d d o o CN oo CN • t f CN CO ^ CO CO CD i - in CN in CO CD o CD CN CD co in CN CN co in co m CD iri CN CN o CD CO IS-"tf m CO - t f CO -tf I s- I s-I s- CT) CO CO CT) CD CD O) CT) CD CN CO CO 1^  • t f CN O O CO O CO •<- in i - CD CO CN O in CD CN I s- co m •tf CD CD CN CN o in co I s-co I s-•tf in co co I s - r--7 CO = CO CN 172 CO 111 X o CL < 3 CO co o c (D SZ SZ _Q_ CO >* LL o <u T " § ^ -) Q 0) ~" Q. E ro CO o CO CD LO CO CO T— cb o> T— o o T— T— LO CO CN ? LO LO co CN CO LO O 00 00 o oo CN CO CO LO CN co LO CD LO LO CN CN CO CN T t 1^-CN CO T t O T t T f 1^  c o o Tt ro CM J£ C n 00 LO OO 00 K CN CO LO CO LO CD CN CO CN cb CN 1^  T t LO CN T f O 00 ib LO CM CN CO LO T t CD O T- CD T t T f CN I T t CO CM -r- CM CD O cd co CD CD CD CD r>- CM T t LO ib cb CO CO CM cq ib ib LO T t csi CD CN T t T f CN CM CD LO LO CN T t ib CN CN LO T t l-~ ib T t T t = co CM CD T t 00 CO CD CD CD CD T t CD CD CO T t CM CD CD CO CD CO CO CM ib cb co in csi CM CM cb CN 00 T f o co cb cb CM CM CD CO 1^-CM CO CD CO CO CO T t CO CM oo oo CD O CO 1^-' CD CO O 00 CD T t 00 N-cb cb CD O T t CO ib ib o o CD CM COl CM lO CD CN in CD CO o T f 00 CN CM LO O CD I in 1-h- I 173 CO LU H < X o ft CL > ft < CO L L _ C — </) O i l ^ F LT§ o -5. co Q O ~ Q. E ro CO T - IS-00 00 CD CD O O CO CO c b cb o co O I s -CD • t f LO CN T - CO O CD LO LO CN CN LO • t f I s -• t f " t f I s - I s -CT) CL z •tf CT) O o co CM CM • X c re CJ) 0 CO CM CJ) o LO cb o oo •tf co cb r- : c o CM o ) co CM c \ i CM o LO CM CO CO CD O • t f LO CN CN co CD • t f IS-CN - t f O CD If) •tf I s - I s -CJ) LO CO d -<-o o " t f co cb CN CD I s - CD cb cb CD o IS- LO CN 00 ^ cb Is-o CN CO CO CJ) CN LO CD • t f CN CD LO CM CO LO "tf CN CN LO o LO I s -• t f LO I s - CO •tf ib I s - I s -C0 I CO ; CM CM CD -tf CJ) CD cci i ^ CJ) CD 00 co LO LO CO - t f CJ) •tf I s - CD LO cb 00 o CJ) LO CN CN •tf cb CN CM O CO CD I s -LO CO I s - I s -co CM CM CO LO • t f T -t-^  d o o CO i -IS- CN LO ib IS-CO cb •>- 'tf-LO CM cb cb LO o CN IS- CO o o CM CM CO CN CD CN oo co CN CO CN CN CD 00 CD I s -00 CD I s - CD I s - I s -LO i CO CN CN LO O CO LO CO CO CD CJ) CO CO CO CJ) CN -tf cb I s-' CO cb CD CD T - O cb cb co CO o Is-CN CO CO CD I s -Is-: cb CN CN I s - CN CO CN CN CO I s - I s -E CO CM LU = ^ > CM CM J£ C re CD CD CJ) • t f CD CD • t f CO CD CD CO CO CO CD CD 00 CO OO o c b IS-CO CN I s - o I s - LO •^ CM • t f O O LO LO cb CN CN CD O CD CO • t f CO IS- IS-CD T -LO IS-d d o o O CD O CD Is-: cb CD CD CD T -cb N : IS- - t f CD CO LO CN CO CO O CN LO i b CN CN LO 00 • t f IS-I s - -tf CD I s -'tf "tf I s - I s -CM i • t f CN CN CO IS-LO • t f d d o o •tf CN CO O 00 i - CO cb cb o o CD I s -co o I s - CD • t f LO • t f CN • t f LO CN 00 • t f • t f CN CN CD •tf LO I s -CO LO I s - T -LO i b I s - I s -7 •tf = CN CN CO I s -CO CO CN LO CO I s -o CO I s -O) d Q CD CO O) CD CJ) CD CD CN CN IS-' CN CO CJ) LO CO ^ • t f CD CN CO LO 00 LO • t f O CO o CN CD CN LO LO • t f CD CD LO CN CN LO LO LO O CO - t f I s - I s -• t f : CN CN CD CN CD CO O) CJ) LO l b CO O Is-cb LO "tf CO LO cb c b co co CM CO <J) LO CM CN LO o i b CN CO IS-CD c o c b CN CN LO O) •tf I s -I s - CO CO o cb cb I s - I s -CT) XL • t f CT) E CM = o l O CM CM x: c re CO CO CO CD I s -CN c b LO CD CO CD cb IS-IS-CD LO CD CO IS-O) CO CN CO CN CO CO O) CO CN CN CO o I s - c o CD CO d T -I s - I s -lO = CN CN •tf -tf O t-d d o o CO -tf 00 N-•tf - t f I s - CO LO 1 -cb cb CO LO CN CN O o CO CM CO CD CO CN CO CO • t f CO "tf CM CM CN I s - I s -LO T -i b ^ ' I s - I s -LO CN CD LO CO LO d d o o CN O CN CD c b i b LO CD cb I s -IS- IS-o co CM CM CN 00 -tf -tf CN CN CO 00 CD T -i b i b I s - I s -LO ; CN CN LO CN CN 174 co UJ x o CL > < CO co LL x: — co o i l o -5. ^ CO Q J3> — Q. E ro CO |v. CD CO CO o i CO co co CD LO CO CO LO LO T t T f CD CO CO o CD cb CO LO CO T f CN CN O CD CO CD CO CO CSI CN O T t CN CO cb c b I s- I s-JL T f o CO CM CM JL c re co o T f CO cb i ^ CO o> LO T t O CO i b T t LO T t cb CO CO LO CO cb cb h- CO co co csi IS-T f i b LO CN CN co csi LO h-T t I s- I s-CN i CD CN CN T t CN T - CD CO CO 00 LO O O CN LO CO LO CN T t O T t CO CO CN CO O i b CN CO CO T t LO i b T t CN CN I s-C0 T t I s-T t T -LO T t T t i b I s- h~ co I CO CN CN LO CO O CO CO 00 co co oo C O i b 00 CO CO CO i b i b CD 00 CO CO O CO cb i b T t o c o CO o c o T t CN 00 CD LO T t T t CN CN O T t i b I s-CO CN CO T f i b i b I s- I s-Y CO : CN CN CN ° CO °? CO § co E£ CO LO CO 00 T t T f CO o O CO cb i b 00 00 T t CO csi T ~ T t CO CO CN LO CO CSI T f T t csi CSI CN LO I s-I s- LO i b Is-' I s- I s-CD = CN CN I s-CO LO cb oo CO CO CO CO I s- o T t cb LO CO 00 CSI i b cb r-- co T - CN csi csi T t o LO cb CN CN CN CD CN CO T t CO f - I s-CO JL T f O) E CM O : o I s -CM CM JL c re CO o co d o o o CD CO CO T f T t I s-CO CO CM cb CD LO CO o cb cb CN CO 00 CD CO 00 CD I s-d CN I s- T f I s- I s-d d CN CN T t CN d I s-C0 CD CN CSI C0 CO I s- I s-IO CN CO 00 O) LO CD 1^ LO CO cb CM CO T t CSI cb cb CD T f T - 00 CO T f I s- CD i ^ i b CN CSI CO CM CO I s -CSI CO CN CO Csi T f I s- I s-I s- ; CN CN I s- LO cp CO cd d CO CO CN CN co oo T t i b CD CD CO CO CM CO T t CM CO LO T t T -T t i b CM CM co T - LO LO 00 ib T t I s- I s-00 CD cq I s-d ib CO CO I s-T t i b T t CO O 00 1^ CO LO 00 i b CM 00 CO I s-co 00 CD CN IS- o CM T t co i b CN LO CO co a> i b i b CSI CM CO CO T t I s-LO I s-co o T t T t I s- I s-00 CM LO CM Is-: CO CO LO 00 LO I s--^' CO T f 00 00 CO CM CO I s-Csi CN I s-CN Is-: CM CD 00 CT> LO c b CM CN T t CN O T t co csi I s- I s-c o o 00 CM CM JL c re CO CO 00 00 i b Is-: co co I s-o CO CD cq I s-Is-: cb CN cb LO CD CO T f i b cb co csi T t CM csi LO LO ob CM CD CO I  IO d d CM CM LO T t d I s-T t I s-T t T t d d I s- i— CO I s-a> CM r--: cb CO O) T t CM i b LO co T t o I s-T t i b CM co o T t cb 00 T t o co csi csi I s-T t CM CO CO T t T -T t i b CM CM CM ib I s-T t I s-LO 00 ib T t I s- I s-co I 00 ; CN CN CO IS-I s - r--d oo oo CD LO co CN I s-C0 LO T t CO co i b o CD CD i b i b 00 I s-co CD O) LO CN CM 00 o CN CO CM CM O) CM 00 CO 1^ CD I s- I s-oo CN CN 175 CO UJ X o Q. >•" < CO CO Lu J Z — (/) g i l ^ co Q 0) — a E ro CO S . 8 o o CO h -T f L O T t CO o o c b r - CN C D O i b cb CO C D LO CN T - ' T t CD CN CN C N L O C D C O csi csi CN CSI C O C O C O L O O C D 1^ 1^ O) E o o : o CN^  cn CM CM c n CD § C D E£ CD 1^  cb T f L O co a ) c b cb T t CD L O L O co C O C D 00 CO LO oo 00 00 i o h -T t CO CM CM C M oo i b CM CN Tt CN l b CD i ^ h -£ 3 00 . CD § CD 0 oo o cb T ~ T t 1^ - T f i b cb oo oo T t h- o i b cb i - co csi csi c o C M C O CD h- o C O T f CN CM CD O C O 1^ r-- r -C N CD cb i b co • CD ; C N C N O C O 00 C D C D C D C D 00 CO CM T t cb T t oo T t T t CD O i b cb C D T f r~- CD o o T t CN C D C D T f CO CN CM CJ) T -CD O i b cb 7 C D CN CN h~ 00 O CN d d o o CD i b C D CO T - CM i b i b C O 1^ -L O 00 cb cb C O Csi h - LO C O C O Csi CN 1^ CO T f CO T t CM CM C D 1^ oo co CO LO cb i b 1^ h~ LO I C D C N C N C O CO h~ CD CD CD T t O CD T f cb i ^ CN i -CO LO cb cb C D I O C D T t C O T f oo cb C N C N LO T t CM CO CD CO LO CO U) E © : CN^ o CO CM c ro CD g CO P CD § CD ° CO CD - I - LO CD CD Tt Tt cb cb CO o csi C D O ) O C D C D C O C O C M 00 C D h~ LO CO CO CM CM C M T -C N T t cb cb CM I O ; co C N L O ° > • " . L O C J C D C D C D m L O C D l b C D C D cb CD h -cb cb CO CO T - CO CN CN CO CD i n CN o T t T t c b C N C N T t O T - , C O i b co i ^ 00 1^ LO T f oo C D C D T t C N T - | ^ C N C N 1^ 1^ T t C N i - o i b cb CD LO CM 1^ LO CN 00 Csi CN T f 00 O CD h-^ 1^ CN CN T t C O csi CO CN CD CO CM CM y O : co CM m g C M P C D § C D E! CN 1 -CO T -c b CO CD cb CN O T t C D cb cb C D C D o ) m T t 1^ cb CN 00 o T t O c b i ^ CM CM o o co c b L O I o co CM S o 00 00 CD CO CD C D CM LO O CD i ^ cb o oo CO CO O LO oo CN O 1^ 1^ CN CN CO oo csi CN CO CD csi C M i ^ cn JC E CM = o CO CM c <0 C D CO O T -cb cb C D C D 00 C D cb LO T -CD CO cb CN CD o in i b cb 00 00 C O i n t-T t T t 1^ 1^ CN CN 1^ -m csi 1^ -m C D LO LO C N csi CN co CN 00 O T -CD 00 I O 1^ -cb LO CD O T f i ^ cb T t O LO T -CM CM CO T t T f 1^ 00 CM CM 1^ -o csi N -oo r--m i n csi t-c o CM C O C D i n co o i r~-' C D C D LO CO o co i ^ cb o LO LO t-cb cb CN i ^ CM T -CD CN 00 CD CN i n c b CN CN CM CD o T t CO h~ I— 176 CO LU H < X o ft CL >^  ft < CO co LL SZ — oo " & o -5. co Q O ~ Q. E ro CO tal 1 CD I s -CM O 1 100.52 | CN I s -CO CN CO LO CO CD oo I s -CN • t f CN I s -o LO CD o m CD o CD CN c o CO CD I s -• t f CD •tf CD LO i n CO • t f CN oo LO LO • t f CD CD CO i n co I s -CN • t f 1 To cci CD cb CD 1 100.52 | o o CD CD CD CD o o CD CD 00 CD CD CD CD CD cb o O o CD CD CD CD CD CD CD CD CD CD 00 CD cb CD Is-: CD I s-: CD d o CD CD CD CD CD CD d o Is-: CD d o OJ > LO CO CO o CO CD o I s -CD LO I s - CO o LO CD CN o I s - CO CD CD CD I s -I s -CM 00 • t f < CO cb CO c d ^ CD Is-: cb i r i N : r-: cb cb |% Lipid CO CD CO CD CD o CM I s -O LO CO LO • t f LO CD CO CO LO • t f I s -LO I s -I s - • t f LO CD CN I s -CO x— co 00 CO CO CD •tf oo CD I s -CO CD CO CO •tf en LO CO CD I s -00 I s -I s -I s -CD |% Lipid CO CO i - : CO • t f CO CD cb h«: Is-: CD 00 Is-: cb Is-: CO ob 1^ i r i i r i I s-' cb cb cb cb cb i b cb cb cb Avg o CM LO CD CD I s -CN CN CO CO CO CD CD CD o 00 CO CD CO i n o CO LO CO CO I s -T— • t f • t f Avg LO LO LO cb LO i b cb cb i r i i r i c b i r i cb cb i r i Dtein LO LO CO CM CD CO I s -I s -00 o CO CO CO I s -I s -00 LO • t f CD o CD CD • t f i n CO CO CD CD 00 CO CO o CN CO i n CO CO I s -CN CN CD m T— CO o CO CO CD CO • t f CN I s -CD |%Pn LO • t f CD LO LO LO cb CD i b i b i b LO c b cb cb i r i i r i i b cb i r i i r i cb • t f CD cb cb i r i cb • t f cb > CD CD CO CO O I s - o CO I s -o CO CN V— CN CN CO CN CO CO o CO • t f LO CD co < T _ T _ CM Csi CN CM CM CN CN CM T ~ CN CN T _ Ash 00 CD CO CO I s -Is-; o o O CN LO CO c o CO CD •tf o CD • t f LO • t f CO c o CD m I s -CD •t f • t f co I s -CO CN CO t— 00 • t f I s -o CN o •tf I s -CN LO CD o CD • t f CO CD CN CO CD LO i n i n CN CO T— T— CM CN T _ CN CN cb CN CN T ~ CN CM CM CN CN CN CN CN CN CM , r " T _ Avg LO LO o CD O CO •tf I s - 00 • t f CO CN I s -CD o CO CO m o o CD CO CN CD LO CM LO Avg CM LO CN CM CN co CN cb CN CN • t f CN cb CN cb CN • t f CN 1^ CN cb CN CM CN • t f CN i r i CN S Q LO LO LO LO CM •tf CD I s -CD CN •tf CO CN CO I s - o CO CN o LO o 00 CD CN o I s -co I s -m •tf o CO LO •tf I s -CO CO i n I s -• t f CM CD CD •tf I s -CO CO I s -CD • t f CO i n CO 00 CN i n I s -CM I s-: CM LO CN LO CN CN CM CM CN c b CN •tf CN LO CN cb CN cb CN c b CN cb CN i n CN Is-: CN cb CN CN cb CN • t f CN • t f CM cb CN I s-' CN cb CN i r i CN CM CN CM CM cb CN • t f CN i r i CN i r i CN : Avg LO O •tf O I s -CD CN CD CD CO CD I s -CO o o CN I s -•tf o o • t f CD co o i n r-- CD • t f : Avg CM I s -• t f I s -I s-: I s -CO I s -cb I s -CN I s -i b cb I s -c b I s -i r i I s -cb I s -cb I s -Is-: I s -i r i I s -•tf I s -ISJC-I LO •tf LO • t f 00 LO CN •tf I s -CD CO 00 CO CO co o •tf CO CD LO CD Is -co CD LO co CD CO CO CO • t f CD CD • t f • t f CD CN I s -CD m CM CD I s - o CD CN CN CO o CD LO CO CN I s -m CN CN I s -CM I s -•tf I s -• t f I s -Is-: I s -Is-: I s -CD I s -LO I s -•tf I s - I s -cb I s -cb I s -•tf I s -CM I s -cb I s -CN I s -cb I s -i r i I s -i r i I s -cb I s -CN I s -c b I s -• t f I s - I s -I s-: I s -cb I s -i r i I s -•tf I s -•tf I s -jSamples o> XL CL z E CM O = © o CM co CM XL C ra 1-CM i CM = CO CM 00 i CN = CO CN T CN = CO CN LO 1 CN = co CN O) J £ O) E T— co IN LU = co CO CM XL C ra V-CN 1 CO = co CM co 1 CO = CO CN CO ; CO CM LO 1 CO = co CN o + 4 c o o •tf = CO CM Xi c ra 1-CN 1 •t f = CO CN co 1 • t f = CO CN •t f = CO CM LO 1 "tf r CO CN 177 co 111 X o tt CL > tt < (0 cn LL O C <D SZ a. c o +-C <D 4 Q Q 0) ~ a. E ro CO Values | -» S 21.192 21.159 21.021 21.188 21.255 21.2181 o Calorifi 1 cal/g 5065 5057 5024 5064 5080 5071 tal co 00 00 o CO I s -T— LO CO oo CO CJ) o CJ) CO o LO CJ) LO cn T t cn To T t 00 00 oo I s-' oo 00 00 LO oo cb oo cb 00 cb 00 T t 00 00 00 cb 00 Avg CO o CJ) T t T t T t o CM T t CM CO Avg 00 d d d cd d CM ipid CO T f o CD CD CO CO CD CM CD I s -CM o o o T t T t CM I s -I s -00 o 00 %L Is-: oo oo T— d CN d CM cb d d CO Is-: d CM d CM Avg I s -T t o oo in CD CO CJ) I s -O cn Avg oS T t d T t d T t d T t d IO d T t c |%Protei I s -CD I s -cn LO CO LO CJ) 00 00 CO T t CO o CO 00 I s -o T t T t LO 00 00 CD |%Protei 00 T t cn T t d LO CO T t d T t d T t d LO d T t d LO d T t d LO cb T t o > CN 00 CN oo I s -I s -co I s -CM CO I s -00 < oo cb 00 00 00 00 Ash T t 00 00 T t 00 o 00 LO I s -cn I s -I s -I s -LO I s -CN 00 00 I s -I s -CD CJ) oo oo oo cb od 00 od oci CO cb od cb Avg cn LO I s -CJ) CD 00 CJ) CJ) O CD CJ) Avg ci cn d O) d CJ) d cn cn d O) S Q cn Is -CM I s -I s -T t co CO I s -IO CD I s -00 co o CJ) o I s -o LO I s - oo •-s Q d d CJ) d CJ) d cn d cn d cn d O) O) cn O) d CJ) en > T t cn CN CO CN o O) o < oi d d d 00 d % Moist cn o CO I s -CO CM CD CO I s -CN LO CO CO CM CJ) cn co cn LO CN CM oo % Moist cn d d d . d d d CO 00 cb d cb Samples Control II 0.002 mg/kg 0.2 mg/kg •I 20 mg/kg 2000 mg/kg II LU = 178 APPENDIX B Summaries of Statistical Analyses Parameter P. Block Significant Difference Block P Treatment Significant Difference Treatment Growth FW 0.404 no 0.013 yes Growth SW 0.502 no 0.687 no Hepatosomatic Index 0.363 no O . 0 0 1 yes Growth Aug-mid 0.588 no 0.338 no Na (plasma) 0.964 no 0.013 no K (plasma) 0.503 no 0.169 no Condition Factor FW 0.713 no 0.011 yes Condition Factor SW 0.536 no 0.910 no Hematocrit Index 0.650 no 0.548 no T 3 F W 0.007 yes 0.002 yes T 3 SW 0.120 , no 0.609 no T 4 FW 0.194 no 0.356 no T 4 S W 0.851 no 0.781 no Protein 0.196 no 0.986 no Moisture 0.815 no 0.425 no Ash 0.569 no 0.599 no Lipid 0.784 no 0.416 no DFI FW 0.317 no 0.561 no DFI SW 0.287 no 0.741 no FE FW 0.832 no 0.024 yes FE SW 0.270 no 0.690 no PEP FW 0.778 no 0.022 yes PEPSW 0.176 no 0.503 no SGP FW . 0.262 no 0.015 yes SGP SW 0.433 no 0.241 no GEU FW 0.246 no 0.193 no GEU SW 0.166 no 0.186 no PPD FW 0.129 no 0.055 no PPD SW 0.189 no 0.732 no 180 APPENDIX C Survey Questionnaire Fax: (604) 822-9250 Risk Perception of Endocrine Disruption Investigator: Patricia Keen, M . A . Candidate. e-mail :plkeenpl@interchange. ubc. ca Research supervisor: Dr. K.J . Hall This short survey is part of my research involving effects of endocrine disrupting compounds for Master's thesis in Resource Management and Environmental Studies. I am interested in the perceived connection between ecosystem health and human health. This questionnaire is designed to provide some information to assess the knowledge of environmental effects of endocrine disrupting compounds (either natural hormones or man-made chemicals that can act as hormones in an organism) within a segment of the medical community. I would like to gain insight into the perception of risk associated with environmental exposure to endocrine disrupting compounds (EDCs). The survey should take about 10 minutes to complete. Agreeing to complete this questionnaire shall be taken as consent to provide this information. A l l information is confidential and intended as a basis from which to design a comprehensive study of perception of risk associated with endocrine disruption. Please at least answer the yes/ no questions and question 9. Additional comments on any of the questions are very much appreciated. 1. How are organisms and humans exposed to endocrine disrupting compounds? Where do you know they come from? 2. Do you believe environmental exposure to endocrine disrupting compounds has a greater effect on other organisms than on humans? • Yes • No • The same effect • Undecided 3. Do you believe human health can be directly affected by endocrine disrupting compounds? • Yes • No • Undecided Why? 4. In your opinion, are effects of exposure to exposure to any other pollutants? • Yes • No • Undecided Why? 5. Do you believe a controversy exists about compounds? • Yes • No • Undecided endocrine disrupting compounds different from environmental exposure to endocrine disrupting 182 What kind of controversy"? 6. Have you been asked by another individual for information concerning endocrine disrupting compounds? • Yes • No • Undecided 6a. Have you noticed more men or women asking you for this type of information? • More men • More women 7. Do you notice any gender bias in the popular press that reports on effects of exposure to endocrine disrupting compounds? • Yes • No • Undecided What gives you this impression? 8. Do you know of public health policy dealing with endocrine disrupting compounds? • Yes • No • Undecided 9. Medical specialty Number of years in medical profession With which ethnic/racial group do you most closely identify? • Male • Female Please provide any comments you may wish: 183 APPENDIX D Poster from i f n Annual SETAC Europe Meeting, Madrid Spain May 2000 184 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items