AE1136 - Health of Coastal Environments: Pathological Risk Evaluation Using Diagnostic and Innovative Molecular & Cellular Tools (PREDICT 2).
Prof. Michael Moore1, Prof. James Readman1, Mr Icarus Allen1, Mr David Lowe1, Dr Jenny Shaw1, Dr Robert Clarke2, Dr Brett Lyons3, Dr Grant Stentiford3, Dr Steven Feist3, Mr John Bignall3 & Dr John Thain4
1Plymouth Marine Laboratory, Plymouth, UK
2 PRIMER-є, Plymouth Marine Laboratory, Plymouth, UK
3CEFAS Weymouth, UK
4CEFAS Burnham-on-Crouch, UK
Executive Summary
The UK is developing an ecosystem-based approach to safe clean management of our seas. Development and deployment of effective “indicators of state”, which can demonstrate the health of ecosystems, is a key requirement for the effective safe management of the coastal seas in the UK. In accordance with this need, the overall aim of this project was to develop and test diagnostic pathology-related indicators of harmful pollutant effect (biomarkers) in accordance with Defra and ICES requirements.
In this project, a suite of biological measurements (biomarkers) operating at different functional levels within the individual, that could indicate pollutant exposure and harmful effects of pollution, were applied to common, coastal invertebrate species (mussels and periwinkles) and flatfish in order to test their suitability for assessing marine pollution at sites around the UK. The potential use of pathological cellular reactions to environmental pollutants (organic chemicals, copper and C-60 fullerene nanoparticles) is explored in relation to predicting aquatic animal health, based on clinical-type diagnostic tests. Biomarkers, such as damage to the genetic material (DNA) in the liver in fish, together with liver histopathology, already provide a powerful set of “health status-related” assessment tools. In order to derive a similar set of tools for mussels, marine snails (periwinkles) and other molluscan sentinel animals, several indicators of cell injury have been selected, including tests for free-radical mediated oxidative damage. These biomarker tests meet the criteria that they are relatively inexpensive, rapid, sensitive, precise, can be easily learnt and readily interpreted, are applicable to a broad spectrum of marine environmental sentinel animals, as well as having prognostic value for development of pathologies and are ecologically relevant.
An integrated system of tests has been developed that enables us to confidently link damage caused by pollutant chemicals at the molecular level to proteins and cellular membranes, with the development of tissue pathology and disease. A further advantage of these tests is that they are probably generic and, consequently, could be applied to a wide range of animals including shellfish and fish.
Derivation of generic explanatory frameworks for prediction of pollutant impact on health is a major goal; and we have developed a conceptual mechanistic model linking molecular damage with injury to cells and tissues. This conceptual model has also complemented the creation of a cell-based computational model for the liver analogue of mussels and other molluscs that simulates cellular reactions to pollutants. Experimental and simulated results have indicated that increased turnover of cell constituents induced by diet restriction or nutritional deprivation (i.e., self-eating or lysosomal autophagy) has a protective function against toxic effects mediated by reactive oxygen species (ROS), including free-radicals. This process may contribute to stress resistance in intertidal animals, such as mussels and periwinkles, which are subject to fluctuating environmental conditions as a result of the tiadal cycle.
Contaminants seldom occur singly in the marine environment and are usually present as complex mixtures. The effects of experimental contaminant mixtures have been assessed in mussels (immune system and larvae) and the results show that harmful toxic synergistic effects occur that may be linked to the increasing complexity of the mixture.
The use of coupled empirical measurements of biomarker reactions and modelling is proposed as a practical approach to the development of an operational toolbox for predicting the health of the environment and this has been manifested in the form of a decision support or “expert system” that can be used to categorise harmful environmental impact. The decision support system has been developed in collaboration with colleagues in Italy and Germany, and can be used to categorise environmental risk in populations of sentinel animals.
Current assessment methods are largely indicative of exposure to chemical contamination but do not necessarily indicate harmful effects on the health of sentinel animals. In contrast, we have clearly demonstrated links between damaging cellular reactions induced pollutant- and early onset pathology. The methods developed and successfully tested in this study provide direct measures of animal health status. Cell-based simulation models have also been successfully developed for flatfish (dab and flounder) liver and molluscan hepatopancreas (liver analogue).
Finally, we have evidence for a statistical correlation between cellular health in mussels and biodiversity of bottom-dwelling animals (i.e., benthic macrofauna). Although this is only a correlation it may be indicative of a general linkage between animal well-being and ecosytsem integrity.
CONTENTS
Executive Summary1
Contents 3
1. Introduction 4
1.1. Aim4
1.2. Objectives4
1.3. Background & rationale5
2. Results 8
2.1. Selection and experimental development of pathology-related biomarkers8
2.2. Pathology-related biomarkers and “health status”in molluscs8
2.3. Multixenobiotic resistance protein (MXR) as an exposure biomarker9
2.4. Oxidative Stress10
2.5. Ecological relevance10
2.6. Pathology-related biomarkers and “health status” in fish10
2.7. Challenge Tests10
2.8. Recovery from Stress11
2.9. Nanotoxicity11
2.10. Application of the Biomarker Approach to Early Life Stages of Molluscs11
2.11. Effects of contaminant mixtures12
2.12. Simulation Modelling of Pollutant Effects12
2.13. Decision-support system for interpretation of biomarker data13
2.14. Field application of biomarkers in flatfish (dab) from Cardigan Bay14
2.15. Field application of biomarkers in mussels from the Tamar Estuary14
3. Discussion 15
3.1. Fish biomarkers15
3.2. Molluscan biomarkers15
3.3. Environmental prognostics: integration of biomarker-based studies and
simulation modelling 17
4. Conclusions19
5. Technical Summary & Recommendations21
5.1. Summary 21
5.2.Recommendations22
6. Data and Publications22
7. References 24
Tables & Figures 30
APPENDIX 1 - Methods
APPENDIX 2 - Pathology Related Biomarkers & Health Status
APPENDIX 3 - Molluscan Hepatopancreas & Fish Liver Simulation Models
APPENDIX 4 - Decision Support System for Interpretation of Multiple Biomarker Data
1. Introduction
1.1. Aim
The current evolving ecosystem approach to marine environmental management requires that the cumulative effect of all conceivably relevant impacting human activities are considered. Although performance indicators play an important and essential role in environmental protection, these indicators do not allow us to say with confidence that the coastal seas are in a healthy state (Defra, 2005). Consequently, a Defra have identified the requirement for a further category of indicators that can demonstrate that ecosystems are healthy (Altieri et al., 2007). These are known as indicators of state (Defra, 2005).
The overall aim of this project was to develop suitable pathology-related biomarkers and demonstrate the effectiveness of their application as potential indicators of state in environmental sentinel animals.
1.2. Objectives
- Develop CEFAS and ICES/OSPAR acceptable diagnostic tools for the early detection of molecular, cellular, physiological and pathological abnormalities in aquatic sentinel organisms exposed to toxic environmental chemicals selected from the OSPAR list (OSPAR Strategy with regard to Hazardous Substances; Ref No. 1998-16).
- Select diagnostic tools that are relatively inexpensive, rapid, sensitive, precise, can be easily learnt, and readily interpreted, that are applicable to a broad spectrum of marine environmental sentinel animals, which also have prognostic value for development of pathologies and are ecologically relevant.
- Develop and validate strategies for the application of diagnostic tools in hazard and risk assessment of environmental chemicals.
- Develop innovative strategies, such as challenge tests (e.g., exposure to a standard thermal or chemical stress), to enhance the capability to assess the effects of chronic exposure to low levels of chemicals in situ (for example in sediments).
- Evaluate the predictive capability of the chosen diagnostic tools for selected endpoints such as pathology and developmental abnormalities, which can be readily related to health of the environment/ecological impact.
- Address deficits in current risk assessment procedures by focusing on appropriate aspects of invertebrate (i.e., molluscan) pathology and assessments of health status in fish.
- Develop simulation models for linking processes of pollutant chemical speciation/particulate binding, bioavailability, mode of cellular uptake, compartmentalisation, toxic targeting and cell injury, with the higher level pathological consequences.
- Based on existing data and specific experiments, evaluate the risk of complex contaminant chemical mixtures, such as those existing in coastal environments.
Biologists have been reducing life to its constituent parts for over 50 years. The new challenge is to reassemble this data to unravel how complex systems, from subcellular processes to organisms, work. Systems biology attempts to reconstruct biological systems by developing and evolving series of overlapping conceptual, numerical and statistical models (Hunter, 2003), a process involving the interaction of experiment and simulation in an ongoing iterative process.
Coastal marine ecosystems are sensitive to exposure to toxic contaminants. Pollutants either individually or in combination may have sub-lethal effects at the cellular, organ and individual level, (e.g. causing changes in genetic, behavioural and reproductive activity). Key species have been identified as indicators of this sensitivity including the edible mussel, periwinkles, crabs and several species of fish (Altieri et al., 2007; Bayne et al., 1988; Stebbing et al., 1992). Biomarkers include a variety of measures of specific molecular, cellular and physiological responses of key species to contaminant exposure. A response is generally indicative of either contaminant exposure or poor health. The challenge is to integrate individual biomarker responses into a set of tools and indices capable of detecting and monitoring the degradation in health of a particular organism. We have propose dEnvironmental Prognostics as a branch of systems biology that is specific to the reactions of organisms to both natural and anthroprogenic stress (Allen & Moore, 2004).
The key enabling concepts of environmental prognostics are:-
1)Acknowledgement that reductionist science acts to disassemble ecotoxicological impacts into constituent processes;
2)Acceptance that biology is a cross-disciplinary science involving mathematics, physics, chemistry, engineering and information technology;
3)Moving towards the notion that biology is an information-based rather than a qualitative science;
4)The process requires the assembly of systems by modelling followed by disassembly and focused experimentation as an ongoing procedure.
The crux of the procedure is the definition and evaluation of models of the system in question. This requires the use of the following heavily interdependent tools: conceptual, statistical and numerical models, empirical experimental work and bioinformatics (Fig. 1). Many biomarkers only exhibit a response in a part of the health status space. They indicate that the response has taken place and may even indicate health status within a narrow range, or what has induced the response, but they do not indicate the health status of the whole range from healthy to terminally ill. In terms of environmental prognostics, the first stage is to relate biomarker responses to health status of individual organisms and then to derive integrated explanatory frameworks.
In order to achieve the above objectives, it is necessary to explore the utility of biomarkers (e.g., DNA adducts, oxidative stress reactions, subcellular responses and tissue level histopathology) as prognostic indicators for putative pathophysiology that will permit prediction of animal health status; as well as the development of realistic integrated conceptual models encompassing reactions of pathology-related biomarkers and oxidative stress processes in cell injury and pathology. In attaining this overall aim we have three primary goals.
Our first goal is to determine whether the selected biomarkers provide appropriate information related to pathological change in molluscs and flatfish, based on experimental treatments with contaminant chemicals and published data in the scientific literature. The second goal is statistical analysis and modelling of chemical and biomarker data, using an index of “health status” as a reference measurement for cellular “well-being” (Allen & Moore, 2004). This data will be subjected to analysis using univariate and multivariate statistical routines, in order to develop an appropriate conceptual framework and statistical models for the role of selected biomarker function and responses to environmental variables, particularly chemical pollutants.
Our third goal is to derive relational frameworks that can be used to integrate disparate data sets and for testing a numerical model used to produce computational simulations of digestive gland function in blue mussels (McVeigh et al, 2004) and hepatocyte function in flatfish liver (Jamal et al., 2006). Such simulation models will complement the experimental data as part of the necessary basis for explanatory frameworks that will facilitate the development of a predictive capacity for estimating risk to the health of sentinel animals associated with the possibility of future environmental events.
1.3. Background & rationale
Biomarkers include a variety of measures of specific molecular, cellular and physiological responses of key species to contaminant exposure (Depledge, 1994, 1999; Depledge et al., 1993; Feist et al., 2004; Moore et al., 2004a, b). A response is generally indicative of either contaminant exposure or compromised physiological fitness. The challenge is to integrate individual biomarker responses into a set of tools and indices capable of detecting and monitoring the degradation in health of a particular type of sentinel organism.
However, what we are currently lacking are integrated explanatory frameworks for evaluating complex environmental information and predicting harmful biological effects and their subsequent consequences for environmental health. And while it is clearly recognised that stress-induced changes at the population / assemblage / ecosystem / human health levels of biological organisation are the ultimate concern; they are generally too complex and far removed from the causative events to be of much use in developing tools for the early detection and prediction of the consequences of environmental stress (Depledge et al., 1993; Feist et al., 2004; Lyons et al., 2004; Moore et al., 2004).
Consequently, it is only at the lower levels of biological organisation that we will have the reasonable expectation of developing a basis of mechanistic understanding of how different environmental conditions can modulate organismal function, which in turn will ultimately help in linking causality with predictability of response (Livingstone et al., 2000; Marigomez & Baybay-Villacorta, 2003). This is in part due to our ability to make certain generalisations about biological organisation and function at the molecular and cellular level, which rapidly disappears as we ascend the hierarchical ladder. Hence, distress signals at the molecular, cellular and physiological levels of organisation should be capable of providing "early warning biomarkers” (molecular, cellular, physiological and behavioural) indicating reduced performance; some of which may be prognostic for impending pathology and severe damage to health of the animal (Depledge, 1994; Depledge et al., 1993; Feist et al., 2004; Galloway et al., 2002, 2004; Moore, 2002).
Responses of the lysosomal-vacuolar system may provide a solution to the question of prognostic biomarkers, since injurious lysosomal reactions frequently precede cell and tissue pathology. Lysosomal perturbations have been widely used as early indicators of adverse effect to various factors, including pollutant exposure (Galloway et al., 2004; Moore, 2002; Moore et al., 2004). Consequently, lysosomal function can be used across a range of animals, including annelids, molluscs, crustaceans and fish to detect responses to environmental stress (Cajaraville et al., 2000; Galloway et al., 2004; Hankard et al., 2004; Hwang et al., 2002; Köhler et al., 1992, 2002; Lekube et al., 2000; Lowe et al., 1992, 1995; Svendseb & Weeks, 1995; Wedderburn et al., 1998).
Lysosomal reactions are involved in normal physiological responses as well as many cell injury and disease processes; these include augmented sequestration and autophagy of organelles and proteins (Cuervo, 2004; Klionsky & Emr, 2000; Moore, 1990, 2002). Stress-induced macroautophagy, such as that triggered by nutrient deprivation, is regulated by the mTOR kinase (mammalian target of rapomycin) in eukaryotic cells from yeast to mammals (Klionsky & Emr, 2000). Such reactions have been widely documented for many adaptive and developmental physiological and disease processes; and lysosomal responses have been shown to be involved in generalised reactions to environmental stress (Cajaraville et al., 1995; Köhler et al., 2002; Moore, 1985, 1990, 2002). The functional stability of the lysosomal membrane is a good indicator of lysosomal integrity and has been used widely to measure responses to environmental perturbation in fish and molluscs (Allen & Moore, 2004; Hwang et al., 2002; Köhler et al., 2002; Moore, 2002; Moore et al., 2004a, b).
Lysosomal functional integrity is a generic common target for environmental stressors in all eukaryotic organisms from yeast and protozoans to humans (Cuervo, 2004), that is evolutionarily highly conserved, and lysosomal membrane stability is a good diagnostic biomarker of individual health status (Allen & Moore, 2004; Bayne & Moore, 1998; Burlando et al., 2002; Cajaraville et al., 1995, 2000; Dondero et al., 2006b; Galloway et al., 2002, 2004; Hankard et al., 2004; Klionsky & Emr, 2000; Köhler et al., 1992, 2002; Lekube et al., 2000; Lowe, 1988; Lowe et al., 1982, 1992, 1995, 2006; Marigomez & Baybay-Villacorta, 2003; Moore, 1976, 1985, 1988, 1990, 2002; Moore et al., 2004a; Moore et al., 2006a,b,c; Nicholson & Lam, 2005; Svendsen & Weeks, 1995; Svendsen et al., 2004; Winston et al., 2002). Dysfunction of lysosomal processes has been mechanistically linked with many aspects of pathology associated with toxicity and degenerative diseases (Cuervo, 2004; Köhler, 2004; Köhler et al., 2002; Moore et al., 2006a, b). Recent studies have shown that lysosomal autophagy provides a second line of defence against oxidative stress (Cuervo, 2004; Moore et al., 2006c), and the capability to effectively up-regulate this process is probably a significant factor contributing to the ability of some organisms to tolerate stressful and polluted environments.
Lysosomal membrane stability has recently been adopted by UNEP as part of the first tier of techniques for assessing harmful impact in the Mediterranean Pollution programme (MEDPOL Phase IV). Other lysosomal biomarkers including lipofuscin in molluscs (age/stress pignment), and lysosomal neutral lipid (chemically induced lipidosis) in molluscs and fish have been adopted as part of the second tier assessment methods (Krishnakumar et al., 1994; Moore, 1988; Moore et al., 2004b).
This biomarker can also be used prognostically to predict liver damage and tumour progression in the liver of various fish species (Broeg et al., 1999 a, b; Köhler et al., 2002; Köhler, 2004), and hepatopancreatic degeneration in molluscs (e.g., blue and green mussels, freshwater bivalves and snails, periwinkles, oysters), coelomocyte damage in earthworms, as well as enhanced protein turnover (i.e., lysosomal autophagy) as a result of radical attack on proteins; and energetic status (i.e., scope for growth) as a predictive indicator of fitness of individuals within a population (Allen & Moore, 2004; Kirchin et al., 1992; Köhler et al., 2002; Moore et al., 2004a, 2006a; Nicholson & Lam, 2005; Svendsen & Weeks, 1995; Svendsen et al., 2004).