A review of biomarkers and other bioassay- based toxicity assessment ..... The term biomarker has been adopted in the context of environmental monitoring and.
Review of biomarkers, bioassays and their potential use in monitoring the Fal and Helford SAC.
W. J. Langston, B.S. Chesman and G.R. Burt Marine Biological Association, Citadel Hill, Plymouth, PL1 2PB
A project carried out on behalf of Natural England (Nominated Officer: Roger Covey, Senior Marine Specialist)
ii
Summary The assessment of the condition of marine SACs, including the Fal & Helford SAC, by Natural England is currently guided by a combination of water quality data (often collected for compliance purposes) and more opportunistic ecological survey data, which are often disparate, both spatially and temporally. Discussion between Natural England, the Environment Agency and the research community has produced agreement that better ways are needed to monitor and assess site condition. Further investigation of the potential role of biomarkers and bioassays, as a means of gauging anthropogenic impact, has been suggested as a potential key work area to complement the more conventional assessment procedures. If such techniques can be trialled and validated rigorously there may be strong arguments for their routine deployment alongside other surveillance methods. Because most biomarker and bioassay responses are readily quantifiable they could form the basis of long term observational series to determine whether the site is improving or deteriorating. If, as expected, biomarkers respond more rapidly to impacts than community changes, they would also be useful in an early warning capacity. Moreover, some biomarkers are selective in their response and may help to diagnose the principal chemical threats and sources, thereby helping to guide appropriate management responses. ‘Site Characterisation’ has proved to be a useful tool, helping to prioritise the water quality concerns and monitoring requirements for European Marine Sites. Current perceptions are that there are three major water quality issues in the Fal & Helford SAC; nutrient enrichment (periodic eutrophication), metals and TBT contamination, which all exhibit distinct gradients. There is evidence of acute biological impact at sites just outside the SAC boundary, for example in Restronguet Creek (metals) and Falmouth (from TBT sources). Acute effects related to nutrient enrichment are periodic in occurrence and, though focused in the upper Fal, can be manifested in other parts of the system, including the Helford. There may also be other chronic influences that we know little about. The Fal and Helford SAC is therefore a suitable test-bed for the application of novel methods to assess threats to nature. The problem is; which biological effects tools to deploy, and how best to use them? To help answer this question, the current report reviews information on the availability and suitability of biomarkers and other biological effects tools, and their potential use in monitoring the Fal and Helford SAC. The rationale and protocols for deployment are also discussed. Key recommendations are 1) The need to deploy an appropriate suite of ‘core’ biomarkers, including diagnostic exposure types as well as general health markers, and 2) the requirement to run biomarkers and other biological effects indicators alongside the most appropriate, contemporary, chemical measurements - preferably in conjunction with higher order measures of ecological response (e.g. abundance and biodiversity indices). Some of these recommendations are currently being addressed in a broader context under Natural England project FST20-18-028 and within the context of the Fal Pilot project (sponsored partly by NERC).
iii
iv
Contents Summary ...................................................................................................................... iii 1. Background ................................................................................................................1 2. Objectives ..................................................................................................................3 3. A review of
biomarkers and other bioassay- based toxicity assessment
methodologies under development or used within SW research community................3 3.1 ‘Biomarkers’ ........................................................................................................3 3.1.1 Biochemical ..................................................................................................5 3.1.2 Molecular and cellular biomarker approaches ..............................................8 3.1.3 Genotoxicity, DNA damage and chromosomal aberrations .......................10 3.1.4. Reproductive markers and Endocrine disruption.......................................11 3.1.5 Physiological Indicators..............................................................................13 3.1.6 Behavioural biomarkers ..............................................................................15 3.2 Bioassays and Direct Toxicity Assessment (DTA) ...........................................15 3.3 Previous biomarker and bioassay deployments in the Fal & Helford SAC.......18 4. Potential role and state-of-readiness of other biomarkers/bioassays used nationally and internationally .......................................................................................................28 5. A biomarker strategy for use in the Fal & Helford SAC .........................................35 5.1 Identification of biomarkers...............................................................................35 5.2 Identification of sites for deployment of biological effects techniques.............39 5.3 Identification of species for measuring biological effects. ................................41 5.4. Methods of interpreting biomarker data to inform assessment of site condition ..................................................................................................................................42 6 Requirements for contextual monitoring to validate biomarker responses (biological and chemical measurements) .......................................................................................46 6.1 Chemistry...........................................................................................................46 6.2 Ecological Condition .........................................................................................47 7. Conclusions and Recommendations ........................................................................48 8. Acknowledgements..................................................................................................51 9. References................................................................................................................53
v
vi
1. Background
Biological Effects monitoring in the context of Natural England’s statutory requirements The EU Habitats Directive is founded on the ‘precautionary principle’ and highlights the need to maintain favourable condition at European sites designated under the Habitats and Birds Directives. Through the Conservation Regulations (1994), which implement the Habitats Directive, there may be requirements for Natural England to provide advice and, in some cases, to carry out assessments themselves to ensure that the ecological integrity of a European site is not adversely affected by proposed or existing schemes. In line with these requirements there needs to be a greater understanding and ability to predict the influence of anthropogenic activity on Special Areas of Conservation (SACs, designated under the Habitats Directive). As part of this process, biological effects techniques may offer a rapid cost-effective complement to traditional ecological and chemical monitoring and are consistent with the ecosystem approach to marine stewardship, to which the statutory agencies are committed. There are, in fact three main areas of Natural England’s work where information from biomarkers and other biological effects measurements could be used: • Condition assessment of sites designated for nature conservation under national and European legislation. • Assessing impacts from plans and projects on sites designated under the Birds and Habitats Directives, as a requirement of the Habitats Regulations 48 and 50 (1994). • Assessment of wildlife “health” in the wider environment
The mechanisms required to deliver the ecosystem approach to pollution management, and relationships to major legislation and policy (e.g. Habitats and Water Framework Directive) are still under development but will need to be in place in the near future (Covey and Laffoley, 2002). Natural England has a statutory responsibility to monitor the condition of Special Areas of Conservation, and report their condition to the European Commission through the Joint Nature Conservation Committee. Current monitoring focuses on measuring a series of pre-determined attributes relating to the habitats and species for which the cSAC has been selected, and reporting changes against baseline condition. Such monitoring will detect change in these attributes, but is unlikely to be detailed enough to directly inform management action without further research. The monitoring is also not precautionary – only detecting change when the impact has occurred at population or community-levels, rather than giving an early warning of potential impact which can then be managed before serious damage occurs. Part of the rationale behind the proposed use of biomarkers and bioassays, is that they may be able to detect effects on organisms and habitats before they become significant in ecological and conservation terms. 1
Water quality is a key parameter in the health of the marine environment. However, the links between particular contaminants and impacts at the species and community level is generally poorly documented and understood. The effects of mixtures of toxic pollutants are even more uncertain. Traditionally, water quality monitoring has been undertaken at the chemical level – looking for exceedences to known thresholds of single chemicals. There has been little consideration of the combined sub-lethal ‘cocktail effects’ of mixtures of chemicals: the rationale for application of biological effects tools is particularly strong in this respect and in terms of developing strategy on environmental protection. A general review of the scope, potential and criteria for using biomarkers (across freshwater, terrestrial and marine environments) has been compiled for Natural England by Long et al (2004). The conclusion from this and similar reviews is that potentially, the use of biomarkers and other bioassays, such as direct toxicity tests, may provide more precautionary and effective monitoring of water quality impacts, with a clearer link to management measures. Thus, as well as using these techniques to assess the bioavailability and toxic effects of particular pollutant groups at individual sites, there is a great deal of interest within Natural England as to their potential role in providing a risk-screening approach across sites, and also in surveillance monitoring, to assess the overall stress imposed on communities through exposure to toxicants. Outputs from such monitoring may supplement current condition tables, or could trigger further investigations at high risk sites i.e. helping to direct management responses. Whilst a great deal of laboratory research has been conducted on the development of individual biomarkers and whole effluent toxicity tests, their value in predicting detrimental effects in the field is still poorly understood. Before such techniques are used in routine monitoring of European Marine Sites, it is important to identify biomarker responses which can be reliably taken to indicate toxic stress under realistic environmental conditions. It would also be desirable to know how the level of biomarker response measured relates to impact on populations or communities, although this remains an elusive parameter to quantify precisely. Thus, with a few exceptions, biomarkers are not yet adequately validated as ‘standalone’ assessment tools and there has to be further consideration of their development and application in the field of environmental protection. Through the SW Marine Science Forum 1 - whose aims include translation of policy into applicable research Natural England, along with the Environment Agency and the SW marine research community, are keen to progress this issue. As part of this brief it was agreed that a small integrated research project, focusing on the Fal & Helford, could serve as a pilot scheme to demonstrate how to achieve delivery of better science, better management and hence a better environment for the SW. The current project - investigation of the potential roles of biomarkers and bioassays to assess toxic stress – represents a key work area of the Fal and Helford pilot study.
1
(www.swmarinescienceforum.org)
2
2. Objectives The overall objective of this project is to consider the role that biomarkers and other biological effects tools might play in condition (and impact) assessments of marine sites, and to develop a strategy for their application in the Fal and Helford SAC. Specific objectives are: 1. To review biomarkers and other bioassay-based toxicity assessment methodologies that have been developed, or are under development, within the SW research community, and to summarise where they have been deployed on the Fal and Helford system; 2. To consider, based on existing national and international reviews, the potential role and state-of-readiness of other biomarkers/bioassays; 3. To take into account the findings of objectives (1) and (2) and design a sampling regime/protocol which deals with: locations to be sampled; the suite of biomarkers to be deployed and the organisms to be investigated (in situ and/or transplanted). Requirements for contextual information (chemical and ecological characterisation) necessary to validate the use of biomarkers/bioassays in SAC monitoring are also considered.
3. A review of biomarkers and other bioassay- based toxicity assessment methodologies under development or used within SW research community 3.1 ‘Biomarkers’ There are many definitions of ‘biomarker’, a term which has its origins in human health diagnostics. The official NIH definition of a biomarker is: "a characteristic that is objectively measured and evaluated as an indicator of normal biologic processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention”. Alternatives include ‘A specific biochemical in the body which has a particular molecular feature that makes it useful for measuring the progress of disease or the effects of treatment’ The term biomarker has been adopted in the context of environmental monitoring and in the process has expanded widely, beyond the original constraints, to encompass almost any response which may be indicative of a biological effect. Somewhat arbitrarily biomarkers may be divided into biochemical, cellular & molecular, physiological, reproductive and behavioural responses, not all of which are entirely objective. Nevertheless, it is this eclectic definition which is considered here since it maps well onto that described by Long et al 2004 in a recent Natural England report.
3
Biomarkers Biochemical, cellular, physiological or behavioural variations, in the tissue or body fluids or at the level of whole organism, that provide evidence of exposure to chemical pollutants, and may also indicate a toxic effect A crucial point is that some biomarkers exhibit relatively chemical-specific ‘exposure’ responses whilst others represent more generalised indicators of ‘health or condition’. Combining both types therefore offers the most comprehensive option for monitoring habitats. It is worth reviewing some of the general ideology concerning the application of biomarkers, summarised from Long et al (2004). Key points to be appreciated are, that: - A clear understanding of the scope and nature of the intended monitoring is needed - There are at least three main roles for biomarkers: - as a screening tool that could trigger further study of a site, - a diagnostic tool which focuses on specific issues identified, - a monitoring tool to follow the ‘health’ of a site over time, perhaps in response to management measures. - Biomarkers with clear linkage to higher order effects would be the most effective - Baseline knowledge of natural variation, sensitivity and reproducibility in biomarker response is needed (Quality Assurance). - Individual biomarkers are unlikely to be robust enough for regulatory action: a suite of biomarkers could provide a body of evidence. A fairly large number of biological effects techniques have been developed for use in marine organisms and some have been adopted by OSPAR for their surveillance programme (OSPAR, 2003, 2004; ICES, 2004). Nevertheless, the application of biomarkers as a monitoring ‘tool’ is in its infancy. In view of their novelty, biomarkers should, in the first instance at least, be validated and used in conjunction with other biotic and chemical indices. Further studies to investigate the relationship between biomarkers and diagnostic responses at the individual, population and community level would be extremely valuable. Concerning their ease of application, it is important to appreciate that the use of biomarkers often requires a high level of expertise to avoid false interpretation, which could be seen as a disadvantage. Particular care is needed to minimise the effect of natural variables, and to use appropriate controls. A comprehensive review of biomarkers would be prohibitively long and we have restricted descriptions to types which are most relevant to marine and estuarine organisms and habitats. Within each class we highlight (in grey boxes) those considered as ‘core’ biomarkers – those which are suitable for deployment or are currently being trialled in the Fal & Helford. Selection will depend on the precise
4
reason for the study, suspected threats, relevance and costs.
3.1.1 Biochemical
• Induction of detoxifying enzymes and proteins – the basis of adaptation and resistance MT induction is primarily an indicator of exposure to metals (e.g. Cu, Cd, Zn, Hg). In acute exposures, induction of MT to high levels (and changes in metal binding patterns) is related to the onset of toxic effects and mortality. In fish, MT induction can be influenced by both abiotic and biotic factors such as water temperature, age and reproductive phase, though these factors are less influential in invertebrates such as mussels. Nevertheless, as with most biomarkers, an understanding of these factors is important to the design and interpretation of field deployments. Metallothionein (MT) induction and intracellular metal-binding patterns are capable of mapping the extent of responses to elevated levels of bioavailable metals. MT protein can be measured accurately by differential pulse polarography; accompanying metals by LC-AAS (e.g. Langston, et al., 2002). This assay is able to ascertain whether or not the population is adapting successfully to the site conditions and may be used to quantify change. MT is best viewed as a potential early warning indicator – a predictive tool for contaminant risk assessment. Phytochelatins (PC) are the equivalent metal-binding components in plants and could be useful in a similar capacity to MT as a marker of metal exposure. To date however, the application and validation of PC has been less extensive and hence the assay is not in routine use. Glutathione S-transferases (GST) are a multigene family of enzymes of the phase II biotransformation system which inactivate toxins by chemically bonding them to the tripeptide glutatione (GSH), making them soluble for easy excretion. GSH levels are responsive to metals but are also responsive to a number of organic hydrophilic contaminants, including PAHs and some types of pesticides (organophosphate, thiocarbamates, triazines). In addition to detoxifying pollutants directly, GSH has antioxidant properties. Because of these responses to such a wide range of contaminant types GSH is more useful as a general stress indicator rather than as a diagnostic tool. Delta-aminolevolinic acid dehydralase (ALAD) inhibition is responsive to lead, (together with mercury and inorganic arsenic), but is limited mainly to use in evaluating lead exposure in haemoglobin-containing species (particularly humans). Acetylcholinesterase activity (AChE) is an exposure biomarker which has been tested
5
often in the marine environment because of its potential application in screening for the effects of pesticides and other pollutants in a range of vertebrate and invertebrate species. Acetylcholinesterase activity (AChE): acetylcholinesterase is an enzyme that was been widely used as a biomarker for neurotoxicity. It may be a particularly sensitive marker for organophosphate and carbamate pesticides, but may also respond to metals, detergents and some algal toxins. Inhibition of esterase enzymes may be estimated using a simple low-cost spectrophotometric assay adapted for invertebrate species (Galloway et al., 2002). EROD (ethoxyresorufin-O-deethylase), is a marker for the activity of the cytochrome P450 system (e.g. CYP1A), whose induction is usually associated with exposure to, and the detoxification of, xenobiotics such as planar PAHs, PCB and dioxins. EROD activity and CYP1A protein concentration are often measured jointly using spectrometry and ELISA, respectively. Induction is mediated through the binding of xenobiotics to a cytosolic aryl hydrocarbon receptor (AhR): EROD activity describes the rate of the CYP1A mediated de-ethylation of the substrate 7-ethoxyresorufin (Scmitt et al., 2002). Occasionally these transformations may produce deleterious side effects due to the formation of carcinogenic or genotoxic compounds (e.g. the formation of benzo(a)pyrene diol epoxide from benzo(a) pyrene). EROD activity is primarily an indicator of exposure. The relationship to higher levels of organization is still unclear but it could be related to detrimental effects, including reproductive activity and mortality. Therefore EROD is best viewed as a potential early warning indicator – a predictive tool for contaminant risk assessment. It is a fairly sensitive, low-cost tool but can be influenced by a large number of abiotic and biotic factors such as water temperature, age and reproductive phase (including oestrogens). Induction can also be inhibited by an overload of contaminants giving a false ‘noresponse’ signal (e.g. some low molecular weight PAHs do not exhibit P450 inducing effects and can suppress the normal pattern of induction elicited by larger PAHs). As with most biomarkers an understanding of these complex factors is critical to the design and interpretation of field data. Benzo(a)pyrene hydroxylase activity has also been suggested as a surrogate measure of the phase 1 biotransformation activity of the mixed-function oxidase (MFO) system towards PAH metabolism (specifically benzo(a)pyrene), but may be inhibited by high loadings of other contaminants. Further validation of this biomarker under field conditions may be useful, including development of immunochemical assays to replace interference-prone fluorometric detection. PAH metabolites. Exposure to PAHs and their metabolites may be screened semiquantitatively by fluorescence measurements of the urine of crustaceans, bile in fish and body fluids of other invertebrates (Watson et al., 2003). To fully understand the risks posed to organisms whose biological fluids contain detectable levels of PAH metabolites, it is necessary to combine this technique with effects biomarkers (physiological and biochemical). Parameters in fish which correlate with elevated PAH metabolites in the bile may include CYP1A induction (Beyer et al., 1996 and Miller et al., 1999) and incidence of hepatic neoplasia (Krahn et al., 1986)
6
Fixed wavelength and synchronous fluorescence spectrometry are relatively rapid, low-cost options for PAH exposure estimates but can be subject to some interferences. Lengthier high performance liquid chromatography (HPLC)-fluorescence or gas chromatography-mass spectrometry with selected ion monitoring (GC-MS-SIM) analysis is generally required for fully quantitative and qualitative assessment. The composition of PAHs and their metabolites can give an indication of sources: for example 1-OH pyrene is characteristic of pyrogenic sources (urban and land-based industry run-off) whilst alkylated naphthalenes and phenanthrenes are more likely to signify oil derived (petrogenic) origins. Heat shock (or stress) protein induction (e.g HSP60, HSP70, HSP90) is proposed as a general biochemical marker of stress whose expression can be demonstrated in response to a wide variety of chemical and physical insults. The diagnostic value of this category of biomarker may therefore be limited. Also a bell shaped doseresponse curve means interpretation may often be difficult. HSPs are typically measured by electrophoresis. Induction of some of these defensive systems may ultimately lead to adaptation - of which there are a number of examples in the Fal relating to the history of long-term exposure to metals (see section 3.3). The adaptations themselves are an indication that the organisms are responding to stressors and could be classed as a ‘biomarker of effect’. Although the adaptation may ensure survival there may be other costs (metabolic, reproductive, production of more toxic bi-products) for the organism/population. Metal-adapted populations of Hediste (Nereis) diversicolor in Restronguet Creek are probably the best example reported to date and suggest a genetic basis to the development of tolerance. Investigational studies of tolerant populations to establish the extent of adaptation to contamination levels, and the basis of resistance, would be relevant in the Fal. For example, mapping the genetic composition of tolerant populations of species such as Hediste, in relation to induction of detoxification systems such as metallothionein, would be a way of linking exposure biomarkers to their higher level effects. • Measures of oxidative stress and antioxidant responses Pollutants may increase the intracellular formation of reactive oxygen species (ROS) which are counteracted in cells by the production of antioxidant defences including enzymes and smaller molecules, such as GSH (described above). Total Oxyradical Scavenging Capacity (TOSC) quantifies the ROS scavenging potential of the test organism/tissue and the ability to cope with the various toxicological processes that are linked to oxidative stress. Individual antioxidant enzymes such as superoxide dismutase (SOD), catalase, peroxidase can also be determined along with Glutathione reductase (GR) which plays an essential role in cell defence against reactive oxygen metabolites by sustaining the reduced status of the important antioxidant, glutathione (GSH).. Total Oxyradical Scavenging Capacity (TOSC): In complex field situations such as the Fal, heavy metals may act in combination with other stressors (e.g TBT,
7
eutrophication) to overwhelm the defence mechanisms of the cell. The extent of damage to anti-oxidative systems can be determined by measurement of a reduction in TOSC as a general marker of oxidative stress (reduced capacity to neutralise various ROS). Under certain chronic exposure circumstances however an increase in TOSC may be seen as a counteractive measure by the organism to pro-oxidant challenge. It is therefore crucial to have appropriate baselines and reference samples. Total Oxyradical Scavenging Capacity (TOSC): TOSC quantifies the Reactive Oxygen Species scavenging potential and hence the ability to cope with the various toxicological processes which are linked to oxidative stress (including pollutants and fluctuating DO levels associated with eutrophication). Methodology is based on the measurement of rates at which different ROS species oxidise substrate (keto-γmethiolbutyric acid) to ethylene gas, measured by GC-FID (Regoli et al., 2000). Reactive oxygen species (ROS) can be determined directly. For example, generation of ROS can be measured in blood cells, using microplate assays to measure SODinhibitable reduction of cytochrome-C and nitroblue tetrazolium. In many respects however TOSC may be a preferred biomarker as it represents a more integrated measure of the organisms resistance towards oxidative damage. • Other biochemical markers Changes in plasma vitamin levels (e.g. vitamin A, B), measured by HPLC, have been proposed as a biomarker of health in vertebrates and are known to be affected by pesticide, PCB and PAH exposure in the laboratory. In the field it is more difficult to discern the causes of variability in vitamin levels as they are susceptible to numerous biotic factors. Mitochondrial activity (correct functioning of the ATP producing organelles) and ATPase activity (measured by electrophoresis) is also thought to be symptomatic of health of the organism but is yet to be validated fully in field applications. Increases in porphyrins, measured by HPLC, may signify impacts on haemoglobin synthesis (animals) and chlorophyll production (plants) and have been linked with the presence of halogenated aromatic hydrocarbons, organochlorines and Fe. Variability is high however and the assay is not suitable for the majority of invertebrates. The same is true of a number of enzymes suggested as biomarkers of stress in fish including sorbitol dehydrogenase, lactate dehydrogenase, glutamate dehydrogenase and glutamate pyruvate transaminase.
3.1.2 Molecular and cellular biomarker approaches • Immune function, Neutral Red retention, peroxisomal proliferation
8
Immunotoxicity Assays: These simple low-medium cost assays measure the immunocompetence of haemocytes from invertebrates (and fish), reflecting both the extent of exposure to immunotoxins and the general well-being of the test organism. For example, they may be used to assess the role of environmental contaminants in mediating bacterial infections in shellfish. There are a variety of determinands including total and differential blood counts, phagocytosis and susceptibility to infection. Immune function in invertebrates can be determined by measuring the ability of haemocytes to phagocytose (ingest) dyed particles. The phagocytic index is then expressed relative to the number of viable cells. In fish the number and size of pigmented macrophage accumulations tends to increase at contaminated sites and it is suggested that they represent immunotoxicological markers and show potential monitors of fish health. Lysozyme, an enzyme with antibacterial and antiviral activity (causing lysis of bacterial cells), has also been proposed as a surrogate marker of immune system function, but further validation is required before it can be routinely applied (Scmitt and Dethloff et al 2000). Laboratory studies indicate this type of assay is sensitive to a variety of contaminants, although temperature, age and nutritional status are among a number of potentially confounding features. Immunosuppression is likely to occur at dosages lower than those of many other commonly used toxicological indices and therefore has considerable early warning potential. Since there are strong links between immunity and disease resistance it seems rational to incorporate and link results for both measures in schemes aimed at ‘health assessments’. More general histopathology may be considered also (see below and section 4). • Other Cellular and molecular markers Membrane integrity and detection of cytosolic enzyme leakage has been applied as a biomarker e.g. the neutral red retention assay (histochemical/microscopic assay for lysosomal membrane stability) Neutral red retention assay measures the retention time of the dye Neutral Red in lysosomes 2 and provides an indication of membrane stability. It has been applied to fish and many different invertebrates including bivalves, polychaetes, crustaceans and echinoderms and suggested as a proxy for the health of the organism. The neutral red retention assay has been shown, experimentally, to correlate with higher order effects (growth, reproduction and survival), though it is only marginally more sensitive than lethality. In the Fal, acute conditions in Restronguet Creek have been confirmed with this biomarker, though effectiveness at distinguishing gradients is less certain and establishing cause and effect is difficult. This general stress assay is also semi-subjective to an extent (depends on visual observation and counting) and 2
Lysosomes contain acid hydrolases and are involved in digestion and sequestration/detoxification of toxic chemicals: membrane stability is reduced in response to a variety of stressors. Peroxisomes are ubiquitous organelles involved in lipid metabolism and oxyradical homeostasis and are characterised by their ability to increase in number and activity ( fatty acid oxidation enzymes such as palmitoyl-CoA oxidase) hence the term peroxisomal proliferation.
9
is probably best used as part of a suite of biomarkers. Histological examination of lysosomal and peroxisomal2 numbers, intracellular accumulation of neutral lipids (volume density) and spectrophotometric determinations of peroxisomal acyl Co-A oxidase activities (‘peroxisomal proliferation’) in fish livers and mussel digestive glands are promising related biomarkers of exposure - particularly to organic pollutants (Bilbao et al 2006a). The same also applies to measurement of the peroxisome proliferator-activated receptor α (PPARα), by immunoblotting. Peroxisomal proliferation has been proposed as an alternative to the MFO biomarker (in molluscs) but requires further validation in this capacity.
3.1.3 Genotoxicity, DNA damage and chromosomal aberrations Comet Assay: Redox sensitive metals such as Cu or Fe, and others such as Cd, can damage DNA by stimulating the production of intracellular reactive oxygen species (Polle & Rennenberg, 1993:Schützendübel & Polle 2002). Many organic chemicals, UV and ionising radiation can also act as genotoxic agents. The single cell gel-electrophoresis (SCGE –‘comet’) assay is one of several techniques which are ideal for screening for possible genotoxicity associated with point-source and diffuse inputs to the system. (see also ‘micronucleus formation’) The comet assay can detect DNA damage e.g. in molluscan blood cells by detecting double and single-strand breaks in the DNA (Singh et al 1988). A suspension of single cells e.g molluscan haemocytes (1 x 105 per ml) is embedded in agarose, exposed to SCGE and visualised with a cyanine dye (SYBR-gold, Molecular Probes). The resultant slides are examined histologically and specially designed software is used to score damage to DNA. Measurement of DNA unwinding rates in alkaline conditions may be a comparable and complementary technique to that of the comet assay. The method involves measurement, by fluorescence, of the ratio of double strand DNA remaining after a given time period; faster unwinding rates are thought to indicate a greater percentage of strand breaks. Provided that further validation is forthcoming, the technique represents, potentially, a relatively simple, if indirect, in vitro approach to screen for genotoxicity. Micronucleus Formation. An alternative estimation of exposure to genotoxins involves examination of tissues microscopically to determine the presence of micronuclei. The assay involves scoring cells for the presence of cytoplasmic micronuclei associated with the main cellular nucleus DNA adducts represent large hydrophobic xenobiotics which are covalently bound to normal DNA nucleotides. Measurement of DNA adduct formation, based on 32P-
10
post-labelling molecular techniques 3 , has been used to signal impact on genetic material from PAHs and other organics, and represents a higher test option (medium to high cost). DNA alteration may be a precursor of cancerous conditions and, as with other genotoxic responses described above, may therefore have far reaching consequences on the health of affected individuals and populations. Cell death may occur following acute exposure, either as a result of direct bonding of the pollutant to DNA or, more likely, following biotransformations to highly reactive metabolites which could attack the DNA. The presence of adducts can reflect previous (months) as well as recent (days) exposure. Background levels, and the degree of natural variation, have to be well defined in all these assays.
3.1.4. Reproductive markers and Endocrine disruption • Imposex, intersex, reproductive hormones, vitellogenin (VTG) and zona radiata protein (ZRP) induction Biomarkers to evaluate pollutant impacts on reproductive performance include histological observations of germ cells, gonads and associated ducts (e.g. imposex and intersex), signal transduction, hormonal metabolism balance (measurement of sex steroid hormones), gamete physiology and viability, and embryo viability and development. Imposex, the imposition of male sexual characteristics on females is a specific and sensitive biomarker of TBT pollution which is still an issue in the Fal & Helford SAC (section 3.3). Among the most sensitive species (neogastropod molluscs such as Nucella lapillus) imposex is manifested by the irreversible growth of a penis and vas deferens in females, and is quantified by corresponding indices of penis length (Relative Penis Size Index, RPSI) and sequence of development in the vas deferens (Vas Deferens Index, VDSI), as first described by Bryan et al (1986) and Gibbs et al (1987). Imposex is induced at TBT levels of ~1 ng l-1; above ~10 ngl-1 sterilisation leads to population decline. The majority of dogwhelk (N. lapillus) populations in the Fal are known to have succumbed due to the high levels of TBT associated with dockyards and marinas. In the Helford populations are known to have declined in the 1980s (Covey and Hocking, 1987), although there are populations near the mouth in which imposex levels can be monitored. Measurement of imposex in other neogastropods (e.g. Ocenebra, Ocinebrina) may be a useful alternative marker to N. lapillus for this form of endocrine disruption. Hinia reticulata and Littorina littorea (neogastropods) also display a similar response to TBT. However, these species are far less sensitive to the compound than Nucella and consequently they are more widespread throughout the Fal, offering monitoring 3
DNA sample subjected to balanced hydrolysation to nucleoside monophosphates which are radiolabelled with 32P-ATP. DNA adducts are separation by multidirectional TLC and expressed as a proportion of the total number of analysed nucleotides.
11
possibilities. Sex steroids including testosterone and oestradiol influence gonadal differentiation and gametogenesis, ovulation, and spermiation in many fish (though their role in other taxa, including molluscs, is less clear). Direct measurements of hormones may provide an indication of reproductive impairment if properly controlled to reduce stress effects and natural variability. However, a variety of chemicals have been identified as xenoestrogens due their ability to induce responses similar to those caused by the natural oestrogen, oestradiol. The synthetic oestrogen ethinyloestradiol, is a particularly strong inducer, but there are several others less potent, like organochlorine insecticides, industrial chemicals such as octyl- and nonyl-phenol, aromatic halogenated hydrocarbons (dioxins, furans and polychlorinated PCBs) and metals. There are a number of tests to establish possible hormone disrupting potencies such as the induction of blood plasma vitellogenin (VTG) in male fish (e.g. Lye et al., 1998). Induction of zona radiata proteins (ZRP) has also been proposed as a potential biomarker for oestrogenic effects in male fish. Indeed, studies with juvenile salmon have indicated that the ZRP response may be more sensitive than VTG to a range of pollutants (Arukwe et al., 2000), providing a further powerful means of detecting exposure to environmental oestrogens. Inhibition of VTG in oestrogen-induced fish has also been proposed as a biomarker of anti-oestrogens. Measurement of hormones and developmental chemicals such as VTG may be carried out using complex chromatography-MS techniques, as well as biochemical (Enzyme Linked ImmunoSorbent Assay (ELISA) / Radioimmunoassay (RIA)) and molecular methodologies. Alkali labile phosphate (ALP) activity has been proposed as a lowercost surrogate biomarker for VTG though it is generally less sensitive and specific. There are a number of emerging in-vitro hormone receptor/reporter gene assays which are also coming on-line (see section 4). It should be stressed however that symptoms of ‘endocrine disruption’ can result from pollutant-induced changes to a number of pathways, not just the oestrogen receptor. Thus, to be of most relevance, in vitro screening for endocrine disrupting chemicals needs to be accompanied by observations of higher order effects (histology of the gonads, viability of embryos, population structure). In fish, measurement of intersex (e.g. ovotestis– the presence of oocytes in male gonad) is probably the most widely observed higher order ‘biomarker’. Recent observations of intersex clams Scrobicularia plana in the Fal and Helford SAC suggest a potential for screening for possible reproductive impairment (structural) caused by endocrine disruptors, and if properly validated could provide a more practical means of monitoring estuaries than the fish model (see section 3.3). The gonadal-somatic index (gonadal weight as a percentage of body weight) has been used as a measure of reproductive maturity in many studies, particularly in fish, to assess gonadal changes with respect to natural and anthropogenic influences. Alternatively, stages in gametogenesis can be monitored histologically using a variety of staged indices. Should any of these reproductive ‘biomarkers’ signify response, contextual
12
information on higher level impacts – e.g. fertility; abundance and viability of embryos - would be justified in order to understand the reproductive consequences at compromised locations. Imposex development in gastropod molluscs is a highly specific and sensitive biomarker of TBT impact. The condition is irreversible and may therefore signify historical exposure. Intersex measurement in fish and bivalves indicate susceptibility to the effects of other endocrine disrupting chemicals, most frequently oestrogens and xeno-oestrogens. Both of these low cost options involve histological examination of gonadal tissue. Both may be linked to a decline in reproductive health. Other components of the endocrine system have been proposed as general biomarkers of reproductive impairment in fish: these include leuteinizing hormone, follicle stimulating hormone (fsh), ketosterone and progesterone (measured by RIA). These may be influenced by of a variety of organic contaminants and synthetic hormones and also tend to be confounded by numerous biotic factors, hence their application in the field has been relatively limited.
3.1.5 Physiological Indicators
Condition indices, histopathological changes, scope for growth & heart rate monitoring have been used as indicators of physiological condition: (e.g. growth and condition indices, tumour formation, cardiac activity in bivalves; incidence of tumours, parasites and melanomacrophage centres in fish). They represent higher level integrated effects of a number of the biomarker responses described above, and have considerable ecological relevance, though arguably they lack specificity in terms of identifying causes Condition indices usually relate to biometric measures such as weight and length, whilst somatic indices indicate the proportional sizes of certain organs. Condition is a non-specific organism- level response which can be influenced by many factors including nutritional and reproductive status as well as contaminant and pathogen exposure. Somatic indices reflect the fact that the size of certain organs can fluctuate with respect to overall body size, and can also respond to many such factors. Common examples include the hepato-somatic index (HSI) - the weight of the liver (digestive gland in shellfish) expressed as a percentage of total body weight – and the gonadosomatic index (GSI – gonad to body weight ratio). Because these indices integrate response at many levels they are used as standard procedures to indicate overall wellbeing in fish and shellfish physiology studies. The GSI has sometimes been used as a surrogate measure of reproductive status. A key requirement when using condition as a biomarker in site comparisons, is the need to use similar sized organisms where possible and to be wary of seasonal factors which can override other effects. Depending on species and the type of contaminant exposure, these indices can decrease or increase as a result of pollution. Interpretation therefore requires caution. Although a greater weight: length ratio might intuitively be seen as a healthier
13
condition (at the individual level), the presence of fewer but larger, individuals could alternatively signify abnormality at the population or community level. Histology of liver, kidney, gut and gills of fish and bivalves, by light and electron microscopy, can reveal the incidence of tumours (some of which may be cancerous) and other pathologies, with obvious links to higher order effects. The frequency of these lesions can be caused by a wide range of contaminants (especially exposure to carcinogens) and may also be linked to disease susceptibility. Therefore establishing cause and effect using this technique is likely to be difficult and requires considerable expertise; even then, conclusions are likely to be partly subjective. The same also applies to examination of fish eggs for malformations: though these can be demonstrated to be pollutant-induced in the laboratory it is possible that other factors such as temperature could be influential in the field (von Westernhagen et al 2006). Scope for Growth (SfG) has been extensively tested in species such as mussels and is based on whole-organism energy-budget calculations. This relatively high-cost assay gives a holistic view of the physiological functioning of the organism. SfG has sometimes been linked to population-level consequences, but tends to be non-specific and is also influenced by disease and food-availability, in addition to contaminants. Interpretation of SfG results needs careful consideration of ‘controls’. Measurement of food assimilation in mussels (one component of the energy equation) has been used as a surrogate for Scope for Growth and assumes that feeding rate is the major variable in the equation affected by water quality. However food availability and quality will also be major influences. Simple indices of shell to body size ratios can sometimes be useful in establishing condition of bivalves in acute situations. The cellular energy allocation (CEA) is based on similar principals to SfG - in quantifying available energy reserves (measured as carbohydrate, protein and lipid) against energy consumption (an enzymic component of the electron transport system) and so deriving a net cellular energy budget. A small number of trials have been conducted with marine mussels: as with SfG the availability of food may influence the ability of this assay to discriminate pollution effects in complex field situations therefore further validation is warranted. Heart rate provides a general indication of the metabolic status of mussels and crabs and modern techniques permit the non-invasive, continuous monitoring of cardiac activity using infra-red sensors attached to the shell (Depledge and Anderson, 1990). These measures may prove useful indicators of acute effects from high levels of organic enrichment, toxins or eutrophication, including impacts arising from low or variable DO, and ammonium found near discharges. Sensitivity to changes in chronic levels of contamination is probably low.
14
3.1.6
Behavioural biomarkers
The influence of contaminants on behavioural responses have been demonstrated in a number of laboratory studies with marine organisms and mostly involve changes in locomotary activity. For example the failure of shrimp to bury increases with Cd exposure concentrations, whilst successful prey capture decreases in a dose dependent manner (Bird, D., unpublished results). Other suggestions for behavioural responses include abnormal behaviour in polychaete worms and starfish and mollusc rightingresponses (where the time taken to assume correct orientation when turned upside down is hypothesised to be proportional to the degree of contaminant exposure). As the latter may be sensitive to temperature and salinity, it may be difficult to establish cause and effect in the environment. This type of assay is more akin to the general bioassay/Direct Toxicity Assessment described below and is perhaps of less value, diagnostically, in field assessments than a number of the biomarkers described above
This is not a comprehensive list of biomarkers or their properties (for a more extensive review, see Long et al., (2004)). Rather, the procedures outlined here have been selected primarily with regard to their ease of use, practicality, current level of validation, costs and relevance to known environmental problems in estuaries and coastal areas in the southwest and in particular the Fal and Helford SAC.
3.2 Bioassays and Direct Toxicity Assessment (DTA) Broadly speaking the terms bioassay and biomarker have become interchangeable to an extent. Traditionally ‘bioassay’ is defined as an appraisal of the biological activity of a substance by testing its effect on an organism over a range of dilutions and comparing the result with some agreed standard. The protocol originates in the pharmaceutical industry as a means of evaluating no-effects levels and safe dose limits. Nowadays it has been assimilated into ecotoxicological terminology as a means of providing a direct assessment of toxicity of a given contaminant source (point source or diffuse). Equally, bioassays can be applied to assess the condition of the environment, either in situ or by using environmental samples in laboratory trials. Bioassays using appropriate species may be considered to assess the extent of acute effects of complex major discharges (where specific chemical analysis may be difficult), or where bioaccumulation results indicate the possibility of raised body burdens. Assays have usually been conducted on effluent and receiving waters in relation to plumes from major discharges i.e. comparing dilutions required to reduce effluent toxicity to zero (Predicted No Effect Concentration ‘PNEC’ dilution) with measured or modelled contaminant values. Standard test methods for Integrated Pollution Prevention and Control (IPPC) discharges currently use Daphnia magna and Selenastrum capricornutum (alga) for freshwater discharges, and oysters Crassostrea gigas, copepod Tisbe battagliai and the diatom Skeletonema costatum for marine assays. A fish species may be added to the list of standard test organisms in detailed assessments though this practice is becoming undesirable on ethical grounds. There
15
are certain limitations in some of these tests in terms of monitoring point sources, in that they usually address only short-term effects and may not account for the large temporal variations in discharge composition. There is also increasing emphasis on the use of bioassays to identify and qualify the toxicity of non-point discharges and indeed the estuarine and coastal environment in general. Many national and international monitoring programmes (e.g. OSPAR) now incorporate such techniques. The range of water quality bioassays is developing rapidly with techniques involving field deployment or laboratory-based exposures, in vitro and in vivo (whole organism) tests, and the use of selective extraction and concentration steps to identify the water fraction responsible for toxicity e.g. Solid Phase Extraction, Toxicity Identification and Evaluation (TIE). Techniques used in a recent inter-comparison of water quality assays in the North Sea included a range of in vivo and in vitro tests (Vethaak, 2006). Endpoints of in vitro schemes included: pure enzyme inhibition (AChE) together with cytotoxicity, genotoxicity (comet assay), EROD and VTG induction, and reporter gene upregulation (oestrogen, androgen, dioxin receptors) in cell cultures. Whole organism tests include growth inhibition in algae (S. costatum), mortality and immobility in crustaceans T. battaglia and Ceriodaphnia dubai, embryo development in zebrafish (Danio rerio) and oysters (C. gigas) and larval formation in mussels Mytilus edulis. Growth and survival of mussel larvae has also been trialled in the context of the Fal Estuary (see next section 3.3). Most of these assays are capable of providing useful information on effluent toxicity directly and, following preconcentration, in some water column samples. Manipulation of water samples might conceivably introduce some anomalous ‘operational effects’ and claims of success at monitoring chronic pollution levels in water, in absolute terms, are currently more circumspect. Nevertheless in terms of comparative toxicity of point sources, and defining spatial and temporal trends this technique is a useful tool in environmental management. Bioassays can also be adapted for monitoring sediment toxicity. Uncertainty over the impact of contaminants in sediments is particular concern in many SACs (sedimentbased ecosystems form almost a third of all SAC interest features). A relevant bioassay/direct toxicity assessment should be conducted in parallel with other forms of evaluation. For example survival and burrowing responses in key infaunal species (clams, polychaetes and crustaceans) are probably appropriate sublethal measures to consider in addition to the less subtle endpoint of mortality. DTA tests of this nature are undergoing evaluation with species such as Arenicola and Corophium and may incorporate a feeding bioassay (e.g. cast production in Arenicola) as well as mortality as endpoints. As there have been few attempts to apply such tests to the Fal and Helford SAC we have trialled the technique for Arenicola as part of the current biomarkers project. Results from these trials are very promising: casting rates were significantly reduced when Arenicola were placed in sediments from at least two sites in the Fal, Restronguet (outside the SAC) and Truro (figure 1). Mortalities, which occurred in Arenicola in the Truro sediment were related to a reduction in the casting rate,
16
showing a predictive, ‘early warning’ potential for the bioassay (own unpublished results). This behavioural assay can also allow a rudimentary classification of sediments which directly relates to impact on infaunal biota at the sites of origin, therefore may be a valuable addition in environmental management.
CASTING Rate Homogeneous Groups, alpha=.05 (Tukey's HSD) 9.0 c
Mean daily casting rate (10d)
7.5 bc
bc
6.0
bc
Mortality rate Helford 0% Gillan 0% Maenporth 0% Falmouth Marina 10% Flushing 0% Weir Point 10% Old Mine 10% King Harry's Ferry 10% Coombe 0% Malpas 10% Truro 20%
bc ab
4.5
ab
ab
ab ab
3.0 a
1.5
0.0 Helford
Maenporth Flushing Old Mine Coombe Truro Gillan Falmouth Marina Weir Point King Harry Ferry Malpas
Figure 1. Arenicola marina casting rate in sediments from the Fal & Helford SAC For mud and sand habitats, which are widespread in the Fal and Helford SAC, infaunal species such as clams Scrobicularia plana and polychaetes Hediste diversicolor might also be suitable and relevant species for bioassays to evaluate sediments, particularly if they are additionally employed as indicators of bioaccumulation. Thus, using transplanted or native individuals, there are possibilities to study effects in the laboratory and in the field. It would also be useful to examine sublethal-effects in these species (including growth in juveniles and burrowing behaviour) in addition to mortality.
The choice of test organism used in bioassays should ideally be tailored to the questions being asked and the conditions of the site. Toxicity has scarcely been studied in some of the less well-represented and perhaps sensitive species and communities of important conservation status – a major gap in terms of Site Assessment. Bioassay-type assessments are likely to be of most use as a first line of enquiry where
17
risks to biota and uncertainties are considered high. This strategy can be used independently where chemistry is complex or poorly understood, or if there are insufficient ecotoxicological data on the substance(s) of concern. Despite some limitations however, DTA could clearly be useful as a supplementary tool, particularly when conducted alongside appropriate chemical analyses and some of the more subtle indicators of ecological and biological condition.
3.3 Previous biomarker and bioassay deployments in the Fal & Helford SAC In addition to the Arenicola bioassay described above and the ongoing biomarker assessment project (described briefly in section 5), the Fal and Helford has been used on a number of previous occasions to help trial and validate biological effects tools or to try and assess toxicity of parts of the system. These include Toxicity of sediments to bivalves Cerastoderma edule and Scrobicularia plana. In the past (prior to the cessation of active mining) the benthic communities of Restronguet Creek (and perhaps Mylor) have been considered to be impacted and impoverished, with molluscs such as cockles representing one of the more sensitive taxa (reviewed in Langston et al., 2003). Cockles appear to be transient in their occurrence and attempts have been made to establish the cause based on sediment bioassays with Cersatoderma edule. The route of metal uptake in C. edule, seems to be primarily via water - rather than directly from sediments - though desorption of sediment-bound metal may be an important source of dissolved metal. Bioaccumulation during experimental exposure of control cockles to Restronguet muds was found to be negligible, if the sediments were supplied with clean flowing sea water (Bryan and Gibbs, 1983). Under static conditions, leaching of Cu and Zn to overlying sea water led to mortalities; body burdens in survivors were comparable to those of moribund animals from Restronguet. Survival of the deposit-feeding bivalve is also compromised in all but marginal sediments towards the mouth of the Creek (where overlying water may be less contaminated, hence permitting larval settlement). Up-to-date studies of a similar type, including other metals and other species, would be useful in helping to interpret the current threat and primary causes and mechanisms of biological damage in the Fal Estuary. The ability to extrapolate from environmental levels (water/sediment) to lethal burdens or sub-lethal stress indices is also an important goal, but unfortunately is hampered by the lack of information and basic research. Metal adapted populations Not all species are excluded from the upper reaches of Restronguet. Populations of the ragworm Hediste (Nereis) diversicolor (and to a lesser extent F. vesiculosus) have become adapted to the high levels of metals through development of detoxification and exclusion mechanisms (see for example Bryan, 1976). These include metal rich granules (particularly in the epidermis) and also thiolic (MT-like proteins). In the case of Hediste there is a clear genetic component to adaptation, and selection of populations for Cu and Zn tolerance occurs up to 1km downstream of the source
18
(Carnon River) at sediment concentrations above 1000 and 3500 µg g-1 dw respectively (Hateley et al., 1989, 1992; Grant et al., 1989). The temporal stability of this metal tolerance has been demonstrated recently in acute and chronic toxicity tests: Burlinson & Lawrence (2006) exposed H. diversicolor from Restronguet and Mylor Creeks to increasing concentrations of copper sulphate and observed behaviour, such as burrowing, eversion of the proboscis, abnormal crawling, and survival rates. Results showed a clear gradient of tolerance, with H. diversicolor from sites closest to Carnon River displaying fewer metal-stress responses and surviving longer. In this study, the endpoint was death of the worms; however behavioural endpoints can be used to assess sensitivity to contaminants without mortality of the animals (Lawrence & Poulter, 1998). To this end, Burlinson & Lawrence are developing and validating a behavioural assay using H. diversicolor (See Section 4). Large differences in the Cu tolerance of nematode communities from different parts of the Fal have been confirmed (in standard toxicity tests), to correlate with their previous history of exposure (i.e. level of Cu in sediment). Specimens from the more contaminated Restronguet and Mylor Creeks display a marked increase in resistance to Cu compared with individuals from Pill, Cowlands and St Just, though in turn these are more impacted than Percuil. Nematode community composition in the Helford Estuary appears to be comparable to that of the (relatively uncontaminated) Percuil River (Millward and Grant, 2000). Sediment bioassays at the community level support these conclusions: thus, standardised nematode assemblages were observed to vary in response to metal gradients when introduced into Fal sediments, indicating that an experimental microcosm approach can be useful both as a sediment bioassay and in the validation of conclusions drawn from field surveys (Austen and Somerfield, 1997). Mapping (and re-mapping) of tolerant populations (Hediste and others) may be a useful tool in the assessment of condition and would add an interesting temporal dimension to monitoring the anticipated ‘recovery phase' following water-treatment measures at Wheal Jane. This method of establishing tolerance is effective at picking out metal-affected sites and would be a useful complement to conventional ecologically-based monitoring methods and biomarkers. The discovery of adaptation in Hediste and Fucus, and toxicity data for organisms such as C. edule and S. plana, illustrates that organisms are capable of a range of specific responses to the impact of metals. Since the demonstration by Bryan and coworkers that epidermal granules were able to immobilize potentially toxic metals in Hediste, and that a reduction in permeability in Fucus can reduce metal uptake, there have been a small number of studies which have explored the mechanisms responsible in other species. Most work has focused on the adaptations to metals in oysters which partly involves localisation of granular Cu (and Zn), in amoebocytes (blood cells) (George et al., 1978; Pirie et al., 1984). The green discoloration of these cells is proportional to Cu content (Boyce and Herdman, 1898; Orton 1923). Localisation of metals within vesicles, in inert form, enables the oysters to tolerate body burdens of Zn and Cu, which are 10 and 100 times ‘normal’, respectively.
19
Metallothionein induction. Another mechanism which may contribute towards metal tolerance in a range of species concerns the induction of metal-binding protein metallothionein (MT) which signals a sub-lethal response to metal pollution and may, therefore, have potential as an early warning indicator of deleterious effects. In the laboratory, oysters produce MT in response to metals such as Cd, Cu, Hg and Ag and there is some indication from measurement of MT induction in native Fal oysters, that in parts of the system they are attempting to adapt to metal stress (Langston et al., 1998). Thus MT levels in gills of oysters from just inside the mouth of Restronguet Creek were more than three times higher than at the adjacent Loe Beach, mirroring to an extent, Cu levels. The measurement of MT in winkles and in transplanted caged mussels has also been shown to be a useful measure of sublethal stress and biological impact (own unpublished data). Mussel transplantation is a useful alternative to the use of native animals, particularly in parts of the Fal where indigenous populations are absent. Adult mussels transplanted to Restronguet Creek are usually able to survive for periods of several months, allowing investigative studies into the induction of adaptive mechanisms, and the conditions under which detoxifying systems are overwhelmed. Using this approach at the time of the Wheal Jane incident 4 in the early 1990s, we were able to establish that not only is MT substantially raised (fivefold) , but also that there are major shifts in intracellular metal-binding patterns (figure 2) which appear to be a precursor (and diagnostic feature) of subsequent mortality. Mussels at the Old Mine site did not survive much beyond the ‘incident’, whereas mussels transplanted to the mouth of Restronguet, Loe Beach, and Flushing were recovered live up to 6 months later. Copper seems to be the main metal responsible for toxicity in mussels and probably explains their exclusion from Restronguet, along with that of other sensitive species. More recent MT data for Fal mussels – some fifteen years after the flooding incident and after the introduction of water treatment at Wheal Jane – is discussed briefly in section 5. Table 1. Metallothionein levels in mussels Mytilus edulis transplanted to the Fal from a clean site (Whitsand Bay) 6 months prior to the Wheal Jane flooding and recovered 3 weeks after the incident (5th February 1992). Metallothionein concentrations of 2-3 mg g-1 dry wt. are typical for mussels collected in sites free from pollution. Site
Mean MT concentration mg/g dry wt.
8. Opposite Old Mine
10.8
7. Opposite Penpol
4.89
4. Wier Point
5.41
3. Loe Beach
3.86
2. King Harry Ferry
2.97
4
This was a major incident in 1991, when metal-rich mine water overflowed into the Fal estuary from the Wheal Jane tin mine.
20
Cu MT 7000 6000
VLMW
ng/g Cu
5000
looe 18/2/92
4000
whitsand 21/2/92
3000
old mine 5/2/92
2000 1000 0 10
20
30
40
50
Fraction No.
Figure 2. Profiles of cytosolic Cu in mussels Mytilus edulis transplanted to the Old Mine, Restronguet Creek, from a clean site (Whitsand Bay) 6 months prior to the Wheal Jane flooding and recovered 3 weeks after the incident (5th February 1992). Cu levels are significantly raised in metallothionein (MT) and in very low molecular weight (VLMW) pools, relative to clean sites (Whitsand and Looe). MT in certain tissues of the shore crab Carcinus maenas is also a potentially useful biomarker, and has been applied in the Fal Estuary. MT levels in gills responded proportionately to Cu and Zn exposure gradients in sediments at three sites (Restronguet, Mylor and Percuil), and were significantly elevated compared with control sites in the Yealm and Avon estuaries, even following chronic exposure (Pedersen et al., 1997). However, this study showed that in other tissues, for example the mid-gut, the nature of metal binding and handling can confound such relation ships, particularly in the case of Zn, and changes in the relative proportions of Znbound versus Cu-bound MT in this organ appeared to be a more useful indicator of metal exposure than absolute MT concentrations (Pedersen et al., 1997). The MT assay, and associated metal binding profiles, provide a useful and specific measure of bioavailable metal and sub-lethal stress, and could be used to measure spatial trends and improvement or deterioration of condition. The development and application of this biomarker in the context of current work in the Fal & Helford SAC is discussed in section 5. Stress proteins (SP70) Stress protein (stress-70) has been investigated as a biomarker of metal exposure and adverse effects in the shore crab C. maenas from the Fal Estuary (Pedersen et al., 1997). Stress proteins can be induced by a number of trace metals including Cu, Zn, As, and Cd (Sanders et al., 1991; Bauman et al., 1993) suggesting that they would be appropriate for use as a biomarker in the Fal. However, Sanders (1990; 1993) showed
21
that there are also other potent inducers of stress-70 (SP70), including elevated temperature, salinity stress, oxidative stress, UV light, xenobiotics, teratogens and hepatocarcinogens, therefore SP70 is a relatively non-specific indicator of cellular stress. Pedersen et al., (1997) found that SP70 levels in gills of C. maenas were highly variable within and between sites and did not reflect the sediment metal gradient of the Fal. Furthermore no correlation was found between MT levels (see above) and SP70. This study represents one of the first to measure more than one biomarker simultaneously and demonstrates the value of a multi-biomarker approach when conducting ecological impact assessments. Endocrine Disruption and related reproductive ‘biomarkers’ TBT and imposex The evidence for TBT impact on dogwhelk populations in the Fal stems from long-term observations of populations (for example, Crothers, 1975, EA and MBA). In 1972-73 dogwhelks were common throughout the estuary as far upstream as King Harry Ferry. By 1984, populations had disappeared from Weir Point, Mylor and Falmouth and only very small numbers were present at the other sites. By 1992, the Castle Drive population, outside the estuary mouth, represented the northern limit of distribution of the species in the vicinity of the Fal Estuary. TBT concentrations in dogwhelks from Castle Drive appear to have reached a peak in 1987, when the ban on the use of TBT paints on small boats was introduced (Figure 3 from Langston et al. 1994). Subsequently, tissue TBT concentrations declined very slowly. Indices of imposex remained high in this surviving population and TBT concentrations were high enough to sterilize the majority of females (figure 3). Since this TBT-affected population lies outside the mouth of the estuary, it seems unlikely that recolonisation of the Fal, to early 1970s levels, will occur in the foreseeable future. Unpublished records of the EA (N. Babbage, EA, pers comm) and MBA confirm that Castle Drive and others (St Mawes and Towan Beach to the east of the estuary) are probably the last remaining viable Nucella populations in the Fal area. A small number of individuals were observed at St Anthony in September 2005 (own unpublished data).
22
2000
20
N. lapillus
Falmouth
16
1500
12 1000 8 500
4
0
TBT in water (ng Sn/l)
Organotin (ng Sn/g dry wt)
TBT Ban
0 Feb-85 Oct-85 Sep-86 Aug-88 Sep-90 Jun-85 Feb-86 Sep-87 Feb-89 Oct-91
Percentage
100
TBT+DBT fem
Falmouth N. lapillus
TBT male
TBT+DBT male
TBT water
6
TBT Ban
80
4.8
60
3.6
40
2.4
20
1.2
0
VDS index
TBT fem
0 Feb-85
Oct-85 Jun-85
RPS Index (%)
Sep-86 Feb-86
% Females
Aug-88 Sep-87
Sep-90 Feb-89
% Sterile fem.
Oct-91
VDS Index
Figure 3 Nucella lapillus: (top) organotin concentrations in tissues of Falmouth (Castle Drive) animals prior to and after the ban in 1987. TBT values in water also plotted. (Bottom) Imposex indices in Falmouth females (From Langston et al., 1994) At Falmouth the European sting winkle Ocenebra erinacea is also affected by imposex. However, since it is slightly less sensitive to TBT than Nucella lapillus a small percentage of females still appear capable of breeding. On the basis of dogwhelk abundance, the Helford too has been substantially affected by TBT pollution (and may still be in parts). Around the time of the 1987 ban on use of TBT-based antifouling paints, Covey and Hocking (1987) noted a ‘dramatic 23
decline in numbers of dogwhelk Nucella lapillus due to TBT’ and recorded that Nucella populations were present only at outer headlands Nare Point and Rosemullion Head. At Maenporth, on the open coast between the Fal and Helford, there are indications of a reduction in imposex levels with the RPSI falling from 33 in 1985 (Bryan et al 1986) to 8 in 2001 (Birchenough et al., 2002). Since TBT remains a concern for the integrity of the site, updates on populations of neogastropods and the extent of TBT impact from major sources should be carried out at intervals using established measurements of imposex indices and chemical analysis, preferably at established sites (Bryan et al 1986; Gibbs et al 1987; Langston et al 1994; Harino et al 2003; EA unpublished surveys). This is one of the more reliable biomarkers to contribute towards site assessment due to the selectivity and sensitivity of the response. Intersex (feminisation). Information on other endocrine disrupting effects in the south west European Marine Sites is negligible. Oestrogens and xenoestrogens are logical priorities for concern, based on observations in freshwater systems throughout much of the UK. Tests to establish possible hormone disrupting symptoms in fish should be combined with efforts to screen for EDs and their effects in benthic invertebrates (histology and population structure, viability of embryos). To this end some recent studies have indicated that the sexuality of clams Scrobicularia plana is vulnerable to environmental influences and is manifested by the presence of oocytes in male gonads (ovotestis). The precise causes of this intersex condition are not fully understood but frequency of intersex in males is seen as a function of anthropogenic disturbance including effluent from Sewage Treatment Works (STW) and possibly also agriculture (Chesman and Langston 2006; Langston et al., in press). Screening of populations around the Fal &Helford, undertaken as part of the current project, indicates the presence of intersex in a number of locations albeit at a low-level, with only one population free of the condition (figure 4). Further screening at sites subjected to differing degrees of anthropogenic influence (including close to major STW) should be considered using this cost-effective assay (together with more fundamental research to establish cause and effect relationships and population consequences). Intersex, Fal & Helford August 2006 30
% affected males
25
20
15
10
5
0 Froe
Penpol
Old mine
Penryn
Truro
Gweek
Polpenwith
Figure 4. Intersex (ovotestis) in the clam Scrobicularia plana and frequency of the condition in male clams from sites in the Fal and Helford SAC, August 2006.
24
Physiological markers The ability to maintain homeostasis of normal physiological status may sometimes be compromised in the presence of contaminants and this appears to be the case at metal polluted sites in the Fal. Thus, crabs Carcinus maenas from Restronguet may be less able to adapt to changes in salinity (osmoregulate) compared to individuals from control sites (Bamber and Depledge, 1997). Multiple biomarkers-Linking lower and higher level effects In an attempt to examine and link effects at increasing levels of biological organisation, Perryman (1996) conducted a long-term study in the Fal Estuary, to compare and contrast monitoring techniques from cellular to higher level effects. At the cellular level, Perryman (1996) investigated lysosomal stability in mussels Mytilus edulis from Percuil, St Just, Pill, Restronguet and Mylor, using the neutral red assay to evaluate contaminant-induced damage. Differences were discernable between Restronguet and the remaining sites, although results did not reflect metal concentrations in sediments. Scope for growth (SfG), an individual organism biomarker, was investigated in M. edulis using measurements of feeding rate, food absorption efficiency, respiration rate and body condition index. The ultimate SfG values were not correlated with metal contamination, however components of SfG appeared to be more useful: respiration rates increased proportionately to increasing metal concentrations, and subsequent mussel mortality was highly correlated to both the metal gradient and body condition index values. Still at the individual level, survival bioassays were carried out with M. edulis larvae to establish reproductive and developmental success. Larvae could not be reared from Fal Estuary mussels, therefore clean site mussels were used to produce larvae against which test water samples could be used. Again, results (numbers and survival of normal ‘D’ shaped larvae) did not correspond to the known metal gradient, although larvae did respond to the extreme metal contaminated water of Restronguet Creek displaying the lowest survival of healthy ‘D’ shaped larvae (larval survival> cell damage> scope for growth>population/size frequency analysis. With the exception of population level, the techniques used in this study detected the extreme pollution of the most contaminated site, Restronguet Creek, but did not accurately reflect the known metal
25
gradient of the sites. The thesis contains lengthy discussion on advantages and disadvantages of the techniques used, and the value of an integrated approach.
H
Pill
Restronguet 1 2 3 2
4
5 2 1 3 4 4 3 6 5 5 7 6
Mylor 5 1
F
2 3 4
G St.Just 1 2 3 4 5
E
5 4 3 2 1
Percuil
D 1
Falmouth
C
B A Extent of the SAC
10 km.
Figure 5. Biological effects studies in the Fal and Helford SAC.
26
Figure 5 (cont.) Key: Biological effects studies in the Fal and Helford SAC Black Circles
Red Circle
White Circle
White Square
Yellow Square
Red Square
Blue Square
White Triangle
Site
Site
Site
Site
Site
Site
Site
Site
Gillan (A)
Percuil 3
Percuil 1,2,4,5
Castle Drive (1)
Devoran (2)
Opp. Penpol (3)
River Kennal (4)
Restronguet
Helford (Treath) (B)
St.Just 3
St. Just 1,2,4,5
Opp. Penpol (5)
Mylor
Maenporth (C)
Pill 4
Pill 1,2,3,5
Mylor Creek (6)
Percuil
Falmouth Marina (D)
Restronguet 3
Restronguet 1,2,4,5,6,7
Flushing (E)
Mylor 2
Mylor 1,3,4,5
Key Species Hediste diversicolor
Key Species Carcinus maenas
Weir Point (F) King Harry Ferry (G) Malpas (H) Key Species Mytilus
Key Species
Key Species
Mytilus edulis
Mytilus edulis
Key Species Nucella
Arenicola
Nepthys hombergi
Nepthys hombergi
Ocenebra
Key Species Nereis diversicolor
Key Species Cerastoderma
Littorina Biomarker Metallothionein
Biomarker Neutral Red
Biomarker Metallothionein
TOSC
Scope for Growth
Stress protein
Biomarkers (mussels/crabs) Phagocytosis Metallothionein
Other measures
Other measures
Heart rate
Community Structure
Community Structure
Esterase enzymes
Population Size Structure
Population Size Structure
Micronucleus formation PAH metabolites Bioassay Intersex/imposex
Bioassay Imposex
Bioassay Metal tolerance
Arenicola cast production, mortality
Bioassay Sediment bioassays -
Bioassay Acute and chronic
mortality
toxicity tests
Feeding rates (mussels) Refs Natural England FST20-18-028 (in progress)
Refs Perryman, 1996
Refs Perryman, 1996
Refs
Refs
Langston et al 1994
27
Bryan, 1976
Refs Bryan and Gibbs, 1983
Refs Burlinson and Lawrence, 2006
Refs Perdesen et al, 1997
4. Potential role and state-of-readiness of other biomarkers/bioassays used nationally and internationally The majority of biomarkers described above are either being used or being developed for use by researchers within the SW. Other biological effects techniques in current use in monitoring programs include: Cell-based in vitro methods (e.g. those involving primary fish cell lines such as RTG2, from trout) are increasing in their scope and application Although there are limitations (particularly surrounding ecological relevance) these assays may eventually help to bridge the gap between chemical measurements and population level effects, particularly in identifying cause-effect relationships. They are relatively rapid, capable of high throughput and also avoid the criticism of sacrificing whole organisms which is often aimed at in vivo toxicity testing. Examples of endpoints measured in these cell-based systems include EROD, VTG, genotoxicity (comet assay), cytotoxicity (crystal violet stain) and activity of various reporter genes e.g. receptors involved with reproductive hormones and aryl hydrocarbons. The H4IIE rat hepatoma cell line bioassay is another in vitro test which has been used to semiquantify the catalytic activity of cytochrome P4501A (CYP1A), as 7-ethoxyresorufinO-deethylase (EROD) activity in cultured rat liver cells exposed to environmental extracts. Similar caveats apply as to standard EROD measurements in biota, and again there is issue over ecological relevance: nevertheless, as a screening technique it is beginning to be evaluated in environmental surveys (USGS monitoring programme, Schmitt and Dethloff, 2000). Haemocytes from, for example, mussels can also be used to perform non-invasive tests, including the neutral red retention assay. Broad spectrum, cell-based tests include Microtox which is a commercially-available acute-toxicity assay based on stress-induced changes in the luminescent properties of bacteria Vibrio fisheri. Mutatox is a genotoxicity assay which uses a muted version of V. fisheri and assumes that, following exposure to genotoxins, mutations will restore some of the bioluminescence. Mechanistic assays, such as DR-, ER- and AR-Calux are much more specific (to dioxin-, oestrogen- and androgen-like compounds respectively) and are discussed later in this section. Fish and shellfish disease and histopathology. This has already been trialled as indicated earlier and is beginning to be used more extensively: the gradual assembly of results and field trends will undoubtedly pay dividends in the long-term. A number of studies have revealed that fish in severely polluted areas have a higher frequency of gross lesions than in similar, less polluted habitats, although their development is probably due to a combination of factors including susceptibility and abundance of pathogens rather than contaminants per se. In addition to contaminants, species, sex, age, nutritional status, season, and food habits may all influence the appearance of tissues. Nevertheless, the prevalence of visible lesions including fin erosion, skin ulcers, eye disorders, tumours, skeletal deformities and parasitism has been proposed as a measure to evaluate the health of fish populations (along with other biomarkers, community metrics, such as species abundance, trophic composition). Lesions may
28
signify that defensive measures at the immunological, physiological, biochemical level have been overwhelmed, though the condition may be reversible and is not necessarily life-threatening (Schmitt and Dethloff, 2000). Predictive capability, based on gross lesions, is currently limited and seldom is there a clear indication of cause - though, in their favour, fish diseases are an indicator of cumulative stress which provide an ecologically relevant endpoint. Histopathological studies may therefore be useful in an early warning capacity but tend to rely on descriptive comparisons: few, to date, have used quantitative methods that can then be compared statistically. However, in the UK, CEFAS includes the incidence of fish and shellfish disease and various histopathologies routinely as part of its regular monitoring protocol under the National Marine Monitoring Programme and is developing schemes based on frequencies and severity to provide numerical comparisons (see for example CEFAS, 2005). In such surveys disease symptoms observed in fish include lymphosystis, epidermal ulceration, epidermal papilloma, hyperpigmentation, macroscopic liver lesions and number of parasites. Also recorded is the presence of pre-neoplastic and neoplastic toxicopathic lesions (possible biomarkers of carcinogenesis) including hepatic foci of cellular alteration (FCAs) and hepatocellular adenoma, hepatocellular fibrillar inclusions (an unusual proliferation of the rough endoplasmic reticulum and/or an extensive formation of microtubules). As well as a variety of possible lesions there may be responses in cell and tissue organisation which signify pollutant-induced impacts. In the digestive glands of mussels for example, pollution and other stressors can cause a shift in cell type from predominantly digestive types towards an increasing proportion of secretory/basophilic cells (Bilbao et al., 2006b). Digestive cells also respond to contaminants by alterations in the lysosomal system, including enlargement and an increase in lipofuchsin production. In fish and, to an extent, mussels, histological appearance of gonadal tissue, including prevalence of oocyte and sperm degeneration (and the rate of apoptosis – see below) could indicate the potential effect of contaminants (Feist et al., 2006), though as recycling of gamete material occurs naturally, to serve metabolic needs, it may prove difficult to establish cause and effect. There are other developing methodologies of potential relevance but, as far as we are aware, they have not yet been proven to answer the marine issues being posed by Natural England. Studies involving protein expression in mussels, using 2D gel electrophoresis, are able to show qualitative differences in response to pollutants and pollution gradients, although without identification of the proteins involved it is difficult to interpret mechanisms of action. In the lab, responses have been induced by PCBs, metals and salinity stress. The absence of a sequenced genome for Mytilus and most other invertebrates therefore currently render this approach impractical from a routine standpoint (Grosvik et al 2006). New and promising developments in the biomarker field are genomics and proteomics. These are novel molecular and cellular biomarkers and include the use of antibodies and mRNA probes, cDNA microarrays (transcriptomics) or advanced
29
proteomics approaches (Maldi-Tof-Mass Spectrometry), to detect many of the proteins and mRNAs whose synthesis is stimulated or inhibited by pollutants. Genomics is based upon the application of DNA microarrays that allow the expression of hundreds (and up to many thousands) of genes to be monitored simultaneously, thus providing a broad and integrated picture of the way an organism responds to a changing environment (Gracey et al., 2001). The entire protein complement of the genome, the ‘proteome’, can now be analyzed for changes associated with specific treatments, using ‘peptide mass profiling’, a combination of two-dimensional gel electrophoresis and mass spectrometry (Shepard et al., 2000). Proteomics research can provide certain protein expression signatures (PESs), which are specific sets of proteins, present or absent, indicating specific toxicity profiles (van der Oost et al., 2003). This approach is potentially more sensitive and easier to use than that based on the use of the traditional cytochemical/biochemical "core" biomarkers suggested for the Fal & Helford monitoring program, though at present widespread validation in the marine field is still required. Thus, in laboratory exposures microarray analysis has been used to investigate the changes in cDNA abundances (gene activity) in response to specific chemicals and to obtain an image of the molecular mechanisms underlying homeostasis. For example, in the case of copper toxicity (in microalgae), there is a strong upregulation of transcripts encoding enzymes involved in oxidative stress defence mechanisms rather than the essential Cu-containing polypeptides seen in low Cu situations. In both cases, there were significant changes in expression levels of transcripts coding for enzymes involved in several metabolic pathways (photosynthesis, pentose phosphate pathway, glycolysis, gluconeogenesis), in general stress response (heat shock proteins) and in intracellular proteolysis (lysosomal enzymes, proteasome components) (Jamers et al., 2006). It is understood that trials are also underway with a mussel microarray. At present however it does not appear to be established widely how these changes in gene expression are linked to changes in protein abundances and activities in the field or how best to deploy this relatively expensive assay in the field. Similar arguments apply to assays for apoptosis (programmed cell death) such as Terminal deoxynucleotidyl transferase mediated dUDP biotin Nick End Labelling (TUNEL). This histological technique allows the detection of 3'-OH groups at the end of apoptotic and necrotic DNA fragments in dying cells. Cefas has been able to detect apoptosis in germ cells of fish using this technique (Feist et al 2006) though much more validation will be needed to distinguish between pollutant-induced and natural variation in cell turnover before this relatively expensive technique is routinely applied in the field. It, too, will have to be used as part of a suite of biomarkers as it appears to be sensitive to a large range of stressors (temperature, food availability, DNA damage) in addition to contaminants including TBT, metals, PAHs and crude oil. It will be particularly interesting to investigate any links between apoptosis in gonadal cells and reproductive output. Measurement of stable nitrogen isotopes offers a direct means of source identification and incorporation into food chains because the two major sources of nitrate in many agricultural areas, fertilizer and manure, generally have isotopically distinct δ15N values (Schmitt and Dethloff , 2000). Comparative measurements of δ15N (and δ13C)
30
abundance have been made in Nereis populations from a number of UK estuarine sites, including Restronguet and Mylor Creeks, though no clear distinguishing features were noted in the Fal samples (Nithart, 2000). This biomarker remains a highly specialised and expensive technique, requiring detailed study of individual sites and food chain relationships, and is not widely available or routinely applicable. Nevertheless it clearly has potential in an investigative capacity - as a means of differentiating animal waste (including human sewage) from fertilizer inputs, interpreting the results of reproductive biomarkers (ie determining sources) and detecting causal mechanisms (sources of nutrients) behind certain eutrophicationrelated harmful algal blooms, fish kills and disease outbreaks. Metabolic profiling of body fluids, using nuclear magnetic resonance (NMR), has also been proposed as a means of diagnosing exposed organisms, though whilst this approach has potential it will be difficult (and expensive) to establish cause and effect relationships. Emerging biomarkers include the CALUX (chemically activated luciferase reporter gene) assay, adapted and validated for use in blood samples to provide evidence of organochlorine and PAH exposure, expressed as a toxic equivalency factor (Murk et al 1997). However whilst this signifies exposure to dioxin-like chemicals there is no information on actual effects. Based on similar lines, ER/AR Calux is a proposed in vitro assay for estrogenic and androgenic (endocrine disrupting) activity, as are the yeast-based oestrogen and androgen screens (YES/YAS assays) described by Desbrow et al. (1998). In trials with chemically-characterized water and particulate samples, Murk et al. (2002) recently compared three of the most widely used in vitro assays - two reporter gene assays (YES and ER-CALUX) and an oestrogen receptor (ER) binding assay. The responses with ER-CALUX and YES were more comparable with each other than with the ER binding data, probably explained by the different principles of these assays. Comparisons with chemical data revealed trends that were consistent throughout, although the accurate prediction of higher order biological responses, based on concentrations and oestrogenic potencies of compounds, has yet to be achieved widely. A complex variety of mechanisms (including the presence of unquantified compounds) may confound straightforward relationships between measured concentrations of known oestrogens and in vitro responses (Cargouet et al., 2004). These in turn are not always directly related to in vivo responses, since they do not take into account processes such as bioavailability and metabolism. Consequently, although there are many advantages of in vitro assays, including faster throughputs and lower cost, ability to predict whole organism response may not always be relied upon. Particular care needs to be taken when extrapolating to reproductive and ecological impacts in nature. In vitro testing of oestrogenicity (e.g. based on ER binding or ER mediated gene expression) needs to be scaled in tandem with relevant in vivo tests to confirm their validity when used as a means of assessment of oestrogenicity in field situations, or when making regulatory decisions regarding the endocrine activity of chemicals. Nevertheless, in vitro assays are potentially useful screening tools and may provide useful comparative insights into the relative sensitivities of (xeno)oestrogens and androgens and their modes of action. Morphologically intermediate papilla syndrome (MIPS) is a morphometric marker for oestrogenic influence specific to the sand goby (Pomatoschistus minutes). This index, based on morphological abnormalities the uro-genital papilla (UGP), is currently
31
under development for eventual deployment as an indicator of (xeno)oestrogenic impact in marine and lower estuarine environments. The uro-genital papilla is an organ which shows secondary sexual characteristics and is known to be involved in gamete deposition. Results to date include the finding that some wild male sand gobies from the Tees, Mersey and Clyde have a UGP shape that is intermediate between male and female. This condition was rare or absent in the Alde (the clean reference estuary), Crouch and Thames. However, the destructive nature of the assay and high cost may limit application, furthermore the biological significance is, as yet, unclear. Sensitive mRNA/ELISA VTG assays, currently being tested and applied in marine fish such as the viviparous blenny and flounder, are more likely to have widespread uptake in monitoring programmes screening for endocrine disruption. Production of spiggin (a nest glue protein) in kidney of female three-spined stickleback Gasterosteus aculeatus is currently being validated by CEFAS as a biomarker for androgenic compounds (natural and xenobiotic). Although stickleback are found in estuaries and littoral marine habitats, the main use of this medium-cost ELISA-based assay is likely to be in freshwaters. An increase in kidney epithelial cell height, measured histologically, in the same organism is also being trialled in a similar capacity though its sensitivity to androgenic compounds is considered to be lower than that of spiggin production. Behavioural assays have been little used in monitoring programmes, although they can potentially offer a non-destructive method of assessing effects of contaminants in environmental samples. Several studies have yielded promising results: when challenged with increasing concentrations of Cu, the grazing behaviour of estuarine and nearshore copepods (Acartia tonsa, Acartia hudsonica and Temora longicornis) altered (Sharp & Stearns, 1997). Grazing activity of the copepods was affected at environmentally relevant levels of Cu, leading the authors to conclude that grazing activity was a more sensitive measure of sublethal Cu stress, in comparison to acute toxicity tests. The potential role of the estuarine amphipod Gammarus duebeni in sub-lethal ecotoxicology studies was examined by Lawrence & Poulter (1998): Pleopod beat frequency showed a complex response to copper which was concentration and time dependent, however swimming endurance against a head flow of (Cu-spiked) water, showed a clear dose-response relationship and was identified as the more consistent technique. Using this assay, significant impairment of swimming endurance was observed when challenged with Cu, pentachlorophenol (PCP) and B(a)P, indicating that this behavioural assay using G. duebeni is a useful indicator of pollution impacts in estuarine areas (Lawrence & Poulter, 1998). In a further study by the same authors, the assay was shown to be responsive at similar environmental Cu concentrations to those periodically experienced at some locations on the Humber estuary, UK (Lawrence & Poulter, 2001). Advantages of this bioassay is that it is relatively quick and easy to perform, and reflects the competence of the animals over a range of integrated physiological processes. During standard toxicity tests, Burlinson & Lawrence (2006) observed a range of behaviours exhibited by Hediste diversicolor and established behavioural categories that indicated metal stress. The authors went on to develop and validate a nondestructive behavioural bioassay which can determine the copper tolerance of ragworms and allow the separation of tolerance phenotypes without mortality to the worm (Burlinson & Lawrence, 2007). As H. diversicolor were not affected by
32
consecutive bioassays, it was proposed that tolerance to more than one metal could be determined for individual worms. The biomarkers currently under development, in addition to proteomics (changes in protein expression) and transcriptomics (changes in gene expression) therefore have high potential in terms of holistic views of pollutant exposure but with a few exceptions, are not yet validated for routine use and/or are relatively high cost options. Linking biomarkers to predict higher-level consequences of toxicant exposure Few studies have related toxicant-induced effects on molecular and cellular responses to physiological effects and fitness impairment. This type of multifaceted approach is important because it will improve our ability to use molecular biomarker responses of organisms to predict higher-level consequences of toxicant exposure, and progress our understanding of mechanisms through which toxicants impair fitness. With the obvious exception of TBT-induced imposex in dogwhelks causing population and community level change, one of the difficulties in embracing the biomarker approach concerns the absence of clear mechanistic links between some of the responses and manifestations of ‘harm’. This will probably only be fully addressed following a substantial number of trials to confirm the ‘weight of evidence’ approach. To date there have been a number of attempts to progress our understanding of the mechanisms through which toxicants impair fitness. These involve both laboratory and field trials to demonstrate value of an integrated, multiple-biomarker approach e.g. to relate molecular, cellular, and physiological effects to fitness impairment. Table 2. Examples of multiple biomarker trials and attempts to link to higherlevel effects.
Organism
Field/ Lab
Responses measured
Reference
Mytilus edulis Nephtys hombergii
Field/Lab
Perryman 1996
Fish (smelt) Atherinops affinis
Lab (Cd)
Lysosomal stability, Scope for growth, Larval survival, Population (size/frequency), Community (multivariate analyses of benthic macrofaunal structure) Responses/metal-gradient relationships Biomarker interrelationships relationships between physiological responses (i.e. feeding and respiration) and growth relationships between biochemical and cellular effects and growth (DNA fragmentation more responsive than metallothionein levels) Indicates potential usefulness an early warning indicator of adverse effects
33
Rose et al., 2006
Freshwater fish e.g. common Carp, bass catfishes, suckers percids, trout and suckers
Field Biomonitoring of Environmental Status and Trends Programme (BEST, USA).
Fish such as flounder dab and plaice; some marine invertebrates Pelagic fish (cod, saithe, herring, mackerel), fish larvae, fish eggs. Microzooplankton, phytoplankton, zooplankton, mussels,
Field (Cefas, NMMP)
Shore crab Carcinus maenas Limpet Patella vulgata, mussel Mytilus edulis
Lab
Red mullet Mullus barbatus Rainbow trout Oncorhynchus mykiss hepatocytes Mosquitofish Gambusia affinis
Field
Dab Limanda limanda
Field
Lined shore crab Pachygrapsus crassipes, longjaw mudsucker Gillichthys mirabilis Mussels Mytilus galloprovincialis oysters Crassostrea gigas Fish Boops boops, Mullus barbatus Uranoscopus scaber
Field
Field and lab
Suite of biomarkers, bioassays, chemistry, community ecology.
Schmitt and Dethloff, 2000
Cytochrome P450 1A enzymes (EROD), VTG, reproductive hormones, immune function, histopathology General indicators of health were external appearance (lesions, parasites etc and various somatic indices (relative weights liver, gonad, kidney) EROD, bile metabolites, VTG, DNA adducts, histopathology, patterns of benthic communities Chemistry Suite of biomarkers, bioassays, chemistry.
e.g. Cefas, 2005
Hylland et al., 2006
Cytochrome P450 1A enzymes (EROD), VTG, MT, TOSC, AChE, GST, BaP hydroxylase, peroxisome proliferation, accumulation of neutral lipids,
Lab Lab
Field
Bioaccumulation, malformation of fish eggs, viral diseases in fish, lysosomal responses in hepatocytes, histopathological alterations, parasite prevalence, Suite of biomarkers: Lysosomal stability (neutral red retention), neurotoxicity (AcHE), metabolic impairment (total haemolymph protein), physiological status (heart rate), induction of MT
Brown et al. 2004
TOSC, MT, Cytochrome P450 1A enzymes (EROD), GST Metabolic inhibition, DNA damage, EROD
Regoli et al., 2002 Tollefsen et al., 2006
EROD, BaP monooxygenase, VTG, porphyrins, PAH bile metabolites, esterase, catalase, micronuclei Histopathology, Comet assay, DNA adducts, PLUS Potential causative mechanisms (composition and abundance of benthic communities, sediment toxicity, cellular biomarkers of toxicity in biota samples) DNA strand breaks, AChE, apoptosis, Cytochrome P450 1A enzymes, choriogenin proteins, histopathology
Casini et al., 2006
DNA strand breaks, micronuclei
Bolognesi et al, 2006
34
Lyons et al., 2006
Anderson et al, 2006
Mussel (Mytilus galloprovincialis
Field
Green mussel, Perna viridis,
Fieled
Seagrass Posidonia oceania
Lab and Field
Biochemical (aldehyde dehydrogenase and catalase activities), cellular (neutral red retention time), physiological (survival in air and condition index) Antioxidant enzymes: SOD, CAT, lipid peroxidation product malondialdehyde (MDA), GST, AChE MFO activities (NADPH-cytochrome reductase, ethoxycoumarin-O-deethylase (ECOD)), antioxidant enzyme assays (guaiacol peroxidase (GPOX), superoxide dismutase (SOD))
Nesto et al, 2005 Lau et al, 2004 Bucalossi et al, 2006
These multi-tiered approaches begin to allow assessments of how biochemical, molecular, cellular, and physiological biomarker responses relate to higher order effects, and help establish which biomarkers would be most useful in predicting adverse effects on fitness. Although the topic is still in its infancy there are positive signs as to the value of this approach, vindicating application to sites such as the Fal and Helford SAC.
5. A biomarker strategy for use in the Fal & Helford SAC This review has, up until this point, focused on some of the most relevant research applicable to the Fal & Helford - as an illustration of what can be done and as a platform for the future. We now consider some of the key issues surrounding the selection of a biomarker strategy for the Fal & Helford SAC 5 .
5.1 Identification of biomarkers ‘Biomarkers’ are frequently proposed as a means of establishing sub-lethal responses (‘early warning systems’) or confirming the cause of biological effects (Lam and Gray, 2003 ; Galloway et al., 2004; Long et al 2004). Often, however, reports stop short of recommending particular suites of biomarkers because these have to be tailored to individual projects. Consequently, their use as routine tools for marine monitoring has not yet been adopted widely although the potential value, in terms of sensitivity, cost, and ability to detect change, appears high. A practical fieldbased demonstration of the use of a selection of biomarkers in the Fal & Helford SAC will be an important developmental component of future deployments. Initially this 5
Since the review was initiated, a practical demonstration of many of the techniques has been put in place in the Fal and Helford (and Tamar Estuaries) under English Nature contract FST20-18-028 being carried out jointly by members of the Plymouth Marine Science Partnership (University of Plymouth and Marine Biological Association) and as part of the Fal Pilot Project (NERC Knowledge Transfer Award). Some of the early results are used here for demonstration purposes.
35
will focus on a few ‘core’ biomarkers for reasons of cost and reliability, in an attempt to address some fundamental questions: Key questions: Can a reasonable judgement be made on community health by utilising a simple suite of biomarkers? What are the most practical options for routine/cyclical appliance? What are the major limitations (costs, replication, variability, seasonality etc)? What contextual information is most useful? Outcomes should aim to contribute towards a strategy for routine application of biological techniques as a management tool - to help to establish cause/effect relationships, form prognoses, and to help monitor the condition of marine SACs on a long term basis. Selected biomarkers could be used in isolation, although ideally these should be used alongside chemical, DTA and ecological assessments to provide the most useful information on causal links and mechanisms. Where funding is extremely limited priorities should depend on the perceived risks. It is also worth considering the deployment of these techniques in transplanted organisms as an alternative to indigenous populations. This may be particularly relevant near to important discharges where native biota are absent. Biomarker data would thus be useful both for screening effluents and, in the long term, in helping to predict responses to changing environmental quality in the SAC (arising from planned schemes or other factors). Long et al (2004) have summarised some of the challenges and factors which should be considered in the selection of biomarkers. The most demanding challenges are in attempting to link biomarkers with population level effects i.e. from Natural England’s viewpoint, how to link apparent ‘health’ of the site (from biomarker results) to functional and structural integrity of the site. As part of the process of field validation policy makers and regulatory agencies will need to be persuaded of the significance and relevance of responses (effects biomarkers are more convincing than exposure biomarkers in this respect in that they signify ‘harm’). Challenges for researchers include the need to develop non-invasive and non-destructive techniques, promoting the use of rare species by using common surrogates, and more detailed understanding of temporal and spatial variation in responses. These challenges can only be met by more widespread trials. If the perceived risks and threats at the site are known, an appropriate range of response types, relevant to these stressors, should form the basis of the suite of biomarkers chosen. For the purposes of the current exercise we will assume that the threats outlined in the site characterisation work (TBT, metals, eutrophication) are the main water quality concerns for the Fal & Helford SAC. The choice of techniques for deployment will lean towards those which may be used to identify and separate the influence of these types of pollution (useful in terms of prioritising management effort to help improve the environment). The threats from increased turbidity/siltation and remobilisation of contaminants from (dredged) sediment would also be a pertinent issue to consider, particularly if proposed plans to increase berthing capacity to host cruise ships at Falmouth come to fruition.
36
Biochemical, physiological, reproductive and behavioural biomarkers, which signal exposure to, and in some cases, adverse effects of these types of pollution might be chosen from the core biomarkers described in section 3.1. Based upon ease of use, practicality, current level of validation, costs and relevance to known environmental issues it has now been agreed to test a subset of these in the Fal and Helford ( table 3) as part of a pilot project on the application of biological effects tools (Natural England contract FST20-18-028). Included in the table are the general characteristics, classifications and caveats described by Long et al. 2004.
37
Table 3. Biomarkers and bioassays for trial application in the Fal & Helford SAC with Mytilus edulis and Carcinus maenas.* Current perceptions of characteristics (e.g. Long et al. 2004; Schmitt and Dethloff, 2000 ) Dose response
Time response
sensitivity
specificity
Ecological relevance
costs
comments
Links to disease needs further work
Lowmed
No causal established
links
medium
Needs further confirmation, though links to damage to lipid membranes, DNA and NRR Represents whole animal response Moderate-links to survival
medium
Disease & parasites also affect Potential Interference ; subjective Links to other biomarkers (e.g. MT)
Biomarkers of Effects (general health) Phagocytosis/Immunocompetence
Characterised for some contaminants
Fairly rapid
moderate
Lysosomal Stability/Neutral Red Retention (NRR)
Characterised for some contaminants
Fairly rapid
Total Oxyradical Scavenging Capacity (TOSC)
Not widely established
Fairly rapid
marginally sensitive acute tox: moderate
Heart Rate Monitoring
Characterised for some contaminants for some contaminants
Fairly rapid
Sediment bioassays *
Can be longer term- days
more than
Responds to various classes of contaminant low Free radicals and reactive oxygen species
Mainly acute toxicity Feeding rate (cast prodn) moderate
low Probably responds to several classes
medium Lowmedium
Dissolved O2 and nutrition status Effect of sediment type possible
Biomarkers of Exposure (specific chemical class) Esterase enzymes ( e.g. OP Pesticides and other neurotoxic compounds) Fluorescence of Aromatic Hydrocarbon Metabolites (exposure to PAHs and their metabolites) Induction of Metallothionein – (bioavailable metal) Micronucleus Formation – (genotoxic agents). Imposex/Intersex chemicals).
–
(endocrine
disrupting
Quantitative
Fairly rapid
high
Moderate-high
Not fully established
low
Quantitative
Short-term (days)
sensitive
high
Not fully established
low
Quantitative
Can be longer term- days or months Reversible to a degree Irreversible (imposex)
Moderate
high
Early warning partly established
Lowmedium
high
responds to several classes Imposex highly specific to TBT
Not fully established
low
high
low
Not widely established Quantitative for TBT & intersex
Imposex sensitive
*Arenicola marina used in sediment bioassays .
38
highly
inhibition by other pollutants?
Exogenous factors need controlling Uncertain variation Should include chemistry
Measurement of changes in the overall condition of organism should be added to this list as this may corroborate findings from other biomarkers and from chemical analyses, and thereby document whole-organism and higher-level effects. For similar reasons sediment bioassays such as that described for Arenicola, and sedimentdwelling bivalves are a useful supplement. Correlation among biomarkers at different biological levels should also be attempted as this may help to establish mechanisms and cause-effect relationships beyond that of individual assays. All biological effects indices are susceptible to a variety of internal, external, and temporal factors, including species physiology, size and age, reproductive status and capture-related stress. Environmental variables include temperature and sometimes salinity. Steps should be taken to reduce the variation during sampling e.g. by surveying at selected times of year (spring and autumn may avoid peak spawning times). Where detailed analysis of risk is required, consideration should also be given to contaminant exposure route and diet. Contaminant exposure history, and variability, in relation to the timing of the study duration may also be an important feature to factor in to the design of the sampling program. To ensure a fair test of the biomarker approach, the design of the sampling programme needs to consider aspects such as spatial coverage, and level of replication – within the constraints of the funding available. It will not be possible to do everything, therefore priorities need to be set according to the key questions being asked. For the current exercise we are assuming that a first level appraisal of environmental quality on the site, and its relevance to Natural England’s condition assessment, are prime objectives.
5.2 Identification of sites for deployment of biological effects techniques A brief review of the most recent information on condition of the EMS and their potential pressures needs to be conducted in order to make a selection. Key documents will include recent ‘site characterisation’ studies and other ‘state of the environment’ reports from EA/NE. The objective should be to select sites (and gradients) known to be impacted by reasonably well-characterised water/sediment quality issues, together with sites which are considered as reflecting reference conditions (from chemical and ecological evidence). Equally important criteria on which site selection should be based are; sampling practicalities and the siting of interest features (particularly in relation to anthropogenic sources). In the Fal and Helford there are designated habitats and species throughout: Biomarker responses should be therefore be examined at a range of sites from around the Fal and Helford system along known contamination gradients. These gradients differ between TBT, metals and nutrients; the final deployment of biomarkers will reflect the need to separate the contributions from each of these stressors towards biological responses. In the current pilot biomarker study seven sites have been chosen for sampling across
39
the SAC (figure 6). The sites were chosen to represent gradients in water quality and a spectrum of conditions, from upper estuary to open coastal sites. This is considered an important element in the trial if the scheme is to be applied to other SACs. In the case of the Fal and Helford the chosen sites also represent locations where there are known and sustainable populations of available monitoring organisms (essential if a rolling program is envisaged – re-evaluating sites on a cyclical basis). The sites cover a representative range of threats and pressures (STW, docks, urban run-off, sediment loadings) to inform on condition of the site in general and in some cases to evaluate general impacts from key discharges. The sites also include perceived reference conditions within the area. As indicated below they cannot possibly hope to denote all representative features. The combination of sites and species chosen represents the best practical option to evaluate the use of biological effects techniques as surrogates of the status of the site as a whole, across a spectrum of conditions. Also, it is designed with a view to providing broader comparisons between SACs on a national basis.
Figure 6. Sampling sites chosen for biomarker trials in the Fal & Helford cSAC. There is some risk of under-representation of assessment because of restriction of sampling to inter-tidal sites and species. The limitation is principally a reflection of costings since sampling sub-littorally would incur boat costs of up to £2k per day. Offshore sampling therefore does not appear to be a practical option for routine monitoring though one-off sampling should be considered. There are of course risks to this strategy but as with any other scheme it is not possible to be all- inclusive. The inter-tidal sampling is in our opinion the most pragmatic scheme to inform site assessment, given the budgetary constraints. The use of mussels in the current proposal does give the option of transplantation to offshore sites (suspended on buoys etc) if this is considered essential in future assessments, or if indigenous populations are absent. It should be noted however that caging is best seen as a complementary method to sampling natural populations rather than a replacement. Caging provides useful information at discrete locations following contaminant exposure of finite duration. Sampling of natural populations assesses life-time exposure.
40
5.3 Identification of species for measuring biological effects. Selection of species will obviously need to be made with reference to availability at the chosen sites, their ecological relevance and also their performance, sensitivity and reliability in trials already undertaken. Additionally, the nature of the water quality pressures at the chosen sites must be a factor in the final selection. Whilst emphasis should be on native species, complementary use of transplanted test organisms might be an option for consideration (for example near discharges where indigenous organisms are of limited availability). By definition there are likely to be conservation concerns for some of the designated species which are at the heart of the monitoring effort in SACs. They may not be widely available or present in sufficient abundance to justify routine sampling for biomarker purposes: in this case the use of surrogate model species may allow for development of a more standardised methodology for broadscale comparisons within and across sites. The choice of study species should therefore include common coastal organisms that are amenable to study. These should include representatives of the major phyla. Natural, transplanted or caged mussels Mytilus edulis, clams Scrobicularia plana and gastropods (Littorina littorea, Nucella lapillus, Patella vulgata) are likely to provide the most practical options. (Other supplementary candidates are ragworms Nereis diversicolor; lugworms Arenicola marina; shore crabs Carcinus maenas; mud snails Hinia reticulate; echinoderms Asterina gibbosa and Asteria rubens; possum shrimp Neomysis integer). For the current biological effects projects taking place in the Fal & Helford SAC (Natural England Contract FST20-18-028 and NERC Fal Pilot study) the focus for most of the biomarker responses are the blue mussel Mytilus edulis and the shore crab Carcinus maenas. Sediment bioassays are taking place with the lug worm Arenicola marina, as described above; information on endocrine disrupting potential will be provided using the edible periwinkle Littorina littorea, the dog whelk Nucella lapillus, and the peppery furrow shell Scrobicularia plana. These species provide ‘surrogates’ to assess, quantitatively, the well-being of a variety of SAC interest features and habitats designated by English Nature, including large shallow bays and inlets, reefs, mudflats, sandflats and estuaries. It must be emphasised again that this does not represent comprehensive coverage but is governed by funding constraints. The ultimate aim is to evaluate biomarkers alongside ecological and other environmental (chemical) data to inform Natural England’s condition assessment. In other words, biomarkers will be an added layer of science to inform assessments, rather than a substitute for currently-used techniques which record the status of specific interest features. The number of individuals collected at each study site will typically be n=8 to allow adequate statistical analysis of results but to prevent over-sampling of local populations. To avoid boat costs, which could be prohibitive, study organisms should be selected to be easily accessible from the shoreline (including Carcinus maenas and Hinia reticulata which may be collected from baited cage deployed at the low water
41
line at each site). Specific interest features within the Fal and Helford include sublittoral Maerl and eelgrass beds. Separate biomarker studies may be needed to evaluate their likely subjection to water quality pressures, if this is deemed sufficiently important from other evidence condition assessments. Deployment of caged mussels may be the most practical technique for such biological effects measurements.
5.4. Methods of interpreting biomarker data to inform assessment of site condition The correct application and quality assurance procedures are vital in deployment and interpretation of biomarkers. This includes ensuring, as far as possible, appropriate levels of replication (i.e. what is the inherent variability of the response), elimination of spurious effects due to collection stress or storage, sex, tissue and season. The selection of an appropriate reference population is also important to compare and scale any ‘effects’ that are detected. There is general understanding of the direction (inhibition or induction) of most biomarker levels in response to stress, though it is not uncommon that defence systems can be induced at low exposure levels, but suppressed by acute pollution. Ideally dose-response relationships should be known in some detail (though in complex natural environments this is virtually impossible to predict for non-specific markers of stress). Some idea of the kinetics of the responses is also desirable, though again this is usually more of a general feeling for timescales and whether effects are transient or irreversible, rather than precise understanding of response times. Given such caveats there are a number of ways in which to present biomarker results. General scoring systems A scoring based ‘weight-of evidence’ approach may be adopted to classify sites. A suite of biomarkers should be chosen to examine effects at different levels of biological organisation, from molecular and cellular alterations, up to the physiological effects, which when assessed in combination can be used as to investigate the overall ‘health’ of the system. Although exposure biomarkers are extremely useful in a surveillance or diagnostic capacity (see below), the incorporation of effects markers adds greatly to the ecological relevance of the classification. A ‘traffic light’ system may be used to highlight assessments of impacts for each biomarker, individually and in combination, (i.e. red indicating severe alteration in biomarker response over a reference condition, orange indicates a less severe alteration, yellow moderate alteration and green little or no difference in comparison to reference responses –see example figure 7). Alternatively these can be scored, for numerical assessments, and an overall classification determined by averaging the ranks. For some biomarkers the cut-offs between ranking can be confidently assigned 42
on the basis of well defined dose-response relationships obtained in the laboratory, though for others the cut offs are more arbitrary – defined statistically rather than on the basis of known end-points. Despite the somewhat subjective nature of certain components in this scheme, it is a useful means of visualising impacts for the purposes of site assessment and policy development. Individual biomarkers as diagnostic tools for specific contaminants Some of the biomarkers listed in table 3 represent a means of assessing biological responses to specific contaminants or classes of contaminants. For example, imposex is a highly specific response to TBT pollution; changes in acetylcholinesterase activity signals exposure to neurotoxic chemicals such as organophosphate (OP) and carbamate pesticides; metallothionein induction is expressed as a result of excess metal exposure. These biomarkers of exposure may be used individually as tools to explore cause-effect relationships and impacts from specific contaminant types (identified as a potential threat by chemical analysis). One example used here to illustrate the diagnostic role of biomarkers is metallothionein, a metal-binding protein induced in Mytilus edulis in response to excess metals (a potential issue in the Fal). Metallothionein (MT) and associated changes in metal partitioning behaviour are specifically induced by metals and are sufficiently sensitive to detect elevated levels of bioavailable metal in the field, or arising from metals in discharges. This assay is able to ascertain whether or not the population is adapting successfully to the site conditions and may be used to quantify change. These signals are sub-lethal early warning indices of stress In section 3.3. we have given examples of this, using mussels transplanted to Restronguet Creek at the time of the Wheal Jane incident. The pattern of MT levels in native mussels from the Fal & Helford system is currently being measured as part of the ongoing Fal pilot study and some preliminary data, depicted using the traffic light system, are shown in Figure 7. These results illustrate that the MT biomarker is induced significantly at Weir Point at the mouth of Restronguet Creek (historically a highly metal-polluted part of the Fal system). Lowest levels were at the local reference site at Maenporth and at Flushing. Analysis of metals in mussel tissue points to sources which are likely to be responsible for these gradients of response. As indicated in figure 8, several metals including Cu, exhibit enhanced bioavailability at this site, presumably because of its proximity to Restronguet Creek.
43
Figure 7. Mytilus edulis, (A) Metallothionein in gills of mussels from the Fal & Helford SAC (spring 2006) (green: baseline levels, yellow: some induction, orange significant induction (P
