Biocomplexity: the post-genome challenge in ecotoxicology

Aquatic Toxicology 59 (2002) 1 – 15

www.elsevier.com/locate/aquatox

Viewpoint article

Biocomplexity: the post-genome challenge in ecotoxicology Michael N. Moore * United Nations Industrial De6elopment Organization (UNIDO), Vienna International Centre, PO Box 300, A-1400 Vienna, Austria Received 12 May 2000; accepted 5 July 2001

Abstract There are four crucial challenges for environmental toxicologists in the next decade: (1) understanding the mechanisms of molecular and subcellular interactions with pollutant chemicals, including genomic and proteomic aspects; (2) the development of predictive simulation models of toxic effects on complex cellular and physiological processes; (3) linking molecular, cellular and patho-pysiological ‘endpoints’ with higher level ecological consequences; and (4) precautionary anticipation of possible harmful impacts of novel developments in industrial processes, including biotechnology and nanotechnology. One of the major difficulties in ecotoxicology is to link harmful effects of chemical pollutants in individual animals and plants with the ecological consequences. Consequently, this obstacle has resulted in a ‘knowledge-gap’ for those seeking to develop policies for sustainable use of resources and environmental protection. The overall problem is: how to develop effective procedures for environmental/ecological impact and risk assessment? However, the use of diagnostic ‘clinical-type’ tests or ‘biomarkers’ has started to provide information on the health-status of populations based on relatively small samples of individuals. Also, biomarkers can now be used to begin to link processes of molecular and cellular damage through to the higher levels (i.e. prognostic capability), where they can result in reduced performance and reproductive success. Research effort to meet this challenge must be inter-disciplinary in character, since the key questions mainly involve complex interfacial problems. These include effects of physico-chemical speciation on uptake and toxicity, the toxicity of complex mixtures; and linking the impact of pollutants through the various hierarchical levels of biological organisation to ecosystem and human health. Finally, the development and use of process-based computational simulation models (i.e. ‘virtual’ cells, organs and animals), illustrated using an endosomal/lysosomal uptake and cell injury model, will facilitate the development of a predictive capacity for estimating risk associated with the possibility of future environmental events. © 2002 Published by Elsevier Science B.V. Keywords: Autophagy; Biomarkers; Biotechnology; Cell injury; Complexity; Ecotoxicology; Impact and risk assessment; Lysosomes; Nanotechnology; Process-based computational simulation models; Xenobiotics

1. Introduction * Present address: Plymouth Marine Laboratory, Prospect Place, The Hoe, Plymouth PL1 3DH, UK. Tel.: + 44-1752633-120; fax: + 44-1752-633-101.

Most of the earth’s living resources are found in specific geographical locations such as the global

0166-445X/02/$ - see front matter © 2002 Published by Elsevier Science B.V. PII: S 0 1 6 6 - 4 4 5 X ( 0 1 ) 0 0 2 2 5 - 9

2

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15

coastal environment and the catchment basins of large river systems (Costanza et al., 1997). Furthermore, a large proportion of the world’s population lives in close proximity to these regions and is frequently dependent upon it for either part or much of its food supply and industrial raw materials. The consequence of this situation is that much of the waste, both industrial and domestic, and various other types of habitat disturbance (e.g. aquaculture, deforestation, erosion of agricultural land, soil contamination, mangrove forest and coral reef destruction, land reclamation, dredging, etc.) generated by the human population occurs in those areas that are of greatest biological and economic significance. Hence, the potential for deleterious impact in both environmental and economic terms is immense, but has in the past generally taken a relatively low priority in the context of the global socio-economic system. However, there is now increasing awareness of the global importance of specific geographical domains, such as the coastal land– sea interface, as major resources and concern for maintaining the diversity of life on our planet. This was a major focus for Agenda 21 of the UNCED, Earth Summit Conference in Rio de Janiero (Quarrie, 1992). Pollutant impact on ecosystem and human health is an urgent and international issue since there is an ever-increasing number of examples of environmental disturbance, likely to affect the biota and humans, by both natural and anthropogenic stress. Important stressors include toxic chemical contaminants, increased UV-B, nutrient enhancement or deprivation, hypoxia due to eutrophication, habitat disturbance and pathogeninduced disease. In fact, environmental disturbance will frequently comprise various combinations of such stresses. Furthermore, it is increasingly recognised that assessment of the impact of environmental disturbance on organisms requires understanding of stress effects throughout the hierarchy of biological organisation, from the molecular and cellular to the organism and population levels, as well as the community and ecosystem level. In the past, damage to the environment has largely been identified retrospectively and in response to acute events such as major disasters (e.g. industrial accidents

like Seveso and Bhopal; and chemical spills like the ‘Amoco Cadiz’ and ‘Exxon Valdes’). Generally these have been measured in terms of human health impacts and visible changes resulting from the loss of particular populations or communities. However, long term and chronic exposure to environmental stress, including chemical pollutants or other anthropogenic factors, will seldom result in rapid and catastrophic change. Rather, the impact will be gradual, subtle and frequently difficult to disentangle from the process and effects of natural environmental change. This latter problem has been a major stumbling block in assessing environmental impact since such investigations began, mainly in the 1960s. While it is clearly recognised that changes at the population/community/ecosystem/human health levels of biological organisation are the ultimate concern, they are too complex and far removed from the causative events to be of much use in developing tools for the early detection and prediction of the consequences of environmental stress (Fig. 1). Stress is defined here as any environmental alteration that extends homeostatic or protective processes into a compensatory state beyond the normal limits of an organism (Bayne

Fig. 1. Conceptual plot of fitness versus health status as related to duration of pollutant exposure (adapted from Depledge et al., 1993). This shows the various pathophysiological states that have been observed in organismal responses/reactions to pollutant stressors.

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15

et al., 1985). In turn, if the stress causes the compensatory limits to be exceeded, then the organism is described as being in a diseased state (Depledge et al., 1993). Consequently, stress renders the individual, and ultimately the population, at a disadvantage through reduced growth, impaired reproduction, and increased susceptibility to degenerative and/ or infectious disease (Bayne et al., 1985). The population will, therefore, be vulnerable to further environmental change and possible extinction. Assessment of the impact of stress on the health of animals and plants is a major challenge given the complexity of the many potentially interactive factors involved, but requires the effective determination of harmful effects on the health status of individuals, as well as the identification of the cause. Rapid resolution of this link to health is essential if environmental management is to have a sound scientific basis for the regulation of the release of toxic substances, nutrients and habitat disturbance. The basis of such regulation, where it exists at present, is often at best sketchy with a heavy reliance on empirical observations and laboratory based toxicity tests using organisms that have limited relevance to the real environmental context. A probable solution to this problem lies in the effective detection of ‘distress signals’ at the molecular and cellular levels of organisation and linking these latter to the higher level consequences (Fig. 1; Moore, 1990; Moore and Simpson, 1992). It is only at these lower levels that we will have the reasonable expectation of developing a reasonable basis of mechanistic understanding of how different environmental conditions can modulate organismal function, which in turn will ultimately help in linking causality with predictability of response. This is in part due to our ability to make certain generalisations about biological organisation and function at the molecular and cellular level which rapidly disappear as we ascend the hierarchical ladder. Hence, distress signals at the molecular, cellular and physiological levels of organisation should be capable of providing ‘early warning prognostic biomarkers (molecular, cellular, physiological and be-

3

havioural)’ of reduced performance, impending pathology and damage to health (Depledge et al., 1993; Hinton and Lauren, 1990; McCarthy and Shugart, 1990). In fact, there is a direct analogy here with the use of clinical tests (biomarkers) in human and veterinary medicine (Moore and Simpson, 1992). Whereas the ultimate health consequences occur at the higher cellular, organism and ecological levels, the underlying root causes and mechanisms are at the molecular level (Maddox, 1998; Slater, 1979). The derivation of potential prognostic ‘distress signals’ will only arise out of an understanding of the mechanistic basis of the cellular and physiological processes that contribute to uptake, biotransformation, molecular damage and cell injury, impairment of protective systems and, ultimately, to degenerative change and the consequences for reproduction and survival (Moore, 1990; Moore et al., 1994; Fig. 1). For this way forward to be fully effective it requires an integrated multi-tiered approach combining both reductionist and synthesist components. The tools to implement this are now becoming increasingly available. Briefly, these include mechanisms of pollutant uptake, biotransformation and radical generation, molecular damage and consequent cell injury, as well as antioxidant protection and repair. These in turn need to be linked with cellular and physiological processes of vesicular transport of proteins, protein turnover, and interactions of the nervous and endocrine systems with effective immune defence function. At the higher organisational levels, differential sensitivity needs to be assessed according to individual genotype, life-history stage, and natural seasonal changes in physiological and/or reproductive status. Finally, this information then needs to be used to develop process simulation models of the type increasingly used in quantitative cell biology and cellular bioengineering (Lauffenburger and Linderman, 1993). Mathematical models can provide insights into the links between molecular properties and cell and organ behaviour and the predictive power of such models can be harnessed to develop tools for risk assessment of toxic chemicals.

4

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15

In summary, therefore, there are four crucial challenges for environmental toxicologists in the next decade: (1) understanding the mechanisms (including the biophysics) of molecular and subcellular interactions with pollutant chemicals; (2) the development of predictive simulation models of complex interactive cellular and physiological processes; (3) linking relevant molecular, cellular and patho-pysiological ‘endpoints’ with higher level ecological consequences; and (4) precautionary anticipation of the possible harmful impacts of novel developments in industrial chemical processes, biotechnology and molecular nanotechnology. All of these challenges are themselves complex and inherently interrelated; for this reason, the problems posed and their possible solutions will be dealt with as a whole, after initially outlining the essential problem in each case.

2. The challenges First, much of the so-called mechanistic research to date has been descriptive, whether at the level of the whole animal, the cell or at the molecular level (Bannasch et al., 1989; Bayne et al., 1985; Hinton and Lauren, 1990; Kurelec and Pivcevic, 1991; Livingstone, 1993; Lowe et al., 1992; Maddox, 1998; Minier and Moore, 1996; Moore et al., 1994; Slater, 1979; Stegeman and Lech, 1991; Winston et al., 1996). Much of our research is described as being mechanistic but is it in reality? It is, in fact, still largely describing the parts and phenomena (‘natural history’ at a cellular and molecular level): that reactions to pollutants do occur; but not how they occur. Environmental toxicologists are not really addressing the core problem of the biophysics and physical chemistry of how complex molecules interact and are affected by contaminant chemicals; and that many of these interactions will be multiple in nature, consequently producing the observed complex responses to stress. Similar arguments also apply throughout the biological sciences (Maddox, 1998). With the imminent completion of the humangenome project, which also includes the characterisation of the genomes of some lower organisms,

the major challenge now facing biologists and toxicologists is to figure out what it all means. In this respect, it is not the genes that carry out the functions in cells and organisms, but the proteins (i.e. cellular machines) expressed by a subset of the genome: in fact, much of the genome is normally switched off in most cells for much of their lifetime. Much of the advance in toxicology during the last two decades has focussed on the identification and characterisation of specific proteins expressed in response to exposure to toxic metals and xenobiotics (Soni and Mehendale, 1998; Viarengo, 1989). Proteomics is the link between genes, proteins and cellular function, including diseased states (e.g. pollutant-induced environmental pathology) (Link et al., 1997). In a complex organism such as a mussel, fish or man, approximately 30,000–60,000 genes are selectively expressed in individual cells to yield the specific sets of proteins required to make functioning liver cells, skin cells or brain cells, etc. However, while the number of genes in an individual organism is static and fixed, the set of proteins that can be produced by an organism at one time or another throughout its life is not fixed; and varies with age and development, cell and tissue type and in response to environmental stimuli. Proteomics aims to measure the protein output of many thousands of genes in order to describe how they are regulated; and how this regulation is altered by environmental adaptations, or disturbed by pollutant induced perturbations and disease processes. Consequently, the task of analysing the complexity of the proteome of an organism is a far greater challenge than anything in genomics (Link et al., 1997). An example of protein interactions in the induction of cellular lysosomal autophagy (‘self-eating’), a frequent response to toxic cell injury, is described below in the section on process simulation modelling (Figs. 2– 4; Table 1). Other scientific approaches are, of course, also required as part of the data collecting process. These include: molecular toxicology; genome analysis; cell biology (including multi-parameter imaging methods (De Biasio et al., 1987) to study signalling pathways and dynamic cellular pro-

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15

5

Fig. 2. Diagrammatical representation of the processes of cellular uptake of xenobiotics; their lysosomal accumulation and harmful effects, including induced lysosomal proteolysis, augmented autophagy and cell injury (Benchimol, 1999; Bonifaciano and Weissman, 1998; Calle et al., 1999; Franke, 1990; Halliwell, 1997; Hawkins and Day, 1999; Hein et al., 1990; Hutchins et al., 1999; Luzikov, 1999; Moore, 1990; Moore and Willows, 1998; Moore et al., 1994, 1996, 1997; Mortimore and Poso, 1987; Rashid et al., 1991; Seglen, 1997; Thevenod and Friedman, 1999; Viarengo et al., 1994; Winston et al., 1996).

cesses); integrated whole animal physiology; improved tools for ecological analysis; and appropriate simulation model development to help in the integration and problem solving process (Maddox, 1998). This latter is particularly important if effective predictive tools are to be developed for use in environmental impact and risk assessments. What is now required is a profound change in current thinking in order to take this critical step towards the biophysics and physical chemistry of molecular interactions in toxicology, and to develop positive interactions with chemists, physicists, process simulation modellers and complexity scientists. Second, there is an urgent need to develop process-based computational simulation models in order to address the complexity of ecotoxicology (Chicurel, 1999; Moore and Willows, 1998). In

chemical engineering, physics and epidemiology, for example, it is well understood that complex systems can be accurately understood only by constructing quantitative mathematical models (Chicurel, 1999; Koo, 1999; Maddox, 1998). Despite the remarkable efforts by molecular cell biologists to identify the chemical components of cells, little has been done as yet to construct models that help to explain what the molecular data mean. Here, one of the most urgent needs is to make comprehensible the complexities of the responses of cells and tissues to external influences, including chemical pollutants. Toxicologists are still largely working in the dark in this respect; and will remain so, until realistic models have been built of the process describing how the specificity of the cell’s response matches that of the external signal or potentially harmful perturbation it receives.

6

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15

Fig. 3. Diagramatic representation of mathematical process-based model for hypothesised cellular and lysosomal uptake, fate and effects of micropollutant xenobiotics (L, micropollutant ligand; C, ligand particulate complex; LF, lipofuscin; P, protein; MDR, multidrug-resistance transporter system; ROS, reactive oxygen species; see Table 1 for all other terms). Taken from Moore and Willows (1998); and Moore, unpublished.

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15

A major difficulty here is that techniques that have proved highly suitable for the identification of parts of cells in the past 30 years are less effective at collecting quantitative information about the working of the molecules, macromolecules and supra-molecular complexes involved in living processes; and their perturbation by xenobiotics and metals (Maddox, 1998). Moreover, there is often an innate resistance to the development and use of mathematical models, since there is still much to discover about the processes and interactions in cells (Maddox, 1998). A further important factor is that many biologists have a distinct aversion to mathematical descriptions. It is, therefore, not surprising that biologists will not readily abandon their perceived, highly productive ‘molecular natural history’ of the cell. Ultimately, however, there will be no choice!

7

There is a prevailing view that modelling cellular processes is a non-productive avenue that takes descriptive and mechanistic data about the cell or organism and transforms it into an unnecessarily complicated and unfamiliar language (Maddox, 1998). Unfortunately, this view misses the point about how models can help to make a coherent whole from disparate data sets. Models are useful conceptual indicators for the design of experiments that rigorously test current paradigms (Biganzoli et al., 1998; Koo, 1999; Lauffenburger and Linderman, 1993). Environmental toxicologists’ collective neglect of modelling has become a serious impediment to progress. However, the recent work of Noble in developing a ‘virtual’ computational heart model, and that of Du¨ chting with a mathematical tumour model, has opened a new avenue for the future in many areas of biomedicine and toxicology (Du¨ chting et al., 1996; Noble et al., 1999).

Fig. 4. Empirical model for the molecular mechanism of cellular autophagy based on the interactions of protein autophagy factors (Apg5p, Apg7p, Apg8p, Apg10, Apg12p, Apg16p) that are expressed by some of the 15 autophagy-defective mutant genes (APGs) in yeast cells (Hutchins et al., 1999; Kim et al., 1999a,b; Kirisako et al., 1999; Liang et al., 1999; Lipton, 1999; Luzikov, 1999; Mizushima et al., 1999; Ohsumi, 1999a,b; Shintani et al., 1999; Tanida et al., 1999; Yuan et al., 1999).

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15

8

Table 1 Glossary of terms used in the model described in Figure 3 Symbol

Definition

Typical unit

Bs BTi Ci Cs CF fB ff flip fL f met L fP Fb Iaut Icyt Il kdeg ke kend kel kex kf kfF kF kl kfp klc kmet kmdr kmdr kr krF krec L LB

Concentration of particulate with free binding sites at cell surface Concentration of total particulate (free+bound) in vacuolar system Concentration of intracellular ligand–particulate complexes Concentration of intracellular ligand–particulate complexes at the cell surface Concentration of ligand–lipofuscin complexes in lysosomes Fraction of endocytosed binding sites degraded Fraction of free ligand in total ligand in vacuolar system Fraction of lysosomal lipid degraded Fraction of endocytosed ligand degraded Fraction of cytosolic ligand metabolised Fraction of lysosomal protein degraded Concentration of lipofuscin (free) Concentration of unsaturated lipid autophagocytosed Concentration of unsaturated cyosolic lipid Concentration of unsaturated lysosomal lipid Rate constant for lysosomal degradation Rate constant for diffusion of ligand out of cell Rate constant for endocytois Rate constant for endosomal–lysosomal transport Rate constant for exocytosis Rate constant for association of ligand and binding site Rate constant for association of ligand and lipofuscin binding sites Rate constant for formation of lipofuscin Rate constant for diffusion of ligand into lysosome/endosome Rate constant for fluid-phase endocytosis Rate constant for diffusion of ligand out of endosomal–lysosomal system Rate constant for metabolic transformation Rate constant for MDR transport of ligand Rate constant for MDR transport from the cytosol into the endosomal–lysosomal system Rate constant for dissociation of ligand–particulate binding site complex Rate constant for dissociation of ligand–lipofuscin binding site complex Rate constant for transport of materials from endosomes back to cell surface via vesicles Ligand concentration in medium Concentration of particulate bound ligand at cell surface: dependant on log P (log Kow) and particulate concentration Free ligand concentration in cytosol Concentration of free ligand in the endosomal–lysosomal system Concentration of total ligand (free+bound) internalised in the endosomal–lysosomal system Avogaddro’s number (6.02×1023 c /mole) Concentration of protein autophagocytosed Concentration of intracellular protein Concentration of protein endocytosed Concentration of protein exocytosed Concentration of protein in lysosome Rate of generation of lipofuscin Rate of generation of particulate

mg mg mg mg mg

Lc Li LTi NAv Paut Pcyt Pend Pex Pl 6F 6s

Taken from Moore and Willows, 1998; and Moore, unpublished.

l−1 l−1 l−1 l−1 l−1

mg l−1 mg l−1 mg l−1 mg l−1 min−1 M−1 min−1 min−1 min−1 min−1 M−1 min−1 M−1 min−1 min−1 M−1 min−1 mm3 min−1 M−1 min−1 M−1 min−1 M−1 min−1 M−1 min−1 min−1 min−1 min−1 M or mg l−1 mg l−1 M or mg l−1 M or mg l−1 M or mg l−1 mg mg mg mg mg mg mg

l−1 l−1 l−1 l−1 l−1 l−1 min−1 l−1 min−1

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15

In ecotoxicology, there has only been very limited use of the type of model described above. Modelling procedures for intracellular behaviour of pollutants will be particularly important in helping to define the problems and in developing hypotheses in this highly complex area. One such preliminary mathematical model has been developed that defines the component processes in endocytosis, lysosomal compartmentation, toxicity and pathology (Figs. 2 and 3; Moore and Willows, 1998). The model focuses on ligand-binding sites associated with endocytosed particulates and the role of the endosomal – lysosomal system in pollutant uptake, toxicity and cell injury (Cheung et al., 1998; Hauton et al., 1998; Lowe et al., 1995a,b; Moore et al., 1994, 1996, 1997; Ringwood et al., 1998; Svendseb and Weeks, 1995; Wedderburn et al., 1998; and Figs. 2 and 3). This model has provided a conceptual framework for pollutant uptake and biotransformation, lysosomal accumulation, protein degradation, cellular autophagy and cell injury, as well as excretion of pollutants and bioavailability. It also highlights key hypotheses for experimental testing and validation of the model (Moore and Willows, 1998). The uptake and accumulation of organic micropollutants and metals by aquatic organisms is governed by their physical chemical speciation (Readman et al., 1984). Since lipophilic pollutants are largely bound to particulate and colloidal organic carbon, it is probable that contaminant entry into cells is directly related to the extracellular and intracellular behaviour of particulates/colloids with adsorbed chemicals (Hermans et al., 1992; Murdoch et al., 1994; Smedes, 1994; and Figs. 2 and 3). The route for uptake of colloids and particulates into cells is by the process of endocytosis and intracellular vesicular transfer to the lysosomal compartment (i.e. cell feeding and intracellular digestion). Lysosomes provide a highly conserved intracellular degradative compartment for proteins and other macromolecules in the cells of almost all eukaryotic organisms; and their accumulation of toxic metals and organic xenobiotics is a well-documented cellular phenomenon (Halliwell, 1997; Rashid et al., 1991). It has also been demonstrated that lysosomal uptake is a significant factor in cell

9

injury. However, in a seemingly converse context, it has been postulated that lysosomal sequestration may have a protective role in cellular detoxication of metals (Calle et al., 1999; Moore, 1990; Nott and Nicolaidou, 1990). However, for organic xenobiotics the detoxication hypothesis has not yet been rigorously tested (Halliwell, 1997; Moore, 1990; Moore and Willows, 1998; Rashid et al., 1991). The model incorporates the main functions of lysosomes as the degradative compartment for vesicular and protein traffic, endocytotic uptake of contaminants, bound to particulates, and their subsequent intracellular behaviour (Moore and Willows, 1998; and Figs. 2 and 3). This includes intravesicular dissociation, diffusion and hypothesised reassociation with lipofuscin; as well as augmented autophagic protein and organelle degradation and generation of reactive oxygen species (Winston et al., 1996; and Figs. 2– 4). Lysosomal toxic thresholds and their possible underlying mechanisms still need to be investigated, in order to determine their significance in the processes of cell injury and pathology (Halliwell, 1997; Moore, 1990). However, there is now a wealth of data that lysosomal dysfunction/damage is a good predictor for higher level pathologies and hence provides a prognostic biomarker (Moore, 1990, 1991). In studies of invertebrates, fish and mammals the process of augmented cellular autophagy (‘self-eating’) and subsequent lysosomal degradation in response to toxic cell injury is a recurrent theme (Franke, 1990; Halliwell, 1997; Hein et al., 1990; Lowe et al., 1992, 1995a,b; Luzikov, 1999; Moore, 1990, 1991; Moore et al., 1994, 1996; Thevenod and Friedman, 1999; and Figs. 2 and 3). Identifying the crucial molecular, structural and functional changes that trigger cellular autophagy is central to understanding elimination of mistakes in organelle assembly and cell injury and, hence, predicting the ensuing pathological consequences (Benchimol, 1999; Bonifaciano and Weissman, 1998; Hawkins and Day, 1999; Lipton, 1999; Luzikov, 1999; Mortimore and Poso, 1987; Seglen, 1997; Shin, 1998; and Fig. 4). Recent advances in this area have largely focused on the yeast cell as an experimental model for autophagy and the identification of a number of proteins, expressed by the APG genes

10

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15

(autophagy-defective mutant genes), that function as autophagy factors and some of their interactions is rapidly providing a mechanistic molecular basis for regulation of cellular autophagy (Hutchins et al., 1999; Kim et al., 1999a; Kirisako et al., 1999; Liang et al., 1999; Mizushima et al., 1999; Ohsumi, 1999a,b; Shintani et al., 1999; Tanida et al., 1999; Yuan et al., 1999). The known interactions of the autophagy factors are shown diagramatically in Fig. 4; and there is increasing evidence for homology between these proteins in yeast and their equivalents in higher organisms, including humans; indicating that they probably represent a highly conserved system (Kim et al., 1999a,b; Liang et al., 1999; Ohsumi, 1999a; Yuan et al., 1999). Although such models are useful, it is necessary to move on from relatively simple chemical engineering-type mathematical models based on differential equations to computational simulation models based on multiple molecular interactions (Schaff et al., 1997). These cellular models must be capable of experimental validation, which should include genetic manipulation of key processes. Visualisation of cellular processes in silico will facilitate the identification of complex subcellular strategies for adaptation to altered environmental conditions. As such, these models will have to include all of the major cellular physiological processes, including those that are essential for cellular defence. These latter must include the various enzymic biotransformation systems (e.g. CYP-proteins and cytosolic esterases), anti-oxidant defences, multidrug resistance, autophagy of damaged cellular components, DNA repair and proteolysis. Mechanisms of toxicant uptake must also be included, such as transmembrane diffusion and the various endocytotic routes into the cell. This will also require the incorporation of QSAR components into the model, which specify the route(s) of uptake, aspects of intracellular behaviour and eventual fate, such as, either metabolism and elimination, or else accumulation within particular cellular compartments (e.g. in lipid droplets and lysosomes). The third challenge is linking relevant molecular, cellular and patho-pysiological ‘endpoints’ with higher level ecological consequences. A key

dual aim of environmental science is to validate (1) rapid and robust and (2) relatively low cost procedures for assessing risk to the health of the biosphere and to use this capability to predict the likely consequences of exposure to potentially harmful toxic pollutants (Fig. 5). Until recently, risk assessment procedures have been oriented towards protecting human health (Fig. 5). Now, it is widely acknowledged that such procedures must also ensure that complex biotic communities in natural ecosystems are protected if the quality of the environment in which we live is to be maintained (Ferson and Long, 1995). Environmental risk assessments are currently based on a suite of information derived from studies on the physicochemical characteristics of compounds, as in the QSAR approach (Rashid et al., 1991), and from laboratory-based toxicity tests. Although these procedures constitute a low cost, pragmatic means of ranking the toxicity of potentially hazardous

Fig. 5. A conceptual framework showing the interconnectedness of environmental pollutant-related processes and their harmful effects as components of a complex adaptive system.

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15

chemicals, they do not directly evaluate the sublethal toxicity, or other adverse effects (e.g. disturbance of ecological relationships) on organisms exposed to complex mixtures of pollutants in the highly fluctuating conditions that prevail in the environment (Howard, 1997; McCarthy and Shugart, 1990). As mentioned above, contaminants are seldom present as a single chemical and usually comprise a complex mix (Fig. 5). Uptake of such complex mixtures is poorly understood and questions of whether components of the mixture influence the uptake, biotransformation and toxicity of other components has not yet been seriously addressed in ecotoxicology (Howard, 1997; Kanzawa et al., 1997; Kortenkamp and Altenburger, 1998; Warne and Hawker, 1995). The process of uptake is often viewed as taking place from solution, with the contaminant crossing cellular membranes by diffusion (Figs. 2 and 3; Moore and Willows, 1998). However, as stated above most contaminant chemicals are bound to particulates and so are seldom in true solution. Also, this is probably of considerable importance in explaining the known compartmentation of many micropollutants and metals within the cells and tissues of aquatic and terrestrial organisms, so it is essential that a mechanistic understanding of the intracellular transport processes, intracellular chemistry and associated biotransformations is developed in the future (Moore, 1990, 1991; Moore and Willows, 1998; Nott and Nicolaidou, 1990). This type of knowledge is essential if we are going to be able to attempt to predict the kinds of organisms at risk and, also, whether particular life stages are more vulnerable than others. There is, therefore, a priority requirement to implement the use of rapid and robust, simple, easy to learn, cost-effective test systems that can identify early diagnostic changes in biota that can in turn be linked to ecologically relevant endpoints (Fig. 5). These latter must be capable of facilitating a predictive ranking of the condition of particular ecosystems, thus highlighting environmental situations where a more detailed analysis is justified (Depledge et al., 1993; Moore, 1991; Moore and Simpson, 1992).

11

The fourth and the final challenge for ecotoxicologists/environmental toxicologists is to try to anticipate the potential impacts of novel products and unwanted by-products of industry (Moore et al., 1997). This includes future developments in the chemical and pharmaceutical industries as well as industries using biotechnology and anticipated arrival of a new ‘molecular nanotechnology’ industry (Drexler, 1992; Gross, 1999; Joachim et al., 2000; Perrin, 1997). Harmful products are likely to include: new targeted drugs and pesticides, particularly those for use in concert with genetically modified crops; natural pesticides resulting from gene transfer into crops; and novel pathogens used for biological control (Darmency, 1994; Dunwell and Paul, 1990; Perrin, 1997). The exponential increase in published papers in the area of nanotechnology since 1990 is a strong indicator that shortly there will be a new industry based on engineering molecules for multiple applications (Drexler, 1994; Gross, 1999). Without strong international controls on such production there could be serious environmental risks, since it is likely that the first practical developments towards nanotechnological application will rely on modified biological molecules and assemblages (Gross, 1999; Kitov et al., 2000). The purpose here is primarily to raise awareness of developments and possible environmental risks in the field of nanotechnology, since it is still in the research phase. The overall problem is, in essence: how to develop effective procedures for environmental/ ecological impact and risk assessment? The use of biomarkers and biological effects indices has proven useful in establishing evidence of exposure to pollutant chemicals and damage to the health of sentinel organisms (Figs. 1, 2 and 5; Depledge et al., 1993). This is obviously of great value in helping to establish causal relationships. However, for impact and risk assessment tools to be effective they must be capable of providing data that relates to ecologically significant processes (Depledge et al., 1993; Moore et al., 1994). This requires a better understanding of particular biomarkers, as they relate to health status (Figs. 1 and 5), in order to improve their interpretative value in monitoring (Moore and Simpson, 1992;

12

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15

Moore et al., 1994). In order to achieve this the science must also address the question of linkages between effects at different levels of biological organisation (Bayne and Moore, 1998; Grundy et al., 1996a,b; Kauffman, 1993; Livingstone et al., 2000). Establishing these linkages is essential, not only for understanding the current status of the environment, but also to provide a rational basis for prognosis for future improvement or deterioration in environmental quality (Fig. 5).

3. Conclusions This broad approach to the complex problem of assessing the health of ecosystems will facilitate the validation, and further the essential new development of robust and rapid tools for assessment (Depledge et al., 1993; Ferson and Long, 1995; Moore and Simpson, 1992; Moore et al., 1994). Future efforts must focus on an integrated approach to the validation of biomarkers that are prognostic for population and community endpoints (Depledge et al., 1993; Moore et al., 1994). As with bioavailability and uptake, exposure to pollutant mixtures must also be considered with the possibility of complex synergistic interactions resulting in emergent and novel toxicities and pathologies (Moore and Willows, 1998). Other environmental factors such as intense visible and UV-radiation (Fig. 5) and hypoxia also need to be considered, since these are likely to be important in terms of potentially harmful interactions with contaminants (Livingstone et al., 2000; McFadzen et al., 2000; Mason, 1990). This includes xenobiotics acting as photosensitisers and the facilitation of cascades of harmful radical production on reoxygenation following a hypoxic interlude, such as that induced by eutrophication (Fig. 5). A major reason for the development of complexity science, and its use in ecotoxicology, is to gain a realistic insight into the limits of reductionism as a very successful universal problem-solving approach (Casti, 1994; Kauffman, 1993; and Fig. 5). Complex biological and ecological processes generate counter-intuitive, seemingly acausal behaviour that is full of novelty (Howard, 1997; Kanzawa et al., 1997; Kauffman, 1993). Trying to

understand the behaviour of a complex adaptive (or dynamic) system, such as an organism, population, ecosystem or biogeochemical cycle, by a reductionist approach often irretrievably destroys the inherent ‘holistic’ nature of the problem (Kauffman, 1993). The recognition that a system is complex tends to be specifically subjective, and not an objective property of an isolated system. However, it can become objective, once the investigative protocol takes into account the larger system with which the target system interacts (Casti, 1994; Kauffman, 1993). In fact, Casti (1994) has stated that ‘complexity science is really a subset of the more general and much larger scale objective of creating a theory of models’. Consequently, there must be a wider recognition that environmental toxicology is dealing with complex adaptive systems (Kauffman, 1993; and Fig. 5). Hence, by implication, there needs to be a rapid acquisition of the new methods of ‘complexity science’ and implementation of these in scientific programmes. A rational way forward can be charted if an integrated multidisciplinary approach to the impact of anthropogenic chemical inputs is adopted as follows: (1) Development of conceptual frameworks and process-based simulation models based on an improved mechanistic understanding of contaminant uptake, biotransformation, toxicity and impact within the biological organisational hierarchy (Du¨ chting et al., 1996; Koo, 1999; Lauffenburger

Fig. 6. Conceptual framework for developing new predictive ecotoxicological tools to address future needs for environmental impact and risk assessment.

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15

and Linderman, 1993; Moore and Willows, 1998; Noble et al., 1999; Figs. 1– 6). (2) Meet the interdisciplinary challenge of mathematically modelling processes in environmental pollution and impact as a complex adaptive system (Kauffman, 1993; Figs. 5 and 6), encompassing contaminant geochemistry, mode of uptake and intracellular behaviour, biochemical toxicology (including proteomics), cellular injury and pathology (e.g. using ‘virtual’ organs and animals; Fig. 3), ecological consequences and human risk. (3) Take a broad view of the current and predicted future problems in environmental toxicology that incorporates both moderate reductionist and synthesist approaches Figs. 3 and 4).

References Bannasch, P., Enzmann, H., Klimek, F., Weber, E., Zerban, H., 1989. Significance of sequential cellular changes inside and outside foci of altered hepatocytes during hepatocarcinogenesis. Toxicol. Pathol. 4, 617 –628. Bayne, B.L., Brown, D.W., Burns, K., Dixon, D.R., Ivanovici, A., Livingstone, D.R., Lowe, D.M., Moore, M.N., Stebbing, A.R.D., Widdows, J., 1985. The Effects of Stress and Pollution on Marine Animals, Praeger, New York, 384p. Bayne, C.J., Moore, M.N., 1998. Non-lymphoid immunologic defenses in aquatic invertebrates and their value as indicators of aquatic pollution. In: Zelikoff, J.T. (Ed.), EcoToxicology: Responses, Biomarkers and Risk Assessment. Published for OECD by SOS Publications, Fair Haven, NJ, pp. 243– 261. Benchimol, M., 1999. Hydrogenosome autophagy: an ultrastructural and cytochemical study. Biol. Cell 91, 165 – 174. Biganzoli, E., Boracchi, P., Daidone, M.G., Marubini, E., 1998. Flexible modelling in survival analysis. Structural biological complexity from the information provided by tumor markers. Int. J. Biol. Mark. 13, 107 –123. Bonifaciano, J.S., Weissman, A.M., 1998. Ubiquitin and the control of protein fate in the secretory and endocytic pathways. Annu. Rev. Cell Dev. Biol. 14, 19 – 57. Calle, E., Berciano, M.T., Fernandez, R., Lafarga, M., 1999. Activation of the autophagy, c-FOS and ubiqutin expression, and nucleolar alterations in Schwann cells precede demyelination in tellurium-induced neuropathy. Acta Neuropathol. 97, 143 –155. Casti, J.L., 1994. Complexification: Explaining a Paradoxical World through the Science of Surprise, Abacus, London, 320p. Cheung, V.V., Wedderburn, R.J., Depledge, M.H., 1998. Molluscan lysosomal responses as a diagnostic tool for detection

13

of a pollution gradient in Tolo Harbour, Hong Kong. Mar. Environ. Res. 46, 237 – 241. Chicurel, M., 1999. The bigger picture. New Scientist 164, 38 – 42. Costanza, R., d’Arge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem, S., O’Neill, R.V., Paruelo, J., Raskin, R.G., Sutton, P., van den Belt, M., 1997. The value of the world’s ecosystem services and natural capital. Nature 387, 253 – 260. Darmency, H., 1994. The impact of hybrids between genetically modified plant crop plants and their related species: introgression and weediness. Mol. Ecol. 3, 37 – 40. De Biasio, R., Bright, G.R., Ernst, L.A., Waggoner, A.S., Taylor, D.L., 1987. Five-parameter fluorescent imaging: wound healing of living Swiss 3T3 cells. J. Cell Biol. 105, 1613 – 1622. Depledge, M.H., Amaral-Mendes, J.J., Daniel, B., Halbrook, R.S., Kloepper-Sams, P., Moore, M.N., Peakall, D.P., 1993. The conceptual basis of the biomarker approach. In: Peakall, D.G., Shugart, L.R. (Eds.), Biomarkers —Research and Application in the Assessment of Environmental Health. Springer, Berlin, pp. 15 – 29. Drexler, K.E., 1992. Nanosystems: Molecular Machinery, Manufacturing and Computation, Wiley – Interscience, New York, 556p. Drexler, K.E., 1994. Molecular nanomachines: physical principles and implementation strategies. Ann. Rev. Biophys. Biomol. Struct. 23, 377 – 405. Du¨ chting, W., Ulmer, W., Ginsberg, T., 1996. Cancer: a challenge for control theory and computer modelling. Eur. J. Cancer Part A 32 (8), 1283 – 1292. Dunwell, J.M., Paul, E.M., 1990. Impact of genetically modified crops in agriculture. Outlook Agric. 19, 103 – 110. Ferson, S., Long, T.F., 1995. Conservative uncertainty propagation in environmental risk assessments. In: Hughes, J.S., et al. (Eds.), Environmental Toxicology and Risk Assessment. ASTM, Philadelphia, PA, pp. 97 – 110. Franke, H., 1990. Substructural alterations of liver parenchymal cells induced by xenobiotics. Exp. Pathol. 39, 139 – 155. Gross, M., 1999. Travels to the Nanoworld: Miniature Machinery in Nature and Technology, Plenum Trade, New York, 254p. Grundy, M.M., Moore, M.N., Howell, S.M., Ratcliffe, N.A., 1996a. Phagocytic reduction and effects on lysosomal membranes of polycyclic aromatic hydrocarbons, in haemocytes of Mytilus edulis. Aquat. Toxicol. 34, 273 – 290. Grundy, M.M., Ratcliffe, N.A., Moore, M.N., 1996b. Immune inhibition in marine mussels by polycyclic aromatic hdyrocarbons. Mar. Environ. Res. 42, 187 – 190. Halliwell, W.H., 1997. Cationic amphiphilic drug-induced phospholipidosis. Toxicol. Pathol. 25, 53 – 60. Hauton, C., Hawkins, L.E., Hutchinson, S., 1998. The use of neutral red retention assay to examine the effects of temperature and salinity on haemocytes of the European flat oyster Ostrea edulis (L.). Comp. Biochem. Physiol. 199B, 619 – 623. Hawkins, A.J., Day, A.J., 1999. Metabolic interrelations underlying the physiological and evolutionary advantages of genetic diversity. Am. Zool. 39, 401 – 411.

14

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15

Hein, L., Lu¨ llman-Rauch, R., Mohr, K., 1990. Human accumulation potential of xenobiotics: potential of catamphiphilic drugs to promote their accumulation via inducing lipidosis or mucopolysaccaridosis. Xenobiotica 20, 1259 – 1267. Hermans, J.H., Smedes, F., Hofstraat, J.W., Cofino, W.P., 1992. A method for the estimation of chlorinated biphenyls in surface waters: influence of sampling methods on analytical results. Environ. Sci. Technol. 26, 2028 –2034. Hinton, D.E., Lauren, D.J., 1990. Liver structural alterations accompanying chronic toxicity in fishes: potential biomarkers of exposure. In: McCarthy andand, J.F., Shugart, L.K. (Eds.), Biomarkers of Environmental Contamination. Lewis Publishers, Boca Raton, FL, pp. 17 –37. Howard, C.V., 1997. Synergistic effects of chemical mixtures: can we rely on traditional toxicology? Ecologist 27, 192 – 195. Hutchins, M.U., Veenhuis, M., Klionsky, D., 1999. Peroxisome degradation in Saccharomyces cere6isiae is dependant on machinery of macroautophagy and the Cvt pathway. J. Cell Sci. 112, 4079 –4087. Joachim, C., Gimzewski, J.K., Aviram, A., 2000. Electronics using hybrid-molecular and mono-molecular devices. Nature 408, 541 – 548. Kanzawa, F., Nishio, K., Fukuoka, K., Fukuda, M., Kunimoto, T., Saijo, N., 1997. Evaluation of synergism by a novel three-dimensional model for the combined action of cisplatin and etoposide on the growth of a human smallcell lung-cancer cell line, SBC-3. Int. J. Cancer 71, 311 – 319. Kauffman, S.A., 1993. The Origins of Order, Oxford University Press, Oxford, NY, 709p. Kim, J., Dalton, V.M., Eggerton, K.P., Scott, S.V., Klionsky, D.J., 1999a. Apg7p/Cvt2p is required for the cytoplasm-tovacuole targeting, macroautophagy, and peroxisome degradation pathways. Mol. Biol. Cell 10, 1337 –1351. Kim, S.-J., Bernreuther, D., Thumm, M., Podila, G.K., 1999b. LB-AUT7, a novel symbiosis-regulated gene from an ectomycorrhizal fungus, Laccaria bicolor, is functionally related to vesicular transport and autophagocytosis. J. Bacteriol. 181, 1963 – 1967. Kirisako, T., Baba, M., Ishihara, N., Miyazawa, K., Ohsumi, M., Yoshimori, T., Noda, T., Ohsumi, Y., 1999. Formation process of autophagosome is traced with Apg8p/ Aut7p in yeast. J. Cell Biol. 147, 435 – 446. Kitov, P., Sadowska, J.M., Mulvey, G., Armstrong, G.D., Ling, H., Pannu, N.S., Read, R.J., Bundle, D.R., 2000. Shiga-like toxins are neutralized by tailored multivalent carbohydrate ligands. Nature (London) 403, 669 –672. Koo, M., 1999. E-R and D in a virtual lab. Helix 1 (4), 18 –20. Kortenkamp, A., Altenburger, R., 1998. Synergisms with mixtures of xenoestrogens: a reevaluation using the method of isoboles. Sci. Total Environ. 221, 59 –73. Kurelec, B., Pivcevic, B., 1991. Evidence for a multixenobiotic resistance mechanism in the mussel Mytilus gallopro6incialis. Aquat. Toxicol. 19, 291 –302.

Lauffenburger, D.A., Linderman, J.L., 1993. Receptors: Models for Binding, Trafficking and Signaling, Oxford University Press, Oxford, NY, 365p. Liang, X.H., Jackson, S., Seaman, M., Brown, K., Kempkes, B., Hibshoosh, H., Levine, B., 1999. Induction of autophagy and inhibition of tumorigenesis by beclin 1. Nature (London) 402, 672 – 6. Link, A.J., Robison, K., Church, G.M., 1997. Comparing the predicted and observed properties of proteins encoded in the genome of Escheria coli. Electrophoresis 18, 1259 – 1313. Lipton, P., 1999. Ischemic cell death in brain neurons. Physiol. Rev. 79, 1431 – 1568. Livingstone, D.R., 1993. Biotechnology and pollution monitoring: use of molecular biomarkers in the aquatic environment. J. Chem. Tech. Biotechnol. 57, 195 – 211. Livingstone, D.R., Chipman, J.K., Lowe, D.M., Minier, C., Mitchelmore, C.L., Moore, M.N., Peters, L.D., Pipe, R.K., 2000. Development of biomarkers to detect the effects of organic pollution on aquatic invertebrates: recent molecular, genotoxic, cellular and immunological studies on the common mussel (Mytilus edulis L.) and other mytilids. Int. J. Environ. Pollut. 13, 56 – 91. Lowe, D.M., Moore, M.N., Evans, B., 1992. Contaminant impact on interactions of molecular probes with lysosomes in living hepatocytes from Dab (Limanda limanda). Mar. Ecol. Prog. Ser. 91, 135 – 140. Lowe, D.M., Fossato, V.U., Depledge, M.H., 1995a. Contaminant induced lysosomal membrane damage in blood cells of mussels M. gallopro6incialis from the Venice Lagoon: an in vitro study. Mar. Ecol. Prog. Ser. 129, 189 – 196. Lowe, D.M., Soverchia, C., Moore, M.N., 1995b. Lysosomal membrane responses in the blood and digestive cells of mussels experimentally exposed to flouranthene. Aquatic Toxicol. 33, 105 – 112. Luzikov, V.N., 1999. Quality control: from molecules to organelles. FEBS Letts. 448, 201 – 205. McCarthy, J.F., Shugart, L.R., 1990. In: J.F. McCarthy, L.R. Shugart (Eds.), Biomarkers of Environmental Contamination, Lewis Publishers, Boca Raton, FL, 457p. McFadzen, I., Baynes, S., Hallam, J., Beesley, A., Lowe, D., 2000. Histopathology of the skin of UV-B irradiated sole (Solea solea) and turbot (Scopthalamus maximus) larvae. Mar. Environ. Res. 50, 273 – 277. Maddox, J., 1998. What Remains to be Discovered, The Free Press, New York, 434p. Mason, R., 1990. Free radical metabolites of foreign compounds and their toxicological significance. Rev. Biochem. Toxicol. 87, 237 – 243. Minier, C., Moore, M.N., 1996. Multixenobiotic resistance in mussel blood cells. Mar. Environ. Res. 42, 389 – 392. Mizushima, N., Noda, T., Ohsumi, Y., 1999. Apg16p is required for the function of the Apg12p – Apg5p conjugate in the yeast autophagy pathway. EMBO J. 18, 3888 – 3896. Moore, M.N., 1990. Lysosomal cytochemistry in marine environmental monitoring. Histochem. J. 22, 187 – 191.

M.N. Moore / Aquatic Toxicology 59 (2002) 1–15 Moore, M.N., 1991. The Robert Feulgen Lecture 1990. Environmental distress signals: cellular reactions to marine pollution. Prog. Histochem. Cytochem. 23, 1 –19. Moore, M.N., Simpson, M.G., 1992. Molecular and cellular pathology in environmental impact assessment. Aquat. Toxicol. 22, 313 – 322. Moore, M.N., Willows, R.I., 1998. A model for cellular uptake and intracellular behaviour of particulate-bound micropollutants. Mar. Environ. Res. 46, 509 –514. Moore, M.N., Kohler, A., Lowe, D.M., Simpson, M.G., 1994. An integrated appraoch to cellular biomarkers in fish. In: Fossi, M.C., Leonzio, C. (Eds.), Non-Destructive Biomarkers in Vertebrates. Lewis/CRC, Boca Raton, FL, pp. 171– 197. Moore, M.N., Soverchia, C., Thomas, M., 1996. Enhanced lysosomal autophagy of intracellular proteins by xenobiotics in living molluscan blood cells. Acta Histchem. Cytochem. (Suppl.) 29, 947 –948. Moore, M.N., Lowe, D.M., Soverchia, C., Haigh, S.D., Hales, S.G., 1997. Uptake of a non-calorific, edible sucrose polyester oil and olive oil by marine mussels and their influence on uptake and effects of anthracene. Aquat. Toxicol. 39, 307 – 320. Mortimore, G.E., Poso, A.R., 1987. Intracellular protein catabolism and its control during nutrient deprivation and supply. Annu. Rev. Nutr. 7, 539 –564. Murdoch, A., Kaiser, K.L.E., Comba, M.E., Neilson, M., 1994. Particle-associated PCBs in Lake Ontario. Sci. Total Environ. 158, 113 – 125. Noble, D., Levin, J., Scott, W., 1999. Biological simulations in drug discovery. Drug Discov. Today 4 (1), 10 –16. Nott, J.A., Nicolaidou, A., 1990. Transfer of metal detoxication along marine food chains. J. Mar. Biol. Assoc. UK 70, 905 – 912. Ohsumi, Y., 1999a. Molecular mechanism of autophagy in yeast, Saccharomyces cere6isiae. Philos. Trans. R. Soc. Lond. 354B, 1577 – 1581. Ohsumi, Y., 1999b. Pictures in cell biology. Autophagosomes in yeast. Trends Cell Biol. 9, 162. Perrin, R.M., 1997. Crop protection: taking stock for the new millennium. Crop Protect. 16, 449 –456. Quarrie, J. (Ed.), 1992. Earth Summit ’92, Regency Press, London, 240p. Rashid, F., Horobin, R.W., Williams, M.A., 1991. Predicting the behaviour and selectivity of fluorescent probes for lysosomes and related structures by means of structure-activity models. Histochem. J. 23, 450 –459. Readman, J.W., Mantoura, R.F.C., Rhead, M.M., 1984. The physico-chemical speciation of polycyclic aromatic hydrocarbons (PAH) in aquatic systems. Fresenius Z. Anal. Chim. 319, 126 – 131. Ringwood, A.H., Connors, D.E., Hoguet, J., 1998. Effects of natural and anthropogenic stressors on lysosomal destabilisation in oysters, Crassostrea 6irginica. Mar. Ecol. Prog. Ser. 166, 163 – 171. Schaff, J., Fink, C., Slepchenko, B., Carson, J., Loew, L., 1997. A general computational framework for modeling cellular structure and function. Biophys. J. 73, 1135 – 1146.

15

Seglen, P.O., 1997. DNA ploidy and autophagic protein degradation as determinants of hepatocellular growth and survival. Cell Biol. Toxicol. 13, 301 – 315. Shin, J., 1998. P62 and the sequestrosome, a novel mechanism for protein metabolism. Arch. Pharm. Res. 21, 623 – 629. Shintani, T., Mizushima, N., Ogawa, Y., Matsuura, A., Noda, T., Ohsumi, Y., 1999. Apg10p, a novel protein-conjugating enzyme for autophagy in yeast. EMBO J. 18, 5234 – 5241. Slater, T.F., 1979. Biochemical studies on liver injury. In: Slater, T.F. (Ed.), Biochemical Mechanisms of Liver Injury. Academic Press, London, pp. 1 – 44. Smedes, F., 1994. Sampling and partition of neutral organic contaminants in surface waters with regard to legislation, environmental quality and flux estimations. Int. J. Environ. Anal. Chem. 57, 215 – 229. Soni, M.G., Mehendale, H.M., 1998. Role of tissue repair in toxicologic interactions among hepatotoxic organics. Environ. Health Perspect. 106 (Suppl. 6), 1307 – 1317. Stegeman, J.J., Lech, J.J., 1991. Cytochrome P-450 mono-oxygenase systems in aquatic species: carcinogen metabolism and biomarkers for carcinogen and pollutant exposure. Environ. Health Perspect. 90, 93 – 100. Svendseb, C., Weeks, J.M., 1995. The use of a lysosome assay for the rapid assessment of cellular stress from copper to the freshwater snail Vi6iparus contectus (Millet). Mar. Pollut. Bull. 31, 139 – 142. Tanida, I., Mizushima, N., Kiyooka, M., Ohsumi, M., Ueno, T., Ohsumi, Y., Kominami, E., 1999. Apg7p/Cvt2p: a novel protein-activating enzyme essential for autophagy. Mol. Biol. Cell 10, 1367 – 1379. Thevenod, F., Friedman, J.M., 1999. Cadmium-mediated oxidative stress in kidney proximal tubule cells induces degradation of Na+/K+ – ATPase through proteosomal and endo-lysosomal proteolytic pathways. FASEB J. 13, 1751 – 1761. Viarengo, A., 1989. Heavy metals in marine invertebrates: mechanisms of regulation and toxicity at the cellular level. Rev. Aquat. Sci. 1, 295 – 317. Viarengo, A., Canesi, L., Moore, M.N., Orunesu, M., 1994. Effects of Hg2 + and Cu2 + on the cytosolic Ca2 + level in molluscan blood cells evaluated by confocal microscopy and spectrofluorimetry. Mar. Biol. 119, 557 – 564. Warne, M.St.J., Hawker, D.W., 1995. The number of components in a mixture determines whether synergistic and antagonistic or additive toxicity predominate: the funnel hypothesis. Ecotox. Environ. Saf. 31, 23 – 28. Wedderburn, J., Cheung, V., Bamber, S., Bloxham, M., Depledge, M.H., 1998. Biomarkers of histochemical and cellular stress in Carcinus maenas: an in situ field study. Mar. Environ. Res. 46, 321 – 324. Winston, G.W., Moore, M.N., Kirchin, M.A., Soverchia, C., 1996. Production of reactive oxygen species (ROS) by hemocytes from the marine mussel, Mytilus edulis. Comp. Biochem. Physiol. 113C, 221 – 229. Yuan, W., Stromhaug, P.E., Dunn, W.A., 1999. Glucose-induced autophagy of peroxisomes in Pichia pastoris requires a unique E1-like protein. Mol. Biol. Cell 10, 1353 – 1366.