Go to PhD Thesis You're here! mscreview.gif (2656 bytes) Under Construction! Go to Scientific Work Go to Site Map

Literature Review

Biological Markers of Water Pollution
With specific reference to glutathione conjugation

Glutathione S-Transferase as a Biological Marker of Aquatic Contamination
Research Thesis in Applied Toxicology
Tomas James Rees, Portsmouth University, UK
In Collaboration with the Water Research Centre, Henley
August 1993

Return to the Index Page   |     Visitor number: to this page.


Tier I biomarkers
Tier II biomarkers

Phase I Biotransformation
Phase II Metabolism
        Glutathione Conjugation
        Other Pathways


In the attempt to define and measure the effects of pollutants on an ecosystem, biomarkers have attracted a great deal of interest. The principle behind the biomarker approach is the analysis of an organism's physiological or biochemical response to pollutant exposure. When compared with chemical residue analysis, biomarkers have an advantage of being a measure of the stress incurred in the organism, and so are more biologically relevant. However whereas chemical analysis for defined pollutants gives well defined concentration data, biomarkers are often more difficult to interpret.

When compared with population parameters, the opposite is the case. Biomarkers are often more easily quantifiable than population parameters, such as growth and reproduction, but whereas measurements of population parameters gives an accurate picture of the ecological effects of the pollutant, biomarkers are not necessarily indicative of a deleterious effect. That is, although all population stresses are necessarily preceded by a biochemical response, all biochemical responses need not be associated with a population stress.

In the identification of potential biomarkers, there are several factors which need to be investigated. First, the correlation between pollutant and response. Second, the correlation between response and effects on the scale of the whole organism and population. Third, the ease with which a response can be measured. Fourth, the specificity of the response; that is whether the biomarker is induced by a single chemical or restricted range of chemicals, or is indicative of a generalised response of the organism to toxicological stress. Both ends of the specificity spectrum can be useful - the important factor is that the response is repeatable.

All biomarkers suffer from complicating factors, that is, factors which affect levels of the biomarker but which are not directly related to pollution. An understanding of the natural variations in biomarkers is important if they are to be used to estimate the pollutant exposure of populations in the wild since the basal levels, as well as the effect on the biomarker of a given pollution exposure, is likely to vary with environmental conditions. For example, in the redbreast sunfish Lepomis auritas levels of two enzymes involved in phase I metabolism, EROD and cytochrome P-450, were found to vary with season. This variation correlates well with temperature with the lowest levels occuring in winter/spring (Jimenez, B.D., Oikari, A., Adams, S.M., Hinton, D.E., McCarthy, J.F., 1990). Similarly, seasonal variations have been found in all plaice liver detoxification enzymes, with the exception of microsomal glutathione S-transferase (George, S., Young, P., Leaver, M., Clarke, D., 1990).

Biomarkers can be very sensitive, and the detection of a response does not necessarily have implications for the health of the test organism. An analysis of fish caged downstream of a pulp mill in a Finnish study revealed that, although there was a clear effect on the enzymes responsible for detoxification, there were no remarkable disturbances in fish ionoregulation or other measures of stress-related status (Lindstrom-Seppa, P., Oikari, A., 1990). The opposite can also occur. In complex pollution situations there may be masking or inhibitory factors that can interfere with biomarker induction, and an unaltered presence of a particular biomarker does not necesarily mean the environment is harmless. Fish caged in the proximity of industrial discharges often exhibit lower levels of biochemical responses than those collected in areas more distant from the source, where the concentration of xenobiotics is actually lower (Oikari, A, Jimenez, B., 1992). It has been suggested that a suite of biomarkers should be analysed to minimise the chances of a false interpretation, and to provide a more rounded picture of the nature of the toxicological effects (Jimenez et al., 1990).

Biomarkers have been distinguished as tier I and tier II, according to their specificity (Sanders, B., 1990). Tier I biomarkers respond to specific contaminants, and often involve enzyme inhibition. Tier II biomarkers respond to general sublethal stress, and often involve enzyme induction (Thompson, H.M., Grieg-Smith, P.W., 1991).


Tier I Biomarkers

There are probably as many tier I biomarkers as there are toxicants. The following is a presentation of a few of the best known.

The inhibition of neural acetylcholinesterase by organophosphate and carbamate pesticides is the primary toxic mechanism of these compounds (Thompson, H.M., Grieg-Smith, D.W., 1991), and the degree of inhibition of the enzyme is closely related to tissue exposure (National Research Council of Canada, 1985), which makes them useful biomarkers for this type of pollution.

Metallothioneins are though to serve a protective function during heavy metal exposure by sequestering most of the free metal ions (Thomas, P., 1990). The toxicity to organisms of heavy metals can be correlated to the levels of metallothioneins in the liver (National Research Council of Canada, 1985).

Lead neurotoxicity can be assessed by the depression of gamma-aminolaevulinic acid dehydratase activity in blood (ibid, 1985).


Tier II biomarkers.

These are generalised stress responses, and their specificity to chemical insult and value as biomarkers varies markedly.

A simple approach is histopathology. For example, the liver plays a central role in toxicological response, and pollutant-induced pathological changes can be observed at all levels of liver histology (Hinton, D.E., Lauren, D.J., 1990). The disadvantage often encountered in histopathology is the difficulty involved in quantifying responses.

All organisms studied to date respond to adverse environmental conditions by synthesising a highly conserved group of proteins known variously as heat shock proteins (HSP) or stress response proteins (SRP) (Thomas, P., 1990). Although little work has been done on their utility as biomarkers, they have been found to be produced in 2 molluscs (Mytilus edulis and Collisella pelta) in response to both heat shock and cadmium (Sanders, B.M., 1988), and in a bacterium (Alcaligenes eutrophus) in response to cadmium and the herbicide 2,4,5-T (2,4,5-trichlorophenoxyacetic acid) (Amy, P.S., Hiatt, H.D, Dupre, R.K., 1993).

Many environmental contaminants enhance oxidative stress in animals, and the presence of specific lesions known to arise as a result of such stress are present in many aquatic animals (Winston, G., 1991). Many redox cycling compounds are important contaminants in aquatic systems (Wenning, R.J., Di Giulio, R.T., Gallagher, E.P., 1988). A breakdown product of lipid peroxidation, malondialdehyde, can be analysed by colorimetric methods and its production has been shown to be enhanced in vitro in response to pollutants such as polychlorinated biphenols (PCBs) and DDT (Kamohara, K., Yagi, N., Itokawa, Y., 1984), cadmium, mercury and paraquat (Wooford, H.W., Thomas, P., 1988). As such it is a potentially useful assay for a wide range of environmental toxins. However nutritional status and age can also influence lipid peroxidation (Thomas, P., 1990).

Another mechanism for the detoxification of oxidising compounds involves ascorbic acid. Levels have been found to fluctuate with a wide variety of toxicological and environmental factors, with the exact response varying according to tissue (Thomas, P., 1990). The utility of this measure is hampered by the number of confounding factors, notably the semi-predictable response due to season. However a marked effect due to oil toxicity has been found (Thomas, P., 1987).

A wide variety of standard mutagenicity tests have been applied to biomonitoring. These include the use of micronuclei counts (Spies, R.B., Stegeman, J.J., Rice, D.W. Jr., Woodin, B., Thomas, P., Hose, J.E., Cross, J.N., Prieto, M., 1990), DNA adducts (Varanasi, U., Stein, J.E., Nishimoto, M., Reichert, W.L., Collier, T.K., 1987) and strand breakage (Thomas, P., 1990). All three methods provide an accurate non-specific assessment of genetic damage. Recovery of strand break damage is rapid (around 12 days), whereas other tests are more persistant (up to 2 months with benzopyrene adducts). A combination of these assays provides a powerful method for assessing short and long term genotoxicity.

Stress response often has a hormonal basis and alterations in the levels of the hormones involved can be used as a biomarker. Cortisol is a steroid hormone associated with immunosupression, growth retardation and susceptibility to disease (Bowman, W.C., Rand, J., 1981). It is produced in response to many stressors, including poor water quality (Thomas, P., Lewis, D.H., 1987).



A special subsection of biomarkers involve the range of systems involved in the metabolism of xenobiotics. These fall somewhere between the designations of tier I and tier II as they are induced specifically in response to contaminants but are relatively non-specific with respect to individul contaminants. Although the mechanism of induction is often completely unknown, their properties give them great potential as biomarkers. Xenobiotic metabolism is often referred to as biotransformation, and the pathways involved are usually divided into phase I and phase II.


Phase I Biotransformation

Phase I is the predominant biotransformation pathway. It generally involves the addition or exposure of functional groups on the xenobiotic, for example by oxidation or hydrolysis. By far the most extensively examined system, from the point of view of biomarkers, is the mixed function oxidase system (MFO) which involves oxidation by a variety of isozymes of cytochrome P-450 (Sipes, I.G., Gandolfi, A.J., 1991).

There are a great many studies of P-450 in marine animals (for a review, see Goksoyr, A., Forlin, L., 1992). In this section only some of the more recent publications are considered.

In an examination of cytochrome p-450 isozymes in fish from San Francisco Bay, cytochrome P-450E showed the greatest and most consistent induction due to pollution load (Spies, R.B., Stegeman, J.J., Rice, D.W. Jr., Woodin, B., Thomas, P., Hose, J.E., Cross, J.N., Prieto, M., 1990). Ethoxyresorfurin-O-deethylase (EROD) activity was found to be enhanced in fish from a stream contaminated by metals, organic chemicals and radionucleides (Jimenez, B.D., Oikari, A., Adams, S.M., Hinton, D.E., McCarthy, J.F., 1990). A major pathway for the induction of P-450 is the Ah receptor, and by labelling this with a photoaffinity ligand, it has been shown to be present in teleost and elasmobranch fish, but not in invertebrates (Hahn, M.E., Poland, A., Glover, E., Stegeman, J.J., 1992). Its presence or absence corresponds to inducibility of cytochrome P-4501A and sensitivity to the toxic effects of TCDD. The induction of EROD (for example by benzo(a)pyrene) can be inhibited by the simultaneous application of an hepatotoxin such as carbon tetrachloride (Oikari, A., Jimenez, B., 1992).

Cytochrome P-450 can be assessed by catalytic activity measurements, though recently an enzyme-linked immunosorbent assay (ELISA) has been developed (Goksoyr, A., Husoy, A.M., 1992). This method showed good correlation with environmental exposure and catalytic assessments, but also gave new and important information where catalytic measurements failed to reveal effects, for example where catalytic activity is low - such as in larvae and extrahepatic tissue such as gills.


Phase II Metabolism

Phase  II reactions are biosynthetic reactions where the foreign compound or a phase I-derived metabolite is covalently linked to an endogenous molecule (Sipes, I.G., Gandolfi, A.J., 1991). There are several pathways of phase II metabolism, and the activities of the various pathways varies with species, environment, and substrate.

Glutathione Conjugation.

Glutathione conjugation involves the addition of the tripeptide (GSH) to an electrophilic site on the substrate, catalysed by glutathione S-transferase (GST). This forms the intitial step in the formation of N-acetylcysteine (mercapturic acid). The rate limiting step in GSH synthesis is catalysed by glutamyl-cysteine synthetase (GCS). Oxidised glutathione (either GS-SG or GS-substrate) can be reduced by glutathione reductase (GR). GST, GSH, GCS and GR have all been investigated as potential biomarkers.

Substrates for GST share three common features: They must be hydrophobic to some degree, they must contain an electrophilic carbon atom, and they must react nonenzymatically with glutathione at some measurable rate (Sipes, I.G., Gandolfi, A.J., 1991). Indeed it has been shown that compounds which are good substrates for GST also lower liver GSH levels soon after administration to the rat (Boyland, E., Chasseaud, L.F., 1970). The turnover of GSH is very rapid, with the half life in rat liver being around 4 hours (Moron, M.S., Depierre, J.W., Mannervik, B., 1979).

Hepatic GSH concentration is significantly lower in fathead minnows and rainbow trout compared to either rats or mice and the fact that a similar Km for GSH is found in all species suggests that the GST pathway in these fish may be limited by GSH availability (Wallace, K.B., 1989). The exact response of GSH to toxins is complicated. A dose dependent decrease in GSH was found in rainbow trout erythrocytes treated with monochloramine (NH2Cl), a widely used drinking and waste water disinfectant (Buckley, J.A., 1982). In rainbow trout exposed to fuel oil, GSH levels were reduced after 3 days and increased after 21 days (Steadman, B.L., Farag, A.M., Bergman, H.L., 1991). Repetitive sublethal doses of the chlorinated pesticide endosulfan caused an increase in GSH levels in the hepatopancreas of the field crab Paratelphusa hydrodromus after 96 hours (Yadwad, V.B., 1989). Gallagher and Di Guilio (1992) found that exposure of catfish to the fungicide chlorothalonil (tetrachloroisophthalonitrile (TCIN)) causes an increase in GCS activity after 144 hours, and a similar increase was found in hepatic GCS activity (Gallagher, E.P., Canada, A.T., Di Giulio, R.T., 1992). Liver GSH also increased in Salmo spp. and Galaxias spp. after 96 hours exposure to chlorothalonil (Davies, P.E., 1985b).

An increase in both GSH levels and GSH synthesis has been found in fish exposed to lead (Thomas, P., Juedes, J.M., 1992), and increased hepatic concentrations of GSH have been found in cadmium and fuel oil exposed striped mullet and sediment exposed catfish (Winston, G.W., Di Giulio, R.T., 1991). Copper has been found to enhance trout liver GSH, but had no effect in the gills (Lauren, D.J., McDonald, D.G., 1987). Copper exposure caused a decrease in mussel tissue GSH content, but cadmium and zinc slightly increased levels (Viarengo, A., Pertica, M., Canesi, L., Biasis, F., Cecchini, G., Orunesu, M., 1988). Mercury has a double edged effect, depleting hepatic GSH by both the formation of Hg-GSH complexes and the inhibition of glutathione reductase (Allen, P., Min, S.Y., Keong, W.M., 1988). Acid soluble thiol (which includes metallothionein and GSH) is raised in response to a wide variety of potential pollutants, such as cadmium and dibenzofuran (Thomas, P., Wooford, H.W., 1984).

Interpretation of changes in GSH levels is complicated by the fact that it also participates in redox defence (see above). Redox cycling compounds, for example paraquat, have been found to induce the synthesis of glutathione peroxidase (GPx) (Winston, G.W., Di Giulio, R.T., 1991). Paraquat has also been found to cause increased glutathione (GSH) levels in mussels (Wenning, R.J., Di Giulio, R.T., 1988). The activity of glutathione reductase is known to increase in rats exposed to PCBs (Kamohara, K., Yagi, N., Itokawa, Y., 1984).

Since glutathione plays a protective role, its depletion will enhance the toxicity of chemicals which are normally detoxified by glutathione conjugation. For example the toxicity to Daphnia magna of CDNB, an algicide used in cooling water (and also the standard substrate in assays for GST), is enhanced by previous depletion of GSH (Dierickx, P.J., 1986), and the prior exposure of Daphnia to CDNB will increase the susceptibility to pentachlorophenate, despite also causing an induction of GST activity (LeBlanc, G.A., Cochrane, B.J., 1985). This suggests the possibility of substantial synergistic effects. Cytsolic fractions can be separated by gel chromatography, and the appearance in fish of metabolites of DDT in fractions other than the GSH pool is associated with fin erosion (Brown, D.A., Bay, S.M., Gossett, R.W., 1985).

Reports of the in situ value of GSH as a biomarker are mixed. GSH levels in the livers of fish taken from coastal water polluted by metals were found to be enhanced by 2 to 3 times (Brown, D.A., Bay, S.M., Greenstein, D.J., Szalay, P., Hershelman, G.P., Ward, C.F., Westcott, A.M., Cross, J.N., 1987). However no variation in either GSH or GST was found in English sole taken from different exposure sites, but there was an induction in both factors which was associated with hepatic lesions (Jenner, N.K., Ostrander, G.K., Kavanagh, T.J., Livesey, J.C., Shen, M.W., Kim, S.C., Holmes, E.H., 1990).

GST has been found in all organisms in which it has been investigated (for example see Stenersen, J., Kobro, S., Bjerke, M., Arend, U., 1987, or Dierickx, P.J., 1984), and it seems likely that it is ubiquitous.

Although GST activity often increases with exposure to contaminants, the induction of enzyme activity is not as marked as that of the MFO system (Jiminez, B.D., Stegeman, J.J., 1990). The activity of GST is known to increase in rats exposed to PCBs (Kamohara, K., Yagi, N., Itokawa, Y., 1984). Similarly, fish injected intraperitoneally with benzo(a)pyrene exhibit a significant increase in GST activity (Fair, P.H., 1986). A marked increase in liver GST activity was found in three different freshwater fish species (S. gairdneri, G. maculatus, G. truttaceus) exposed to chlorothalonil for 96 hours (Davies, P.E., 1985b). Similarly exposure of catfish to chlorothalonil resulted in an increase in gill GST after 72 hours. After 4 weeks of exposure of rainbow trout to cadmium there was an initial decrease in hepatic and nephrotic GST, followed by a net increase in hepatic GST activity (Forlin, L., Haux, C., Karlsson-Norrgren, L., Runn, P., Larsson, A., 1986). However, 6 days of exposure to cadmium results in no alteration in the GST activity of plaice liver (George, S.G., 1989). GST activity in the hepatopancreas of the field crab Paratelphusa hydrodromus is enhanced after 48 hours exposure to endosulfan, with maximal activity ocurring after between 96 and 192 hours (Yadwad, V.B., 1989).

In contrast, no effect on GST activity towards either CDNB or DCNB was found in rainbow trout exposed for 14 days to the insecticide endosulfan, despite a significantly enhanced MFO activity (Jensen, E.G., Skaare, J.U., Egaas, E., Goksoyr, A, 1991). 3-Methylcholanthrene likewise induced cytochrome P-450 in a variety of freshwater and estuarine fish, but had no effect on GST (James, M.O., Heard, C.S., Hawkins, W.E., 1988). Exposure of English sole to contaminants such as B(a)P and contaminated sediment resulted in no increase in GST activity, despite an enhancement of aryl hydorcarbon hydroxylase (AHH) activity (Collier, T.K., Varanasi, U., 1991). Exposure of sunfish to the hepatotoxins carbon tetrachloride or beta-naphthoflavone caused a reduced hepatic GST activity (Oikari, A., Jimenez, B., 1992).

GST was found in the marine mussel Mytilus edulis and activity towards CDNB was found to be unaffected by size but significantly enhanced in shore grown (as opposed to rope grown) mussels, and in mussels taken from polluted environments (Sheenan, D., Crimmins, K.M., Burnell, G.M., 1991). GST activity in the freshwater pea mussel Sphaerium corneum was found to be enhanced by 70% after exposure to lindane (Boryslawskyj, M., Garrood, A.C., Pearson, J.T., Woodhead, D., 1988).

In Daphnia magna, studies on a series of six chlorinated phenols (CP) have found that there is a clear relationship between lipophilicity, toxicity, and inhibition of GST activity (LeBlanc, G.A., Hilgenberg, B., Cochrane, B.J., 1988). This appears to be a pharmacodynamic effect since the internal concentration of the various CPs was relatively constant.

The importance of GST in detoxification has been shown by Varanasi et al. (1987) and Davies (1985b). Varanasi et al. (1987) showed that the primary factor determining the difference in genotoxicity of B(a)P in two fish species was the relative activity of GST. Davies (1985b) found that the chlorothalonil LC50 values for three fish species, Salmo gairdneri, Galaxias maculatus and G. auratus was in good agreement with the total hepatic GST activity towards the substance (Davies, P.E., 1985).

The efficacy of GST as a biomarker in field situations has been investigated in three fish species collected downstream from a PCB incineration plant along the Rhone River (Monod, G., Devaux, A., Riviere, J.L., 1988). GST activity in these fish was found to be significantly increased In general induction of GST activity followed the same trend (as distance from the plant increased) as the induction of the MFO system, but the ratio of induction showed considerable interspecific variation. An assay of rainbow trout caged for three weeks downstream from a pulpmill showed enhanced activity in gill GST but not hepatic GST, with the effect decreasing with distance from the discharge site. Kidney GST was also slightly affected, with fish closest to the discharge having lowered GST activity and those further away having a raised GST activity (Lindstrom-Seppa, P., Oikari, A., 1990).


Glucuronidation represents one of the major phase II conjugation reactions in the conversion of both exogenous and endogenous compounds to polar, water soluble compounds. The widespread occurrence of this detoxification mechanism in species, the broad range of substrates that are accepted, and the diversity in the nature of acceptor groups make conjugation with glucuronic acid qualitatively and quantitatively the most important conjugation reaction (Sipes, I.G., Gandolfi, A.J., 1991).

Glucuronidation by UDPGT (Uridine Diphosphate GlucoronylTransferase) is the major pathway for the elimination of synthetic pyrethroids in rainbow trout (Bradburg, S.P., Coats, J.R., 1986). Similarly, the major metabolite of p-nitrophenol in the rice field crayfish Procambarus clarkii is the beta-D-glucuronide (Foster, G.D., Crosby, D.G., 1986). Studies with plaice hepatocytes have shown that glucuronidation is the major phase II pathway for PAHs with glutathione conjugation contributing significantly, and sulphation being very low (Leaver, M.J., Clarke, D.J., George, S.G., 1992). The presence of UDPG in cell-free tissue extracts from rainbow trout increased biodegradation by phase I enzymes by glucuronidation of the products (Edwards, R., Millburn, P., 1986).

As in mammals, fish UDPGTs have been shown to be induced by a number of xenobiotics (Clarke, D.J., George, S.G., Burchell, B., 1991). UDPGT has been found to be induced by beta-naphthoflavone, but not phenobarbital (Bolinger, R.A., Kennish, J.M., 1992). Four weeks exposure of rainbow trout to cadmium caused an inhibition of the glucuronidation reaction in the kidney (Forlin, L., Haux, C., Karlsson-Norrgren, L., Runn, P., Larsson, A., 1986). 3-MC strongly induces UDPGT in plaice hepatocytes (Leaver, M.J., Clarke, D.J., George, S.G., 1992). Comparisons of these results are made very difficult by the fact that there are a multitude of tests for UDPGT activity in use, mainly involving the use of different substrates.

Two field trials of the use of UDPGT as a biomarker have been conducted. A 40% increase in UDPGT activity was found in salmon caged for 21 days in an area contaminated by petrochemicals (Nikunen, E., 1985). Activity of UDPGT in rainbow trout livers was found to be similarly increased at distances between 4 and 11 km from a pulp mill outfall (Oikari, A., Kunnamo-Ojala, T., 1987).


Methylation is a common biochemical reaction for the metabolism of endogenous compounds but is not usually a quantitatively important pathway for xenobiotic biotransformation (Sipes, I.G., Gandolfi, A.J., 1991). However, it is an important mechanism for the detoxification of sulphide (Bagarinao, T., 1992), and levels of arsenite methylation have been found to be enhanced in algae from polluted lake water (Baker, M.D., Wong, P.T.S., Chau, Y.K., Mayfield, C.I., Inniss, W.E., 1983).

Other Pathways.

Other pathways of phase II metabolism include acetylation, amino acid conjugation and sulphation (Sipes, I.G., Gandolfi, A.J., 1991). These have not been investigated as potential biomarkers although they play a large role in detoxification. Their targets include those xenobiotics with amine, hydroxyl and carboxylic acid functional groups, respectively.



The enzymes involved in biotransformation are well suited to simple bioassay tests by virtue of the fact that they have evolved to metabolise novel chemicals and change their properties. This is often utilised to develop a spectrophotometric assay whereby an aromatic chemical is metabolised and the change in U.V. absorbance is measured. However the development of monoclonal antibodies and ELISA techniques have meant that it is now possible to develop an assay for any protein which can be successfully purified. In this case, biomarker levels are not measured as activity, but concentration. The advantages of the ELISA assay have been enumerated by Goksoyr (1991). It has high sensitivity, detectability and specificity. It is simple and allows the assesment of large numbers of samples simultaneously. It is also less sensitive to denaturation of the biomarker (Goksoyr, A., 1991). In theory this suggests that, for any biomarker which fulfills the other criterion, a practical assay may be developed. However, there may be a problem in relating ELISA and spectrophotometric results as inhibition or partial degradation of the biomarker by pollutants may cause anomalous results. The relationship between ELISA measurements and environmental pollution will therefore need to be investigated.

For general purpose use, tier I biomarkers are not responsive to a wide enough range of pollutants to be used as biomarkers, but can be extremely useful in assessing pollution in well defined situations.

In general, SRPs are good indicators of exposure, but as a measure of toxic effects they are less well proven (Bradley, B.P., 1993). Measurements of oxidative stress and genotoxicity tests are very good indicators of toxicity to the organism, however little work has been carried out into the predictability of the response to pollutants. It is to be expected that these measurements would only respond to a limited range of toxins. More general measures, such as cortisol and ascorbic acid, respond to such a wide range of influences that their utility as biomarkers is limited.

Biotransforming enzymes, on the other hand, are very good indicators of exposure. There are a great many studies which have succesfully used MFO activity as a biomarker of environmental pollution (Goksoyr, A., Forlin, L., 1992) and the good correlation with pollutant levels, coupled with the specificity for chemical stress, gives these test great utility.

In contrast to MFO activity, very little work has been done with phase II enzymes as indicators of environmental pollution. This gives rise to a potential for error in the interpretation of results. Since phase I metabolism generally increases toxicity, whereas phase II generally detoxifies, the relationship between the two is often crucial to determining the toxicity of a compound to the test organism. For example, Varanasi et al. (1987) have found that whereas both English sole and starry flounder have similar levels of P-450 activity, English sole suffered 2-4 times the number of DNA adducts on exposure to benzo(a)pyrene. Since the starry flounder has twice the GST activity of English sole, they reasoned that detoxification of the activated products of P-450 activity by glutathione conjugation was the key factor in determining genotoxicity. Given this, the reduced expression of flounder GST activity by agents which induce cytochrome P-450 (such as PAH - see above) has important consequences.

Similarly, Edwards and Millburn (1986) have found that, across a range of species, the toxicity of pyrethroids was in part determined by the differences in rates of phase I metabolism. Phase I metabolism, in turn, was enhanced by higher activity of phase II pathways which removed the products.

Although the activity of UDPGT in marine animals is well substantiated, the lack of a standard test makes it very difficult to predict a response to contamination. By comparison, the use of a relatively standard assay for GST means that there a good assessment of predictability can be made. In general, GST is reasonably responsive to pollution, but the exact nature of the response varies according to pollutant, time scale, and species.

Biotransforming enzymes are unfortunately not altogether representative of a toxic effect on an organism. Although a poisoned animal may often show enhanced biotransformation (unless the poisoning is such that the synthesis of proteins is retarded, or the pollutant is cytotoxic), an effective biotransformation system may detoxify the poison to the extent where no further effects on the organism can be observed.

In summation, tier II biomarkers are representative of toxic effects, but either limited in their responsiveness or highly unpredictable. Of the biotransforming enzymes, MFO is responsive to pollution in the environment and predictable, but its representivity can be uncertain. This uncertainty can be reduced by a measurement of phase II activity. Due to the disparate nature of the work on UDPGT to date, this essentially means an analysis of GST.

The minimal suite of biomarkers for the detection of stress on an ecosystem due to organic contaminats would therefore appear to be MFO and GST. The next immediate step is the identification of a suitable test organism and assay (whether spectrophotometric or immunological). The potential of a GSH assay as a single measurement for a general purpose indication of toxic stress and biotransformation should also be investigated.


This page was last updated on Tuesday, 22 July 2003 21:48.