Artículos

 

Induction of metallothioneins in Mullus barbatus as specific biomarker of metal contamination: A field study in the western Mediterranean

 

Inducción de metalotioneínas en Mullus barbatus como biomarcador específico de contaminación metálica: Estudio de campo en el Mediterráneo occidental

 

J. Benedicto*, C. Martínez-Gómez and J. Campillo

 

^* Instituto Español de Oceanografía Centro Oceanográfico de Murcia, Varadero 1, 30740 San Pedro del Pinatar, Murcia, Spain. * E-mail: benedicto@mu.ieo.es

 

Recibido en junio de 2003;
aceptado en julio de 2004.

 

Abstract

Red mullet (Mullus barbatus, Linnaeus, 1758) is one of the fish species used as bioindicator in the Mediterranean Pollution Monitoring Programme (MEDPOL). This paper presents the results of a field study on the induction of hepatic metallothioneins (MT) in red mullet as specific biomarker of metal contamination. Specimens of M. barbatus were collected in six areas with different degrees of anthropogenic activities along the Iberian Mediterranean coast. The environmental exposure to heavy metals of the populations was determined through the analyses of surface sediments collected in each study area. The MT levels obtained showed statistically significant differences between areas, but not between specimens of different sex. Furthermore, the highest MT levels were found in the most polluted areas. Hepatic MT levels in red mullet reflect the metal contamination present in surface sediments, especially for Cu, for which a significant linear regression was found. The MT levels are better correlated with Cu and Cd concentrations than with Zn and Hg concentrations. The MT values obtained in this study are the first to be reported for this species in the Iberian Mediterranean and they should help in the interpretation of MT levels and to establish MT background levels for M. barbatus in the western Mediterranean.

Key words: biomarker, heavy metals, metallothioneins, Mediterranean Sea, Mullus barbatus.

 

Resumen

El salmonete de fango (Mullus barbatus, Linnaeus, 1758) es una de las especies utilizadas como bioindicador en el Programa de Seguimiento de la Contaminación del Mar Mediterráneo (MEDPOL). Este trabajo presenta los resultados de un estudio de campo sobre la inducción hepática de metalotioneínas (MT) en salmonete de fango como biomarcador específico de contaminación metálica. Los especímenes de M. barbatus fueron capturados en seis áreas de la costa Mediterránea Ibérica con diferentes grados de actividad antropogénica. La exposición de las poblaciones a la contaminación metálica se determinó analizando los sedimentos superficiales de cada una de las áreas de estudio. Los niveles de MT obtenidos mostraron diferencias significativas entre áreas pero no entre especímenes de diferente sexo. Además, los mayores niveles de MT registrados correspondieron a las áreas más contaminadas. Los niveles de MT reflejan la contaminación metálica existente en los sedimentos superficiales, especialmente por cobre, para el que se demostró la existencia de una regresión lineal significativa. Los niveles de MT están mejor correlacionados con cobre y cadmio, que con zinc y mercurio. Los niveles de MT obtenidos en este estudio son los primeros medidos en esta especie en el Mediterráneo Ibérico y podrán ayudar a interpretar los niveles hepáticos de MT así como a establecer los niveles basales para M. barbatus en el Mediterráneo occidental.

Palabras clave: biomarcador, Mediterráneo, metales pesados, metalotioneínas, Mullus barbatus.

 

Introduction

Pollution of the marine environment with heavy metals from land-based sources (industrial, agriculture, mining, etc.) is one of the specific problems of the Iberian Mediterranean coast that needs to be addressed. A topic of research that has received considerable attention in the last years is the binding of metals by metallothioneins. Metallothioneins (MT) are cytosolic, low molecular weight, metal-binding proteins that appear to be present in virtually all vertebrates and invertebrates. MT synthesis is induced by the exposure of organisms to excess levels of common metal ions such as Cu, Zn, Cd, Hg, Co, Ni and Ag (Bucheli and Fent, 1995). The mechanism of metal detoxification by MT occurs via metal-initiated transcriptional activation of MT genes (Hogstrand and Haux, 1991). Although under normal conditions only Cu and Zn appear to be associated with MT, a wide range of metal ions can bind to MT in vitro or after excess exposure (Nielson et al., 1985). Microcosm experiments have demonstrated that fish spending some time in a polluted area will accumulate pollutants and will display biological effects which can be documented by biomarker analyses such as MT, cytochrome P₄₅₀ proteins (CYP1A) and 7-ethoxyresorufin O-deethylase hepatic activity (EROD) (Goksoyr et al., 1996). The induction of hepatic MT levels in fish due to heavy metal exposure is a well-established biomarker (George and Olsson, 1994), and it is included in several monitoring programmes (JAMP/OSPAR, MEDPOL/PAM) as an early indicator of biological effects of metal pollution. The use of MT for pollution monitoring is based on the assumption that an increased level is primarily caused by exposure to contaminants (Hylland et al., 1998).

Red mullet, Mullus barbatus, is a demersal fish associated with muddy sediments and recommended for use as bioindica-tor in the Mediterranean Pollution Biomonitoring Programme (MEDPOL) (UNEP/RAMOGE, 1999). For the study of biological effects of pollution, the use of a biomarker test battery is recommended to determine the global effects on the target species. Studies carried out using M. barbatus have already shown that this fish is a good bioindicator for monitoring CYP1A, EROD, acetyl cholinesterase (AChE), catalase, glutathione S-transferase (GST), superoxide dismutase, glu-tathione peroxidase and DT-diaphorase activities, as well as DNA adducts (Mathieu et al., 1991; Burgeot et al., 1994, 1996). Several projects have evaluated the suitability of MT as biomarker of metal pollution in marine fish (Chang, 1995; Hylland et al., 1992, 1996); however, few works have been published in which this biomarker has been applied to M. barbatus.

In order to differentiate nonlethal responses to anthropogenic perturbations from natural variability, it is necessary to establish baseline databases for each type of biochemical measurement that is used (Engel, 1988). On the other hand, field validation is essential for the application of any biochemical or physiological index as a biomonitoring tool. Taking into account these premises, the aim of our study is to measure hepatic MT in M. barbatus from several areas along the Iberian Mediterranean coast and to assess the possible correlation between MT levels in each area and the metal burdens in the total fraction of surface sediment.

 

Material and methods

Specimens of M. barbatus were caught by trawling during the postspawning season (October 1999) in six areas along the Iberian Mediterranean coast exposed to different degrees of anthropogenic activities (fig. 1). Collected specimens were immediately sacrificed, sexed, weighed, their length measured and liver removed. Sexual maturity was determined on board. Liver samples were stored in liquid nitrogen. Ten livers were used for MT analysis in each group studied for each area. At each study area, sediment samples were collected using a box-corer to determine metal contents. The number of sediment samples for each area were unequal due to logistic reasons (Portman = 9, C. Palos = 9, Segura River mouth = 12, Tabarca Island = 4, Castellón = 9 and Columbretes Islands = 9). Water bottom temperature and salinity were also recorded.

Sediment samples were lyophilized, dried at room temperature and the <2-mm fraction was separated. Once the total fraction was dried at 105°C to constant weight, the sample was digested in a microwave oven system using Teflon reactors. For Hg determination, 0.5 g of sample was digested with 3 mL of 65% nitric acid applying the following programme: 40 psi for 2 min and 100 psi for 15 min. For Cu, Cd and Zn, 0.5 g of sample were digested with 1 mL of aqua regia and 5 mL of hydrofluoric acid, applying the following programme: 50 psi for 5 min and 100 psi for 10 min. The quantitative analyses were carried out by atomic absorption spectrophotometry (Perkin Elmer model 4110 ZL), using graphite furnace (Cd), flame (Cu and Zn) and cold vapour (Hg). Each sediment sample was measured in duplicate. The accuracy of the analytical procedures was tested and controlled using certified reference material (PACS-1, BCSS-1 and CRM-277) and participating in the intercomparison exercise IAEA-405 organized by the International Agency of Atomic Energy.

To extract the MT, purified liver fractions were washed in cold 1.15% KCl solution and homogenized with six vertical strokes at 3500 rpm in a Potter-Elvehjem teflon-glass homoge-nizer tube on ice using a homogenization buffer (20% glycerol [v/v], 0.1 mM PMSF, 1 mM EDTA, 1 mM DTT and 50 mM Tris-Cl [pH 7.8]) in a ratio of 1:5, w:v. The crude homogenate was centrifuged at 30,000 g for 20 min at 4°C, and the partially purified fraction obtained was used to quantify MT. MT content was measured applying a spectrophotometric assay method (Viarengo et al., 1997; UNEP/RAMOGE, 1999) modified for fish, based on the estimation of the sulfhydryl content of MT protein using Ellman's reagent (Ellman, 1959). MT levels were determined by assuming its molecular weight to be 6000 DA (Livingstone, 1993). The accuracy of the analytical procedures was tested by participating in the intercomparison exercise organized by Piemonte University, under the direction of Dr. A. Viarengo within the framework of the MEDPOL programme.

To eliminate the effect of specimen size on MT content (Benedicto et al., 2001; Rotchell et al., 2001), only data for specimens in the 14-18 cm size range were used. A confidence limit of 95% was established in the statistical analysis. Differences in hepatic MT levels in M. barbatus of different gender and collected from different areas were tested statistically using two-way ANOVA on loge-transformed data. For each gender, differences in MT levels between areas was tested applying one-way ANOVA (Sokal and Rohlf, 1981). Differences in metal concentrations in sediment samples from different areas were also tested using one-way ANOVA. In all cases where significant differences were found (P < 0.05), the Student-Newman-Keuls (SNK) test was applied in a pairwise comparison. The relationships between the metal mean concentrations in sediment and MT mean levels in each area were evaluated using multiple regression analysis on loge-transformed data using hepatic MT levels as dependent variable (Snedecor and Cochran, 1980). In addition, the relationships between sediment metal content (Cu, Zn, Cd and Hg) and MT levels in specimens from the same area were evaluated using Pearson's regression analysis (Siegel, 1956). Statistical treatments also included cluster analysis (CA), using the Bray-Curtis similarity index (SI). Statistical methods were performed using PRIMER v.5.2 and SPSS v.9 software.

 

Results

Hepatic MT levels in M. barbatus, expressed as ng g^-1 of hepatic tissue, and metal concentrations (mean ± standard deviation) in the total fraction of sediment collected in each area, referred to a 1% COOX and expressed in µg g^-1 dry weight, are presented in table 1. The two-way ANOVA showed significant differences in MT levels between areas (P = 0.000), but not between specimens of different sex (P = 0.068) from the same areas. For both genders, the one-way ANOVA showed significant differences in MT levels obtained in Cabo de Palos (females, P = 0.001; males, P = 0.001) and the other study areas (fig. 2). The highest concentrations for the four metals analyzed were found in Portman and the lowest were found in Columbretes Islands, except for Cu, which showed the lowest concentration in Cabo de Palos. The results obtained with the SNK test showed the existence of significant homogeneous subsets among areas with regard to metal concentrations and MT content (table 2). The results of the SNK test were confirmed by the application of CA, grouping the variable "area" according to metal concentrations and MT levels. The CA showed that areas were divided into several groups, the Portman area being the most differentiated (fig. 3). All the previous results show the existence of a gradient of contamination concentrations and of MT levels. The application of CA to MT, Cd, Cu, Zn and Hg revealed that MT levels are better correlated with Cu and Cd (SI = 78% and 62%, respectively) than with Zn or Hg concentrations (fig. 4). This fact explains why MT levels in Cabo de Palos were lower than in Columbretes Islands, in spite of the fact that the metal load was greater in Cabo de Palos, with the exception of Cu (table 1).

In a multiple regression of MT levels against Cu, Zn, Cd and Hg in female red mullet, Cu alone explained most of the variation resulting from the four metals in the regression (r = 0.860, P < 0.028). This result was supported by a parametric correlation analysis of MT levels against Cu, Cd, Zn and Hg. MT levels correlated better with Cu (r = 0.740, P = 0.028) than with Cd (r = 0.368, P = 0.202) or Zn (r = 0.276, P = 0.284) (fig. 5).

The temperature and salinity of the bottom water during the sampling were as follows: Portman, 16.11°C and 37.38; Cabo de Palos, 14.88°C and 38.06; Segura River mouth, 20.35°C and 37.51; Tabarca Island, 20.23°C and 37.65; Castellón, 20.85°C and 38.02; and Columbretes Islands, 18.53°C and 36.89. The temperature values showed differences between areas of up to 5°C. Salinity values were homogeneous in all sampling areas (37.72 ± 0.05 SE). Pearson correlation analyses of salinity and water temperature with MT mean values were not significant (P = 0.541 and P = 0.382, respectively).

 

Discussion

In a review of the available literature, a correlation between sediment metal concentrations and hepatic MT levels in fishes (Hylland et al., 1992) has not always been evident; however, this might be due to the effects of spawning migration, differences in contaminated diets or a combination of both factors. Seasonal differences in hepatic MT observed can be associated with an increased influx of Zn in relation to vitellogenesis. Previous works done with M. barbatus have shown that MT levels increase considerably in the prespawning period, mainly in female specimens (Benedicto et al., 2001). In our field experiment, where the fish were sampled during a non-sexual maturation period, a significant correlation was established between MT levels in red mullet and heavy metal content in sediments. At Portman, where the highest metal sediment concentrations were found, specimens collected showed the highest mean hepatic MT levels, while at Columbretes Islands and Cabo de Palos, where the lowest metal sediment levels were found, the mean MT levels had the lowest values for both females and males. Our finding that MT correlates best with Cu has also been observed in others studies (Hogstrand et al., 1991; Rotchell et al., 2001). Under chronic exposure to high metal levels, hepatic MT can increase four- to five-fold, although two- to three-fold increases are more common in most systems (Hogstrand and Haux, 1992; Hylland et al., 1992). In our study, the highest mean MT levels were about double that of the lowest mean, for both females and males.

There were no obvious indications of sex influence on the hepatic concentrations of MT in red mullet sampled in the postspawning period. This is in accordance with that observed by others authors in fish (Linde et al., 1999). In other studies, where sex influence on MT levels was identified, it was shown that specimens had been sampled during different periods of sexual maturation (Hylland et al., 1998) or during the prespawning period (Hylland et al., 1992).

There are some indications that temperature (Olsson et al., 1996, in Hylland et al., 1998) may affect MT levels in marine fish. An increase in MT levels with decreasing temperature has been observed in other fish species (Olsson et al., 1996, in Hylland et al., 1998; Rotchell et al., 2001). However, the results of our study did not show such a relationship between bottom water temperature and MT levels. The lowest MT levels were found in Cabo de Palos, where the temperature was lowest (14.88°C). In Castellón, where the highest temperature value was registered, MT levels were relatively high. Our study did not provide evidence for the influence of temperature and salinity on MT levels of M. barbatus.

One may ask whether the presence of some other kind of contamination is distorting the real effect of metal contamination on MT levels registered in the study areas. Several authors suggest that high sediment levels of pesticides and organic contaminants inhibit the uptake and accumulation of metals in fish (Brown et al., 1982; Cofino et al., 1992). The effects of such pollutants on metal uptake, or on fish MT levels, are not clear. Recent studies indicate that MT induction in fish should be considered a general stress response, particularly sensitive to heavy metals (Viarengo et al., 1999). Other authors have observed that differences in hepatic MT levels in Limanda limanda were not affected by the impact of organic compounds (George and Young, 1986; Ariyoshi et al., 1990). The possible influence of these factors on our results is being evaluated at present by determining organic compounds and pesticides in surface sediments and fish muscle tissues.

If we consider Cabo de Palos as a reference area from a metal contamination point of view, the background MT levels for M. barbatus collected in postspawning period would fit within the range of 90.96 ± 4.28 µg g^-1 liver. This level is in accordance with the value obtained for Platichthys flesus, 79 ± 17 µg g^-1 liver, from a reference site located in the Forth Estuary, eastern Scotland (Sulaiman et al., 1991). Besides, in a recent study (Lionnetto et al., 2001), hepatic MT levels in M. barbatus, from Italian clean marine environments, ranged from 85 to 29 µg g^-1 liver.

As mentioned above, previous studies have shown that the response of biomarkers may be affected by abiotic and biotic factors. A recent study with P. flesus demonstrated that for MT, 50-60% of total variability could be explained by factors directly related to season, gender and maturation state (Hylland et al., 1998). By removing the influence of age class, season and maturation state on the variation in MT, it becomes possible to demonstrate more clearly the relationship between MT levels and metals recorded in sediment (Rotchell et al., 2001). In this study, in which we have controlled all these factors, we can conclude that MT is a good biomarker of heavy metal contamination in M. barbatus, showing significant differences between areas located along a contamination gradient. Our results are the first to be reported for red mullet from the Iberian Mediterranean coast and they should help in further interpretations of MT levels and related biological effects, as well as to establish MT background levels for M. barbatus in the western Mediterranean.

 

Acknowledgements

We wish to thank the Secretariat General of Fishery of the Ministry of Agriculture, Fishery and Food for financial support of this research.

 

References

Ariyoshi, T., Shiiba, S., Hasegawa, H. and Arizono, K. (1990). Profile of metal-binding proteins and hemo-oxygenase in red carp treated with heavy metals, pesticides and surfactants. Bull. Environ. Contam. Toxicol., 44: 643-649.         [ Links ]

Benedicto, J., Martínez-Gómez C., Campillo, J.A., Martínez, F. and Marull, E. (2001). Influence of biotic factors on metallothionein levels in Mullus barbatus collected from the Iberian Mediterranean coast. Rapp. Comm. Int. Expl. Sci. Mer Médit., 36: 182.         [ Links ]

Brown, D.A., Jenkins, K.D., Perkins, E.M., Gosset, R.W. and Hershelman, G.P. (1982). Detoxification of metals and organic compounds in white croakers. In: W. Bascom (ed)., Coastal Water Research Project, Biennial Report. 1981-1982. Long Beach, California, pp. 157-172.         [ Links ]

Bucheli, T.D. and Fent, K. (1995). Induction of cytochrome P450 as a biomarker or environmental contamination in aquatic ecosystems. Crit. Rev. Environ. Sci. Technol., 25: 201-141.         [ Links ]

Burgeot, T., Bocquéné, G., Truquet, P., Le Deau, L. and Galgani, F. (1994). Induction of EROD activity in red mullet (Mullus barbatus) along the French Mediterranean coast. Sci. Total Environ., 142: 231-220.         [ Links ]

Burgeot, T., Bocquéné, G., Porte, C., Dimeet, I., Santella, R.M., García de la Parra, L.M., Pihol-Leshowicz, A., Raoux, C. and Galgani, F. (1996). Bioindicators of pollutant exposure in the northwest Mediterranean Sea. Mar. Ecol. Prog. Ser., 131: 125-141.         [ Links ]

Chang, K.M. (1995). Metallothionein: Potential biomarker for monitoring of heavy metal pollution in fish around Hong Kong. Mar. Pollut. Bull., 31: 4-12.         [ Links ]

Cofino, W.P., Smedes, F., de Jong, S.A., Abarnou, A., Whinnet, J.A., Schmiolt, D. and Wilson, S. (1992). The Chemistry Programme. Mar. Ecol. Prog. Ser., 91: 47-56.         [ Links ]

Ellman, G.L. (1959). Tissue sulfhydryl groups. Arch. Biochem. Biophys., 82: 70-77.         [ Links ]

Engel, D.W. (1988). The effect of biological variability on monitoring strategies: Metallothioneins as an example. Water Resour. Bull., 24(5): 981-987.         [ Links ]

George, S.G. and Young, P. (1986). The time course of effects of cadmium and 3-methylcholanthrene on activities of enzymes of xenobiotic metabolism and metallothionein levels in the plaice, Pleuronectesplatessa. Comp. Biochem. Physiol., 83C: 37-44.         [ Links ]

George, S.G. and Olsson, P.E. (1994). Metallothioneins as indicators of trace metal pollution. In: K.I.M. Kramer (ed.), Biomonitoring of Coastal Waters and Estuaries. CRC Press, Boca Raton, pp. 151-171.         [ Links ]

Goksoyr, A., Beyer, J., Egaas, E., Grsvik, B.E., Hylland, K., Sandvik, M. and Skaare, J.V. (1996). Biomarker responses in flounder (Platichtys flesus) and their use in pollution monitoring. Mar. Pollut. Bull., 33: 36-45.         [ Links ]

Hogstrand, C. and Haux, C. (1991). Binding and detoxification of heavy metals in lower vertebrates with reference to metallothioneins. Comp. Biochem. Physiol., 100C: 137-141.         [ Links ]

Hogstrand, C. and Haux, C. (1992). Metallothionein as an indicator of heavy-metal exposure in two subtropical fish species. J. Exp. Mar. Biol. Ecol., 139: 69-84.         [ Links ]

Hogstrand, C., Lithner, G. and Haux, C. (1991). The importance of metallothionein for the accumulation of copper, zinc and cadmium in environmentally exposed perch, Perca fluviatilis. Pharmacol. Toxicol., 68: 492-501.         [ Links ]

Hylland, K., Haux, C. and Hogstrand, C. (1992). Hepatic metallothionein and heavy metals in dab Limanda limanda from the German Bight. Mar. Ecol. Prog. Ser., 91: 89-96.         [ Links ]

Hylland, K., Sandvik, M., Skre, J., Beyer, J., Egaas, E. and Goksoyr, A. (1996). Biomarkers in flounder (Platichthys flesus): An evaluation of their use in pollution monitoring. Mar. Environ. Res., 42: 223-360.         [ Links ]

Hylland, K., Nissen-Lie, T., Christensen, P.G. and Sandivk, M. (1998). Natural modulation of hepatic metallothionein and cytochrome P4501A in flounder Platichtys flesus, L. Mar. Environ. Res., 46(1-5): 51-55.         [ Links ]

Linde, A.R., Sánchez-Galán, S., Kleiun, D., García-Vázquez, E. and Summer, H. (1999). Metallothionein and heavy metal in brown trout (Salmo trutta) and European eel (Anguilla anguilla): A comparative study. Ecotoxicol. Environ. Safety, 44: 168-173.         [ Links ]

Lionetto, M.G., Giordano, M.E., Cariacato, R., Pascariello, M.F., Maronosci, L. and Schettino, T. (2001). Biomonitoring of heavy metal contamination along the Salento coast (Italy) by metallothionein evaluation in Mytilus galloprovincialis and Mullus barbatus. Aquat. Conserv. Mar. Freshwat. Ecosyst., 11: 305-310.         [ Links ]

Livingstone, D.R. (1993). Biotechnology and pollution monitoring: Use of molecular biomarkers in the aquatic environment. J. Chem. Technol. Biotechnol., 57: 195-211.         [ Links ]

Mathieu, A., Lemaire, P., Carriére, S., Drai, P., Giudicelli, J. and Lafaurie, M. (1991). Seasonal and sex-linked variations in hepatic and extrahepatic biotransformation activities in striped mullet (Mullus barbatus). Ecotoxicol. Environ. Safety, 22: 45-47.         [ Links ]

Nielson, K.B., Atkin C.L. and Winge, D.R. (1985). Distinct metal-binding configurations in metallothionein. J. Biol. Chem., pp. 5342-5350.         [ Links ]

Rotchell, J.M., Clark, K.R., Newton, L.C. and Bird, D.J. (2001). Hepatic metallothionein as a biomarker for metal contamination. Age effects and seasonal variation in European flounders (Pleuronectes flessus) from the Severn Estuary and Bristol Channel. Mar. Environ. Res., 52: 151-171.         [ Links ]

Siegel, S. (1956). Non-parametric Statistics for the Behavioural Sciences. McGraw-Hill, New York.         [ Links ]

Snedecor, G.W. and Cochran, W.G. (1980). Statistical Methods. 7th ed. Iowa State Univ. Press, Ames.         [ Links ]

Sokal, R.R. and Rohlf, F.J. (1981). Biometry. W.H. Freeman, New York.         [ Links ]

Sulaiman, N., George, S. and Burke, M.D. (1991). Assessment of sublethal pollutant impact on flounders in an industrialised estuary using hepatic biochemical indices. Mar. Ecol. Prog. Ser., 63: 207-212.         [ Links ]

UNEP/RAMOGE, (1999). Manual on the biomarkers recommended for the MEDPOL Biomonitoring Programme.         [ Links ]

Viarengo, A., Ponzano, E., Dondero, F. and Fabbri, R. (1997). A simple spectrophotometric method for MT evaluation in marine organisms: An application to Mediterranean and Antarctic molluscs. Mar. Environ. Res., 44(1): 69-84.         [ Links ]

Viarengo, A., Burlando, B., Dondero, F., Marro, A. and Fabbri, R. (1999). Metallothionein as a tool in biomonitoring programmes. Biomarkers, 4(6): 455-466.         [ Links ]