You are on page 1of 15

Ecotoxicology, 13, 667–681, 2004 Ó 2004 Kluwer Academic Publishers. Manufactured in The Netherlands.

An In Situ Bioassay Integrating Individual and Biochemical Responses Using Small Fish Species
´ BRUNO BRANCO CASTRO,1,2 OLI´MPIA SOBRAL,1,2 LUCIA GUILHERMINO1,3 AND RUI RIBEIRO1,*
1

Departamento de Zoologia da Universidade de Coimbra, Instituto do Ambiente e Vida, Largo Marqueˆs de Pombal, 3004-517 Coimbra, Portugal 2 Centro Interdisciplinar de Investigaca˜o Marinha e Ambiental, Rua dos Bragas 289, ¸ 4050-123 Porto, Portugal 3 ICBAS-Instituto de Cieˆncias Biome´dicas de Abel Salazar, Universidade do Porto, Departamento de Estudos de Populaco˜es, Largo Abel Salazar 2, 4099-003 Porto, Portugal ¸
Accepted 3 December 2003

Abstract. The interest in the ecological relevance of risk assessments and, thus, in in situ bioassays has been increasing in the last years. The present study developed a time- and cost-effective in situ bioassay, aiming at obtaining, in a short period of time and with a minimum of resources, a set of ecologically relevant toxicological information in a site-specific approach. Poecilia reticulata and Gambusia holbrooki were chosen as test species. Post-exposure feeding inhibition and the biomarkers acetylcholinesterase, lactate dehydrogenase and glutathione S-transferases were the endpoints tested. The battery of biomarkers as a whole was sensitive to the in situ exposure in an acid mine drainage impacted effluent, although responses varied between test species. Post-exposure feeding inhibition was the most sensitive endpoint, and its association with biomarker responses was discussed. The linkage between individual responses, such as feeding, and biomarkers suggested that, at least in this case, biomarkers can be relevant at higher levels of biological organization. Altogether, the proposed short-term in situ bioassay seems to be a promising tool, since it represents a reasonable compromise between sensitivity, time/cost-effectiveness and ecological relevance. Keywords: in situ bioassay; feeding inhibition; enzymatic biomarkers; Poecilia reticulata; Gambusia holbrooki

Introduction In situ bioassays are useful tools in ecotoxicology, mainly because they integrate ecological relevance in toxicity testing, by incorporating field fluctua*To whom correspondence may be addressed: E-mail: rui.ribeiro@zoo.uc.pt.

tions (within semi-controlled conditions) in a cost-effective way. Fish have broadly been employed in in situ testing, and test chambers and general procedures have long been described (e.g., Wilde and Parrot, 1984; Jones and Sloan, 1989). Ideally, researchers should work with toxicity estimates derived from reproductive and demographic endpoints, especially in risk assessment.

Kosmala et al. Sturm et al. Taylor et al.. and may. ever since its introduction in mosquito-control programs (Holcı´ k. Therefore. 1992. which could work as an early warning tool. combining organismal (postexposure feeding) and biochemical (biomarkers) responses. as a result of being endpoints at a low level of biological organization (Peakall. 1992). 1998.nor time-effective. Fenet et al. 1996.g. poeciliids are indigenous inhabitants of tropical inland waters. 1996. Pacheco and Santos.. and there are substantial possibilities for using these methodologies in tropical countries. However.e.. namely in field caging experiments (Lindstrom-Seppa and Oikari. Fenet et al. Heath. 1999).668 Castro et al. Stien et al. This has been accomplished by using biochemical or other subcellular responses (i. Although small fish do not provide such large amounts of tissue for biomarker quantification. and glutathione S-transferases (GST). 1991. 1996. a preferable alternative is to perform short-term field exposures and. such as salmonids (Lindstrom¨ Seppa and Oikari... 1990. As) (Lopes . 1998) or eels (Van der Oost et al. such as growth or behavioral responses (swimming activity. An acid mine drainage impacted aquatic reservoir was the location selected for deploying the in situ bioassays. In general. 1995). It occupies a key ¸ intermediate position in the trophic chain of lakes.. as well as in biomonitoring surveys (e. both measuring biochemical markers in several fish species. P. ¨ ¨ Soimasuo et al. 1999). Goksøyr et al. Any eventual depression in feeding rate will have implications in the energetic input. Garcı´ a-Berthou. subsequently. Mn. Therefore. 1995. Feeding responses are usually regarded as behavioral endpoints.. Materials and methods Study sites The aquatic system surrounding Sao Domingos ˜ mine – an abandoned copper mine – in the SE of Portugal. The mine effluent is strongly acidic due to the continuous oxidation of abandoned mine tailings. 1998. and difficulties arise from a practical viewpoint. these studies were performed with large species. Al. 2000. 1999). biomarkers have questionable ecological relevance. where rapid and ecologically relevant toxicological information is most needed..). Such an in situ bioassay is neither cost. 1998. Pb. 1998). was the location chosen for the deployment of the in situ bioassays. The biomarkers chosen were acetylcholinesterase (AChE). 1999). and thus in the growth. although effects on such variables may extend beyond their behavioral outcome. Behavioral endpoints are commonly recognized as individually based responses resulting from the integration of a multiplicity of biochemical and/or physiological processes (Peakall. 1998. Mosquitofish (Gambusia holbrooki) and guppy (Poecilia reticulata).. Pacheco and Santos..g. Zn. in situ bioassays using biochemical markers should combine them with more comprehensive sublethal variables. Goksøyr et al. Co. 1996. 1998... as well as for usually having shorter life cycles. Comprehensive field work has been developed in this area. 2000). and contains large amounts of heavy metals (in decreasing order: Fe. feeding. Wall et al. less complex test-chambers). since it preys on insects and zooplankton. afterwards. lactate dehydrogenase (LDH)... Wall et al. both belonging to Poeciliidae. In spite of their rapid responsiveness and sensitivity to contaminant exposure. Cd. The aim of this study was to develop a sensitive and ecologically relevant short-term in situ bioassay with small fish species. prolonged field exposures are required to estimate sublethal/chronic toxicity. Garcı´ a-Berthou. Cr. 1996.. holbrooki is an abundant fish species in European freshwaters. 1998). Additionally.. 1990. 1998. G. reticulata is a standard test-organism. Ni. recommended by EEC (1992) and OECD (1982) guidelines for toxicity testing. etc. reproduction. 1995. or survival of the individual.. flatfish ¨ (Goksøyr et al. they are preferred for logistic reasons (e.. Fenet et al. 1998. as test endpoints. Cu. Van der Oost et al. while being eaten by piscivorous fish (some of commercial value) (Cabral et al. were used as test species in the in situ bioassay. Stephensen et al.. have adverse effects at the population/community level (Kumar and Chapman. 1998. in an effort to clarify the significance of the set of biomarkers at higher levels of organization. It was also the objective of this work to establish a relationship between biochemical and organismal responses. biomarkers) as test endpoints. to quantify relevant sublethal endpoints capable of responding in such a short period of time. Almaca..

Fish were all from the same strain with the purpose of reducing genetic variability.g. Germany) ad libitum. Schematic representation of the study area (Sao ˜ Domingos mine). two times a day (the amount of food added was consumed 2–3 min subsequently to its addition). 1999.5-l polyethylene terephthalate bottles (transparent). water chemical characteristics of REF were similar to those of contaminated sites.5 cm long) were separated from the rest and kept in aerated containers with filtered (150 lm) reference water. which are easily available and can be obtained at almost no cost. Adult males of appropriate size (2. reticulata). In situ bioassay chamber design and general protocol In situ bioassay chambers were made from 1. Ribeiro et al. and mammals (Ribeiro et al. Initially. it may possibly occur in future studies elsewhere. 1995. Only male guppies were used in the in situ bioassay. 1999. guppies were subjected to a 3-day acclimation period. located in a good water quality upstream lake (Fig. . pesticides. followed by gradual acclimation to filtered (150 lm) reference water.. 1999. was already expressed in other studies (Garcia et al.5 cm standard length. Furthermore.. it has been successfully used as a reference site in previous in situ bioassays (Pereira et al. 2000).0 mm). and thus increasing experimental precision. ecology and water chemistry of the system (Pereira et al.0–2. and a reference site (REF). After some dilution due to a small watercourse.. 2000). S2 and S1. reptiles. showing the location of reference and contaminated sites. Lopes et al.0–2. they were acclimated to field temperature fluctuations (25–30 °C). G. holbrooki were caught in the reference lake (REF) using hand nets (mesh size 1. until the deployment of the in situ bioassays (less than 24 h). there is extensive literature on the biology. industrial discharges or urban runoffs). In the improvised field laboratory. Pereira et al.. The source of contamination (pH and heavy metals) is isolated and well identified. since no other significant contamination sources are known (e. along the contamination gradient (S3. for almost 2 weeks.. to a temperature of 25 ± 2 °C and a photoperiod of 16 h L:8 h D. 1999). especially when dealing with biochemical responses. thus eradicating the chance of accidentally introducing an exotic species into Portuguese waters. 1995.. 1).. Although the sorption of organic contaminants was not a matter of concern in the present study.. from most to least contaminated). amphibians. 2.In Situ Bioassay With Small Fish 669 et al. 1999. were obtained from a local supplier and acclimated in the laboratory. such as enzymatic biomarkers. Guppies were kept in glass aquaria with ASTM hardwater and were fed a commercial flake food diet (TetraMenu-TetraWerke. Furthermore. 2000. birds. 1995). Four sites were chosen to perform the in situ assays: three in the mine effluent. Concern on this matter. prior to testing. 2000). Test organisms Male adult guppies (P. An in situ pre-exposure of the test-chambers in local water can surpass this Figure 1. This lake has no significant sources of contamination and it supports substantial communities of fish. the untreated effluent enters the Chanca River reservoir (part of ¸ the Guadiana River basin) near an ancient dam.. Apart from pH and metal concentrations.

and pH measurements were performed using a WTW 340-A pH meter. test-chambers and fish were brought to the field laboratory. 2). Short-term in situ bioassay with adult fish The 4-d in situ bioassay was performed at S1. ref. Two sets of two air-filled 1. and therefore a clear assessment of their condition. satisfied both these demands. 45% synthetic hydrocarbon. the mesh-size used in in situ bioassay chambers should be small enough to prevent test-organisms from escaping. gills. allowed proper water flow. A 1... No food was added to the chambers during the test. Physical– chemical parameters were measured and registered at the beginning and end of the in situ bioassay. chambers were held in vertical position (cap up) and withdrawn from the water except for the last 5 cm (approximately). Ideally. organisms were introduced through the top opening of the bottle using a funnel-like apparatus. local zooplankton).5-ml microtubes attached to the outside of the chamber with transparent adhesive tape acted as floaters (Fig. Taiwan. Portugal). For both species. S2.0-mm mesh. allowing water flow inside the test-chambers. shortcoming. in future studies. as the one used in the present study. head. TN122/WS. Two sets of test chambers were placed in each site as described above: 2 · 3 chambers per site for the guppy bioassay. and large enough to allow a proper water flow/renewal and aeration (Sibley et al. The top opening is potentially useful if. Organisms underwent a 24-h starvation period prior to the deployment of the in situ bioassay. guaranteeing the buoyancy of the bioassay chamber in an appropriate position. 2). . in local water. A min/max thermometer was used to monitor water temperature fluctuations during the assay. The mesh was sealed to the chamber with white thermal glue (supplied by Elis-Taiwan.5-l polyethylene terephthalate bottles. growth bioassays). which were previously placed near the water surface. feeding the fish becomes necessary (e. In situ bioassay chambers were made from transparent 1. Dissolved oxygen concentrations were determined with a WTW OXI 320 oxygen meter (Reagente 5). Organisms from the first set of chambers were killed by decapitation for biomarker quantification (see below). 2 · 4 chambers for the Gambusia bioassay. although it did not retain autochthonous food sources (e. S3 and REF with adult male P. Three openings (2 lateral and 1 at the base). Following the placement of the chambers in situ. A polyethylene terephthalate screw cap was then placed on the top opening. allowing a proper observation of the fish through the top opening. 1999). holbrooki.g.0-mm nylon mesh (Fig. Porto. Conductivity was measured with a WTW LF 330 conductivity meter (Reagente 5. After unscrewing the cap. and chambers were anchored to the shore with nylon rope. with a chemical composition of 50% ethylene–vinyl–acetate copolymer.. Two lateral rectangular openings and a circular opening at the bottom were made on the bottles and covered with 1. reticulata and G. for further processing. This enabled ‘‘trapping’’ the organisms in a confined volume of water. covered with 1-mm nylon mesh.670 Castro et al. 1999). At the end of the test.g. three fish per test chamber were used. with a SenTix electrode (Reagente 5). and dorsal muscles were immediately removed. and 5% polyethylene wax). which has been shown to be non-toxic to cladocerans (Pereira et al. The in situ bioassay started with the introduction of the organisms (3 per test chamber) in the chambers. This procedure was performed in ice-cold Figure 2..

fish were kept in the field laboratory in aerated containers with respective local water until the beginning of the feeding trials (between 2 and 8 h later). These experiments also showed that. respectively.. in average. this time gap should be reduced to a minimum and all treatments should be given an identical waiting time. until the enzymatic assay. and that a small number of particles would be preferable for improved visualization. A Labsystem Multiskan EX microplate reader (supplied by Paralab. frozen samples were thawed and homogenized using an Ystral GmbH Dottingen homogeniser (Reagente 5). a U is a nmol (for AChE and GST) or lmol (for LDH) of substrate hydrolyzed per minute. Therefore. which is an adaptation to microplate of the spectrophotometric method described by Vassault (1983). Subsequently to the in situ exposure. but in situ exposures pose too many technical difficulties. Post-exposure feeding trials Ideally. Ideally. which could lead them to split the food flakes in two or more pieces in their feeding strikes. Nevertheless. at 25 °C. like guppies or mosquitofish. a 3-min test period seemed adequate to monitor changes in the fish feeding behavior between sites. where high mortalities resulted in an insufficient number of organisms to perform the test. All enzymatic assays were performed in triplicate. (1996). . Biomarker quantification In the laboratory. dorsal muscle in Tris– NaCl buffer (0. through direct individual observation. and activity was expressed as Units (U) per mg of protein. The time elapsed between the beginning of the trial (addition of food) and each feeding strike (i. about 3 min per fish. pH ¼ 7. The activity of LDH was determined according to Diamantino et al. the 3-min feeding trial was started with the addition of 12 particles of TetraMenu. pH ¼ 7. Small fish. GST activity was determined by the technique described by Habig et al.1 M. After a 5-min period for the fish to become calm. although Gambusia were not so voracious.1 M. adapted to microplate. as described in Guilhermino et al. feeding movement where ingestion actually occurred) was registered. pH ¼ 6. Porto. head in phosphate buffer (0. would have difficulties ingesting particles larger than 1 mm. (2001). The feeding trial was stopped when the 3-min period expired or when the fish completed 12 successful strikes. between 500 and 1000 lm (obtained using a simple sieving procedure). Samples were always kept on ice. and the individual fish feeding behavior was monitored.2) for lactate dehydrogenase determinations. (1961) adapted to microplate. Fish were then individually transferred to polyethylene terephthalate vessels with 20 cm of diameter (1 organism per vessel) containing 330 ml of filtered (150 lm) reference water. under these conditions.2) for acetylcholinesterase analysis.2) and took. and a 90-min feeding trial. Fish from the second set of chambers were used in post-exposure feeding trials (see below). while muscle and head homogenates were centrifuged at 5000 · g for 3 min. The activity of AChE was determined by the method of Ellman et al. pH ¼ 7.9 mg/ ml for head and muscle samples.1 M. Supernatants were then normalized to approximate protein contents of 0. Portugal) was used. A similar feeding trial was described by Little and DeLonay (1996) involving video recordings. Samples were minced using a scissors and frozen at )20 °C (head and muscles) or in liquid nitrogen (gills). After removal of blood vestiges. each tissue was placed in the appropriate buffer for each biomarker: gills in phosphate buffer (0. (1974). Preliminary experimentation had shown that healthy fish would eagerly eat the food particles. Gill homogenates were centrifuged at 9000 · g for 30 min (at 4 °C). using bovine c-globulin as standard. Food flakes were evenly spread over the surface of the test water. particles should be large enough to allow a proper observation. and thus the choice of a minimum flake size (500 lm). healthy guppies would eat all the particles in less than 1 min. Since some sort of feeding impairment was expected on the contaminated mine effluent. except for S3. thus misleading the observer.1 M. The concentration of protein was determined in triplicate by the Bradford (1976) method adapted to microplate. feeding impairment should be assessed during the test. as well as throughout all subsequent processes. two types of feeding trials were performed following the in situ exposure: a 3-min continuous-observation assay.e.In Situ Bioassay With Small Fish 671 phosphate buffer (0.5) for glutathione S-transferases determinations.3 and 0.

High temperatures were registered in all sites. where it decreased to 3.2–8. After removal of the fish. faecal pellets were removed with a plastic pipette. LDH and GST) were separately analyzed for differences between sites using one-way analysis of variance (ANOVA).020) in S1. Subsequently to the continuous-observation assay. reticulata (Fig.672 Castro et al. mortality was significantly higher than REF only in S3 (the most contaminated site). In comparison to the REF station.2–7. p ¼ 0. mortality was only registered in S2 and S3. usually employed in survival-time analysis.471). Nevertheless. reticulata was observed along the contamination gradient.5–5. holbrooki. the surplus of ethanol was removed with a pipette.2 3. p ¼ 0. for calculation purposes). Data resulting from the continuous-observation feeding trial were analyzed using non-parametric techniques (Log-rank and Wilcoxon statistics). since fish were from a different set of test-chambers. holbrooki. Range of physical and chemical parameters registered in the field during the in situ exposures Site REF S1 S2 S3 Temperature (°C) 18–34 18–34 18–34 18–34 D. Prior to this procedure. and a progressive acidification and increased conductivity were observed in the most contaminated sites (S2 and S3.0 pH 7. through gaps formed in the mesh-bottle linkage (Table 2). The worst case happened in REF with G. Results Short-term in situ bioassay with adult fish The range of physical and chemical parameters registered during the in situ bioassays are presented in Table 1.06 ± 0. if present in the bottom of the test vessel.87. high mortalities in S3 resulted in an insufficient number of organisms to perform the feeding trials in fish from this particular site. Statistical analysis Mortality was tested for significance using the Fisher exact test.95. and GST was not significantly affected (ANOVA: F(3. where four organisms escaped. at a temperature of 25 ± 2 °C. and food particles were transferred to pre-weighed foil cups and dried for 48 h at 60 °C.84 (average ± SD. As stated above. reticulata AChE activity was inhibited (ANOVA: F(3. thus gathering the uneaten food particles.24) ¼ 3. P.5-ml microtubes. (mg/l) 8. In G. In both cases. where 70.27. during the 90-min period. Food intake (after arcsine transformation) was also tested for significance with oneway ANOVA. Correlation analysis was used to analyze the relationships between food intake (after arcsine transformation) and biomarkers.3 7. holbrooki (Fig. 4). Preliminary experimentation indicated that. except in S3.0–9. p ¼ 0. holbrooki adults. a pre-quantified amount – 12. With G. a Tukey multiple comparison test followed the ANOVA. see Table 1).O. In the laboratory. Food particles were then carefully collected from the mesh and stored in 70% ethanol (to avoid putrefaction) in 1. Biomarkers response was considerably different between P.26) ¼ 0. Some fishes escaped from the test chambers. Food intake was expressed as the proportion of food consumed relatively to the initial amount of food (12 mg.5 3.7 mg/l. Porto.2 Cond.7–5. Dry weights of the uneaten food were then obtained on a METTLER UMT2 microbalance (Soquimica. n ¼ 5) mg/organism – of 500– 1000 lm food particles was added.7–5. Biomarker data (AChE.3–6. Portugal) to the nearest microgram. thus setting off the long-term (90 min) feeding trial. which allow the analysis of censored data. the test water was filtered through a 200-lm nylon mesh. 3) and G.4 7.26) ¼ 7.8 6. and in both cases it was significantly different from REF. All feeding trials were performed in similar conditions. These analyses were performed with the mean values of the biomarkers. only LDH activity was Table 1.4–9. when applicable. A mortality increase (Table 2) of P.6% of the organisms died. LDH activity was significantly stimulated (ANOVA: F(3. Dissolved oxygen (DO) was more or less constant during the in situ exposures.9 4. healthy fish would consume a major percentage of this quantity of food.001) in the most contaminated sites (S2 and S3). (lS/cm) 185–185 260–266 350–458 365–558 .

012).4 0. Food intake (the proportion of food consumed per mg of initial 0. and S3 – sites along the contamination gradient.In Situ Bioassay With Small Fish 673 Table 2.15) ¼ 16. The 90-min post-exposure feeding trial was also successful only in P. reticulata but not with G.2 0. Significant differences were not found between sites.16. Mortality and recovery of adult fish at the end of the short-term in situ bioassay Species P. for each biomarker (pO0:05). p ¼ 1. The 3-min continuous observation assay monitored the feeding activity of the fish.23. LDH activity (U/mg prot) significantly affected (ANOVA: F(3. No feeding trials were performed with fish from S3. S1 and S2 feeding curves were registered (Fig.6 11. AChE activity (U/mg prot) 120 100 80 60 40 20 0 0.1 25. 5a).3 Post-exposure feeding trials Post-exposure feeding trials were successfully conducted with P. S2.339). since the latter did not eat any TetraMenu particles during the feeding trials. Ndead – number of dead fish at the end of the bioassay. Different letters (a.4* 71. while fish from S1 did not eat all the particles within the 3-min interval. either for AChE (ANOVA: F(3. a.05) between REF. LDH. as revealed by the significant differences registered between sites (ANOVA: F(2.35) ¼ 2. reticulata after the 96-h in situ exposure. S1. holbrooki. reticulata.6 a.6** 0 0 30. relatively to the reference (REF) site (Fig. Error bars represent standard deviation. REF – reference site. ***pO0:001). Fish exposed in REF ate all the food particles quickly (<80 s).35) ¼ 4. holbrooki a 140 a. Significant differences (log-rank and Wilcoxon statistics: p  0. All other fish from S2 were hypoactive.1 0.0 180 GST activity (U/mg prot) 160 140 120 100 80 60 40 20 0 REF S1 S2 S3 Sites Figure 3. Ninitial – number of fish at day 0.093) or for GST (ANOVA: F(3. Organisms from S1 occasionally refused food particles by spitting them out (not accounted as successful feeding strikes). **pO0:01. with only one fish demonstrating interest in the food particles.5 a 0. p ¼ 0. Enzymatic activities of AChE. and showed no curiosity towards the food. reticulata Site REF S1 S2 S3 REF S1 S2 S3 Ninitial Nrecov Ndead 18 18 18 18 24 24 24 24 18 18 16 17 20 24 23 21 1 2 4 12 0 0 7 15 Mortality (%) 5.45 · 10)4). b) represent statistically significant differences between sites. Nrecov – number of fish retrieved at the end of the test (including dead ones).9. 5a.0 70. Almost no feeding response was recorded in fish from S2.31.b . 5b). eating one at 41 s and refusing two others. with a clear inhibitory effect being observed. and GST from P.b b 0. because of high mortalities in that site. p ¼ 0.b b b Fisher exact test: significantly different from REF (*pO0:05. G.4*** food supplied) was impaired in the contaminated sites (S1 and S2). as illustrated by Fig. p ¼ 0.32) ¼ 1.

REF – reference site. S2.6 REF S1 S2 Time (s) Food intake (mg dry food consumed/mg initial food) 0. p ¼ 0.5 0.0 a a.9 0. Figure 5.1 0. b. b) represent statistically significant differences between sites. c) represent statistically significant differences between feeding curves (Log-rank and Wilcoxon statistics: pO0:05). p< 0. c) represent statistically significant differences in food intake.7 0.5 LDH activity (U/mg prot) 1. Individual observations (triangles) and means (circles) are represented for each site.4 0. and GST and food intake (r ¼ 0.3 0. (b) Food intake (mg dry food consumed/mg of initial food) in P. LDH and food intake (r ¼ )0. Different letters (a. Short-term in situ bioassay with adult fish The in situ bioassay chambers allowed successful exposure and retrieval of the fish. for each treatment.2 0. holbrooki after the 96-h in situ exposure. for each biomarker (pO0:05). and GST from G. Feeding versus biomarkers Discussion Since no data on feeding was available for G. (a) Feeding activity of P.3 0. p < 0. Numbers in brackets stand for the number of organisms used in each of the treatments. although a small . holbrooki. only P. REF – reference site.2 0.6 0. between sites (pO0:05). S1 and S2.8 0.674 Castro et al.0 0.0 Averages Individual observations Number of observations (6) b (5) c (7) 180 GST activity (U/mg prot) 160 140 120 100 80 60 40 20 0 REF S1 S2 S3 REF S1 S2 Site Sites Figure 4. reticulata results were used in this analysis.001). Error bars represent standard deviation. LDH. Different letters (a. Number of ingested food particles (cumulative) 80 70 60 50 40 30 20 10 0 c (5) 0 20 40 60 80 100 120 140 160 180 b (7) 140 (a) AChE activity (U/mg prot) a (7) 120 100 80 60 40 20 0 0. A significant correlation was found between AChE and food intake (r ¼ 0. S1 and S2 – sites along the contamination gradient.b b (b) a (n) 0. Each set of points represents the cumulative number of food particles ingested by the organisms during the 3-min feeding trial. S1. Different letters (a.831.654.001).003).4 0. Error bars represent standard deviations. b.b a. Enzymatic activities of AChE.1 0. reticulata after in situ exposure in REF. and S3 – sites along the contamination gradient.823. reticulata quantified through a post-exposure feeding trial.

This subject should be further exploited in future studies. Fenet et al. it is apparent that low pH and copper (this may be true for other metals) induce a metabolic shift at the muscular level.. 1998. 2000). 2000). promoting anaerobic metabolism. unanswered. and its activity can be used to assess the metabolic state of the organisms.. 1998. respectively. The two opposite response patterns of LDH activity between the two species tested remain. leading to mortality of the fish.. 1987. a clear increase in mortality was observed along the contamination gradient in both species. McLoughlin et al. Garcia et al. The results from the in situ bioassay suggested that mosquitofish AChE was insensitive to metals and to low pH.. including PCBs and PAHs. Garcia et al. it is still successfully employed as a specific biomarker of exposure to organophosphorous and carbamate pesticides (Beyers and Sikoski. (2000) described this mechanism in brown trout exposed to copper at low pH. Gill et al.In Situ Bioassay With Small Fish 675 minority escaped. which are detoxified via glutathione conjugation (Phase II detoxification mechanism) (Habig et al. 1987. 1992. (1991). This result suggests an adverse effect for guppy on this key enzyme. Beaumont et al. and that ionic depletion was further enhanced by Al (in combination with low pH). Nevertheless. Although several types of contaminants can affect AChE (Guilhermino et al. Heath.. The increase observed in the muscular LDH activity of P. in both guppy and mosquitofish. LDH stimulation.. but resulted instead from a general stress response in combination with an electrophysiological disruption. such as feeding impairment and other behavioral alterations (Lacroix and Townsend. however. (1991.. metals and acidity also produce numerous sublethal effects... after observing high lactate concentration and glycogen depletion in the red and white muscle. since both inhibition (as seen in the mosquitofish) and stimulation (as seen in the guppy) of LDH activity can be explained by the existing literature. 2000). 1991) did not significantly affect LDH activity from skeletal muscle of rosy barb. reticulata may result from this metabolic shift. GST are commonly used biomarkers for hydrophobic organic chemicals. both copper (Gill et al. although its response to metals has revealed to vary with the tissue (Gill et al. These authors also suggested that this phenomenon was not due to the limitation of the oxygen supply to the muscles. 1998. but in vivo responses of AChE seem to vary. 1994. Whether it is a general stress response or a more complex mechanism. Blaxter and Hallers-Tjabbes. 1995. although the exact role of ammonia was not fully understood. 1992). may be a consequence of a local hypoxia ‘‘syndrome’’. McLoughlin et al. Beaumont et al. as shown by Diamantino et al.. and consequent increase in the number of deaths. on the other hand. which was consistent with tissue hypoxia (Heath. In the present study. 1995). Peakall. Nevertheless. 1992) and cadmium (Gill et al. 1991. while guppy AChE was inhibited in the most contaminated sites.. Heath.. 1996. Metals usually inhibit AChE in vitro (Gill et al. resulting from an ammoniamediated toxicity mechanism (may be a general stress response). Stien et al. Lacroix and Townsend (1987) have shown that pH alone can be responsible for severe ionic imbalance. 1992) have shown that in vivo AChE activity can be unaffected. with harmful consequences in the central nervous system and neuromuscular function.. LDH is a glycolytic enzyme. which is not the present case. Both metals and low pH are known to affect ion regulation (Lacroix and Townsend. 1995). 1974. 1995). Biomarker alterations are also among the sublethal outcomes caused by metals. The underlying physiological mechanism for in vivo AChE stimulation or inhibition by metals is still unclear. The inhibition of LDH activity could be due to the direct effect of metals in the enzymatic activity. low oxygen levels might have worked as a confounding factor in this variable. stimulated or inhibited in different organs of rosy barb. 2000. 1991. It is considered a nonspecific indicator of cell integrity when quantified in plasma samples. 1992. LDH muscle activity responded differently in tested species. 2000). 2000). Van der Oost et al.. In S3. as reviewed by Heath (1995). at the muscular level. 1992. . by copper and cadmium. Besides mortality. (2001) and Gill et al... Hyperammonaemia appeared to be related with all of these metabolic changes. 1998. Low pH and heavy metals in the mine effluent caused significant mortality in the most contaminated stations. Sturm et al. LDH has been shown to be related with growth (Pelletier et al. Low mortalities were to be expected but the strong evaporation led to a progressive acidification. Gerhardt.

In the present study. however... 1995. 1998). or a less sensitive response than other related biomarkers (Huuskonen et al. although. strange) food source in the feeding trial. Stien et al. ParisPalacios et al. as a whole. 2000) or in field-caging studies (Soimasuo et al. 1998). which analyzed the number of feeding strikes (Buckler et al. no significant effects were recognizable in GST activity. Little and . Diamantino et al. it is fundamental that the use of native species is encouraged. Stephensen et al.. lindane. and permethrin).. Stien et al. 1995. 1999. Although metals are not natural substrata for these enzymes. which are responsible for the induction of these enzymes (Canesi et al. that these biomarkers will respond to several other types of contaminants. a large set of biochemical endpoints should be used to properly check-up the normal physiological condition and. 1999. mosquitofish and guppy were used due to their availability. For locally collected fish.. (2000) have used ChE and GST for evaluating sublethal effects of several toxic chemicals (zinc.. LAS. namely in vivo. However.. Similar behavioral assays were proposed by other authors. 1995. 2000)... 1996. the battery of biomarkers chosen revealed to be sensitive to the effluent exposure. 1996). Nevertheless. Here.. 2001.. such as indigenous zooplankton. 2000). 1998). Based solely on the results. recognizing the utility of biochemical markers apart from their more popular use as contaminant-specific indicators.. Little and DeLonay (1996) recommend such a procedure when it is intended to use a novel (i. in in situ monitoring of effluents. although a decrease in their activity was apparent... Post-exposure feeding trials The post-exposure feeding trials were not successful with G. and in a time-effective manner. 1991. 2000) and mussels (Canesi et al. it would be sound to recommend guppy. both general and highly specific biomarkers can be integrated in the bioassay. McLoughlin et al. Paris-Palacios et al. 1998. as a preferable test-species. but virtually any small fish species has potential to be included in the bioassay. which were fed with this same type of food during laboratory rearing. Peakall (1992) first suggested this approach. GST for detoxification/oxidative stress.... 2000. reticulata were successfully conducted using TetraMenu.g. GST have successfully been included in biomonitoring studies (e. Their results showed that these biomarkers were more sensitive than mortality and may work in a complementary way. have shown that metals can also inhibit GST.676 Castro et al. GST are among the antioxidant defenses of vertebrates and invertebrates.. Ribeiro et al. post-exposure feeding trials with P. and LDH as an indicator of the metabolic condition of the organisms. it could be preferable to use a different food source. pirimiphos-methyl. Ideally. a simple battery can also provide a sensitive response. not mosquitofish..e. Thus. Garcia et al. for example. especially if further complemented with other measures (e. The set of biomarkers used in the present study were chosen as indicators of key physiological functions: AChE for neurotoxicity.. 2000) and from the work of McLoughlin (2000). In general terms. recent studies have shown in vivo GST induction in fish (Paris-Palacios et al. Serafini and Romeu (1991). holbrooki. However. at the same time. after the in situ exposure.. 1996.. maybe because fish caught in the field may require a pre-acclimation to the food used. The 3-min feeding trial aimed at analyzing the feeding behavior of the fish. it was more sensitive than mortality. as seen here. The battery here proposed did not intend to be specific for acid mine drainage contamination and there is evidence. 1996. feeding inhibition). Van der Oost et al. 2000. 1999) exposed to heavy metals. Other authors have shown that a more or less diversified battery of biomarkers can be successfully used. Fenet et al. and. and metals are known promoters of oxidative stress (Heath. from our laboratories research (Guilhermino et al. It is therefore possible that GST metabolize oxidative stress-originated byproducts. Other authors have also found a lack of response of fish GST to contaminants (Jensen et al. 1994. to identify the causal nature of the physiological disturbance due to the comprehensive contaminant specificity of the selected biomarkers. Although their practical application may thus be questionable. whether it is in biomonitoring (Van der Oost et al.g. although this might require additional optimization of methodologies (the use of live local food is recommended – see below).

2001). also hypothesized on the mechanism leading to feeding impairment.. LAS. Beauvais et al. the most sensitive parameter measured. but also a reduction in the metabolic efficiency can occur (Waiwood and Beamish. Feeding inhibition revealed to be. 1978. a combination of several or all of these factors. in the present study. and a reduction in energy input will affect growth and reproduction. and permethrin). (1998) questioned the possible consequences in the fish and phytoplankton communities of feeding inhibition observed in populations of a primary consumer (Daphnia). In rainbow trout. 1978. Feeding activity in this species has also shown to be an extremely sensitive response to pH alone and to aluminium (Buckler et al. Brewer et al. (2000). including metals and low pH. in S2. AChE activity. and could also affect the feeding behavior of the fish (Dell’Omo et al.. GST activity. and several authors have focused their attention on this issue. The most probable cause of feeding inhibition is. feeding inhibition). The energy status of the organisms is also a main factor when analyzing reproductive endpoints. GST and LDH) with ecologically relevant individual endpoints (in this case. Feeding versus biomarkers The approach proposed in the present study was to attempt linking subcellular responses (AChE. This approach is not new. Ultimately. Several contaminants. feeding impairment was observed where no AChE inhibition occurred (site S1). where mortality. in REF. Taylor et al. in this study. Feeding inhibition was almost always the most sensitive endpoint and it was affected by all of the contaminants tested.In Situ Bioassay With Small Fish 677 DeLonay. 2000. 1978). as the first step to assess the ecological relevance of biomarkers. and no feeding. as seen by the results of the 90-min feeding trial. as a plausible consequence of feeding inhibition observed in key benthic detritivores.. Blaxter and Haller-Tjabbes. food consumption and gross conversion efficiency were significantly impaired by copper and low pH (Waiwood and Beamish. In both cases. 1997. The inhibition of AChE is responsible for an impairment of neuronal and neuromuscular function. in all of the feeding trials. Similar feeding trials were successfully conducted by Waiwood and Beamish (1978) and Kumar and Chapman (1998). Some suggested that damages to sensory organs and receptors induced by metals and low pH cause loss or impairment of sensorial perception of the fish (namely olfaction) (Waiwood and Beamish. after zinc exposure.. although . Kumar and Chapman. Copper. 1997. feeding was significantly impaired. food intake can be considered to be a rough measure of energy input. Maltby and Naylor (1990) and Forrow and Maltby (2000) addressed the problem of reduced efficiency of detritus processing and nutrient cycling. Fish were increasingly hypoactive along the effluent sites. not only the feeding rate of the fish can be affected by heavy metals. however. Jensen et al. 1998). affect the feeding of fish. pirimiphos-methyl. such as Gammarus. Lacroix and Townsend (1987) have shown that pH alone can depress feeding responses in juvenile Atlantic salmon. lindane.. revealing decreasing interest in the food particles. and feeding rate were compared in terms of their sensitivity to five different compounds (zinc. 1995). sluggish feeding. 1992). However. in an in situ experiment. Thus. respectively. Impairment of feeding rate can also have serious consequences at the community level. Kumar and Chapman (1998) have shown evidences of growth impairment and reduction in lipid content associated with reduced feeding rate in rainbow fish exposed to an organophosphate (profenofos). and was subjectively qualified as active feeding. and more work is required to clarify this issue. Other authors that have observed feeding inhibition provoked by metals. for copper and profenofos. Feeding inhibition of Gammarus pulex was also the most sensitive endpoint tested in a study by McLoughlin et al. in S1. 1978). The underlying mechanism of this decrease in the fish interest to feed is unclear. Taylor et al. (1998) suggested gut poisoning as the toxicity mechanism of cadmium in feeding-impaired Daphnia. The observed reduction in the feeding activity had consequences in the amount of food consumed. 1996). Guppy feeding behavior was impaired in the contaminated stations. Maltby and Naylor (1990) showed a correspondence between feeding rate depression and long-term reproductive impairment in Gammarus pulex.and pH-induced growth reduction has been related to feeding inhibition in rainbow trout (Waiwood and Beamish.

which is especially useful to scientists in developing countries.. De Coen and Janssen (2003b) have proposed a successful quantitative approach for predicting population level effects from short-term biomarker endpoints in Daphnia. a battery of biomarkers is strongly preferred. 2000. and population-level responses. based on some sort of underlying physiological mechanism. previous studies have only exploited individual biomarkers. namely because: (i) in situ exposures reduce the uncertainty related with sample manipulation. alkalinity. pesticides. especially for AChE and GST. ecologically based changes result from the integrated alterations in growth. As previously stated. In a previous study by the same authors (De Coen and Janssen. Since biomarkers will respond differently to different contaminants. especially while biochemical endpoints are not fully linked with alterations at ecologically relevant levels of biological organization. where this type of information is most needed. Pelletier et al. significant correlations between biomarkers and food intake were observed. (1995) have shown a strong positive relationship between glycolytic enzymes (including LDH) and growth rate. such as salmonids. while. although future comparisons with population-level responses are needed. 1997. but it is clear that these three biomarkers alone were able to detect physiological alterations due to low pH and high heavy metals concentration and were relevant at higher levels of organization. Recently. . although their role as early warning tools is recognized. they were able to establish a relevant connection between cellular energy allocation (CEA) and population level responses in Daphnia. as well as which pose more drastic consequences to the test organisms. such as population or community. contaminant-specific biomarkers can elucidate which classes of contaminants are present. depending on the purpose of the study: behavioral changes and general physiological biomarkers can be introduced in the assay in order to understand the mode of action of the pollutant(s). but this was feeble for LDH. cellular.678 Castro et al. 2001). subcellular. 1997. etc. etc. At least for AChE. when linking biomarkers with endpoints at higher levels of biological organization. Amidst all of these integrative phenomena. Conclusion The aim of this work was to develop an in situ bioassay with small fish integrating feeding and biomarkers. Laboratory research must also continue. the set of biomarkers used can be broadened. There is novelty in this approach. Brewer et al. also requires an extrapolation basis to higher-level parameters in order to be regularly used in environmental effects assessment. it is difficult to comprehend what is the ecological significance behind subcellular or molecular alterations. especially to build a robust database on the effects of contaminants (metals. Nevertheless. feeding and reproduction of the organisms. but did not study the response of this association to toxicants. hydrocarbons. Many other parameters can be assessed at the end of the assay. which. 2003a). Future research is therefore required to further establish a reliable connection between early warning responses.. such as biomarkers and other individual or subindividual parameters (including feeding). we emphasize the role of this assay as a short-term and cost-effective early warning tool. The present study presented a first step in that direction. due to its behavioral nature. (ii) caging (or in situ) experiments are usually performed with large species. (iii) endpoints at two levels of organization are included in the assay.) on the selected endpoints. Plus. In the present study. The connection between early warning signals (subindividual responses) and ecologically relevant parameters (population and community effects) is vital in order to enable the use of biomarkers in environmental effects assessment.. This parameter is most likely related to feeding. The proposed assay should be used as an early warning tool. allowing a much more realistic scenario of exposure to contaminants. 1995).) and water chemistry (pH. Beauvais et al.. It is difficult to understand the significance of biomarkers at higher levels of biological organization level. Toxicant-induced changes at the individual level are an outcome of the integrated molecular. at the same time. and physiological alterations (Heath. Jensen et al. there could be an underlying mechanistic linkage with feeding behavior. such as AChE inhibition and behavioral changes (Dell’Omo et al.

. A. C. A. Almeida. L. W. L. Survival. S. Physiological measures of neurotoxicity of diazon and malathion to larval rainbow trout (Oncorhynchus mykiss) and their correlation with behavioral measures. Beauvais. Environmental and biological factors influence the relationship between a predator fish. (1999). Analytical Biochemistry 72. Diamantino. H. (1997).M. (1961). to a sub-lethal concentration of copper in soft acidic water: effects upon muscle metabolism and membrane potential..V. G. Exposure of brown trout.F. Moreira for assistance during fieldwork and two anonymous reviewers for significantly improving the relevance of the manuscript.C. Chem. 1632–1641. and Brumbaugh.Sc. Chem.L. Andres. Brewer. and Shore. 43–58.J. A. the in situ approach makes this a site-specific tool for ecotoxicological studies.. EEC. Ellman. Chem. Future work should further attempt to elucidate the link between biomarkers and effects at the population or community levels. C. K. and Featherstone. 19. S. Little. sublethal responses. Hydrobiologia 382. Netherlands J. 40. Toxicol.W. De Coen.L. P. References Almaca.. Toxicity of sodium molybdate and sodium dichromate to Daphnia magna Straus. 40.. W.J. ref. and Bontoux.M. C. Beaumont. In this study. S. Beauvais. Environ. 203-216.. L. G. 22. Viarengo. Saf. Filippelli.R.M. J. and tissue residues of Atlantic salmon exposed to acidic pH and aluminium. J. and Little. Fenet. (1998).V. Toxicol. Chem. Directive 92/32/EEC. Soares.. S. Butler. 31. and Ellersieck. Dell’Omo. 553–560. Castro was a recipient of a M. (1976). C. Behavioral dysfunctions correlate to altered physiology in rainbow trout (Oncorhynchus mykiss) exposed to cholinesterase-inhibiting chemicals. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Beyers. 935–939. Aquat. D.E.. and Guilhermino. (2001).M. C.R. ref. and its main prey in rice fields of the lower Mondego River valley (Portugal)... Biochem. 26.S. (2000). Ecol. G. Heavy metals and glutathione metabolism in mussel tissues. 51. R.. since it identified a contamination gradient at an impacted reservoir. Bryenton. J. E. and Maltby. (2003a). Effects of exposure to an organophosphate pesticide on behavior and acetylcholinesterase activity in the common shrew Sorex araneus. 88–95. E.In Situ Bioassay With Small Fish 679 The results here presented have shown that the 96h in situ bioassay is sensitive and cost/time-effective. E. 13. and Hallers-Tjabbes. R. M. Buckler.L. 253–259. Environ. Environ. Ecotoxicol.K. A multivariate biomarker-based model predicting population-level responses of Daphnia magna. Ecotoxicol. E.C.M. 16. Toward a mechanistic understanding of contaminant-induced changes in detritus processing in streams: direct and indirect effects on detritivore feeding. Toxicol.W. Brewer. Seventh amendment of directive 67/548/EEC. Diamantino. A.E. 45.. 46. Toxicol. 2100–2106. Environ.J. and Taylor. Fish Species and Varieties Introduced into ¸ Portuguese Inland Waters. 259–72. This work was partially funded by Fundacao para a Ciencia e a ¸ ˜ ˆ Tecnologia (project SENSOR. C. (2003b). .M. Acknowledgements B. T. Toxicol. allowing the rapid attainment of ecologically relevant information.L. Aquat. Little.H.. Aquat. (1994).. it was also shown that biomarker responses were related to feeding inhibition.T. and Soares.B. Jones..C. Blaxter.B. Acute toxicity to fish. Furthermore. and Sikoski.. 67–76. Laboratory and field–caging studies on hepatic enzymatic activities in European eel and rainbow trout.. V. D. 137–143. Acetylcholinesterase inhibition in federally endangered Colorado squawfish exposed to carbaryl and malathion. (1995). Arch.D.. Chem. Environ. (2000). Toxicol.W. 41–51. 70–76. Mieiro. Toxicol. (2000). De Coen. EEC (1992). Courtney. S. M.. (1995). Jones. Toxicol.G. M. M. Environ. D. Gallo. L. Pharmacol. Aquat.M. The missing biomarker link: relationships between effects on the cellular energy allocation biomarker of toxicant-stressed Daphnia magna and corresponding population characteristics... Chemosphere 45. Salmo trutta. (2000).A. W. Cabral. Janssen. Canesi. grant from Fundacao para a Ciencia e a Tecnologia ¸ ˜ ˆ (Portugal) – PRAXIS XXI.M. PRAXIS/ PCNA/C/BIA/0157/96) and by the European Commission (project TROCA-WET. (1992). Almeida. 19.R. A new and rapid colorimetric determination of acetylcholinesterase activity. Chem.B. This is a first step to support the idea that a battery of biomarkers may be an indicator of biological effects at higher levels of organization. C. Lisboa: Museu Nacional de ´ Historia Natural. (1998). Environ. Cleveland.. Lactate dehydrogenase activity as an effect criterion in toxicity tests with Daphnia magna Straus. Environ. 272–276. Leonzio.. Guilhermino. 248–254. (2001).C. Authors wish to thank S. E. R. Contam. 1875–1880. Annex V – Part C: Methods for the determination of ecotoxicity. L. Toxicol. ERBIC18-CT98-0264). 7. 22. Janssen. Gambusia holbrooki. Saf. and Marques.K. E. Casellas. S. Environ.R. J. T.. Toxicol. The effect of pollutants on sensory systems and behavior of aquatic animals. DeLonay.E. Bradford. 2195–2201. Forrow. P.

Hylland. 2021–2027. T.. Chem. Responses of juvenile Atlantic salmon (Salmo salar) to episodic increases in acidity of Nova Scotia rivers.L. 1–15. Guilhermino. E. Saf. (1999). Biomarkers 5. 16. Characterization of cholinesterase from guppy (Poecilia reticulata) muscle and its in vitro inhibition by environmental contaminants. Ecol. A. Toxicol. A. Guilhermino. Aquat.M. Peakall. Whole effluent toxicity testing with Oncorhynchus mykiss (Walbaum 1792): survival and behavioral responses to a dilution series of a mining effluent in South Africa. 13–23.R. Biochem. Fish introductions in Europe with particular reference to its central and eastern part. Carvalho..S. B. Chem. and Garric.A. 48.A. and Soares. Can.W. Vol.P. A. Ecotoxicol. 393– 397. Migeon. ¨ ¨ (1996). Environmental Toxicology and Risk Assessment. Total Environ.. (2000). F. (1992). Saf. D. A.E. and Townsend. Biochemical and (ultra)structural hepatic perturbations of Brachydanio rerio (Teleostei. Goncalves. 113C. 16. G. 42. (1974). . A. 501–505. Preliminary observations on the ecological relevance of the Gammarus ‘scope for growth’ assay: effect of zinc on reproduction. and Guilhermino. (2000). Contam. Aquat. Environ. R. 151–155. Environ. M. Chem. Response of xenobiotic metabolizing enzymes in rainbow trout (Onchorhynchus mykiss) to endosulfan.M. 23. Lindstrom-Seppa. (1989). E.. Tewari. (1990). Maltby. M. 4. Garcı´ a-Berthou. D. 1727–1732. Jones.M... 727-738. Little (eds). Silveira.4-dichloroaniline and sodium bromide with the 21-day chronic test and using dehydrogenase activity of Daphnia magna Straus. M. A. B. Holcı´ k. 247.J.P. Food of introduced mosquitofish: ontogenic diet shift and prey selection. M. Biol. 100C. J. Nogueira. Chemosphere 32.U. In T. Koponen. A. pp. A linkage between biomarkers at two levels of biological complexity. OECD Guideline for Testing of Chemicals 203.. (2000). Donato. Yin. (1998). Egaas. J. (1998). P. Comp.M. Jensen. Sci. and Carvalho. London. Grøsvik. Water Pollution and Fish Physiology. Goksøyr.V. L. (1996). Chem. Environ.B.. Bull.and longterm effects of copper on the rosy barb (Puntius conchonius Ham.N.. UK: Chapman & Hall. L. Inc. 216–233. Beyer. (1997). Toxicol. (rosy barb). Lopes. P. Paris-Palacios.) to resin acids and pulp mill effluent: laboratory and field experiments.V. (1998). Lopes. Sandvik.. Huuskonen. A.M. Should the use of inhibition of cholinesterases as a specific biomarker for organophosphate and carbamate pesticides be questioned? Biomarkers 3. (1991). Physiol. OECD (1982). E. H. (1994). Fish. and Santos. M. Wood. Sci. Garsdal. Animal Biomarkers as Pollution Indicators. Arch. J. J. Pabst. and Pande. Skaare. R.... 249. A... Short. Toxicol. J. 21. Chem. F. S. A. Mar. Toxicol. Saf. Discriminating the ecotoxicity due to metals and to low pH in acid mine drainage. (1991). Pollut. W... Soares. and Soares. A. (1990). Ecotoxicol. A. and Ribeiro. and Jakoby. Tewari. (1992). 1475–1484. L. An in situ river exposure vessel for bioaccumulation studies with juvenile fish.. R. Heath. Impact assessment of a wastewater treatment plant effluent using the fish biomarker ethoxyresorufin-O-deethylase: field and on-site experiments. (1991).V. Inhibition of acetylcholinesterase activity as effect criterion in acute tests with juvenile Daphnia magna. 36–45. Kosmala. 2085–2092. E. Comp. Toxicol. 135– 147.G. and Naylor.. Vernet. Habig. Lacroix. I. (2000). Sci. (1995). E. J.. Paris: OECD.H. (1996).M. E.. A. 157–163.J. Garcia. J. Fish.. 109–124. J. USA: CRC Press. and Skaare. S. Aquat. 19. 55. D. Maltby. USA: American Society for Testing and Materials. 4. Environ. 294–306. Biotransformation ¨ ¨ and other toxicological and physiological responses in rainbow trout (Salmo gairdneri Richardson) caged in a lake receiving effluents of pulp and paper industry. S. J. Flammarion. 137–141. 309– 316. Fish Biol. Lopes. Toxicol. M.). Guilhermino.S. Biomarker responses in flounder (Platichthys flesus) and their use in pollution monitoring.. and Chapman.M. 81–93. and Roy.G. K. L. J.V. Can.. pp. Environ.M.. R. 207–214.T. and Yu. L. 187–204. McLoughlin. and Soares. Little. Barros. L.C. ¸ (1999). Biagianti-Risbourg. S..C.U. P. 17. 1799–1806. C. ASTM STP 1262.. A. and Goksøyr.V. C. In vivo and in vitro effects of cadmium on selected enzymes in different organs of the fish Barbus conchonius Ham. Measures of fish behavior as indicators of sublethal toxicosis during standard toxicity tests. Aquat. Funct. P. Jensen.. P.. detected by enzyme activities and immunochemical methods. Acetylcholinesterase inhibition and altered locomotor behavior in the carabid beetle Pterostichus cupreus. 7130–7139. Castro. Saf. T. J. 44.. and Baatrup. Biochemical and genotoxic responses of adult eel (Anguilla anguilla L. 205–213. Guilhermino. L..S. H. Environ. E. 19–28. Soares. Gill. Lindstrom-Seppa. Profenofos toxicity to the eastern rainbow fish (Melanotaenia duboulayi). M. H. Comparative study between the toxicity of 3. Gerhardt. Environ. Environ. L.M. Toxicol. Chemosphere 28.M. (1998).. La Point. L. In vitro and in vivo inhibition of Daphnia magna acetylcholinesterase by surfactant agents: possible implications for contaminant biomonitoring. (1987). Gill. Glutathione S-transferases: the first enzymatic step in mercapturic acid formation.C. and Oikari. Pacheco. B. Egaas. G. K. 33.E. 274–284. Evaluation of sensitivity and specificity of two crustacean biochemical biomarkers. Price. Silva. 2nd edn. Cyprinidae) exposed to two sublethal concentrations of copper sulphate. 50.E. Physiol. (1999). Ecotoxicol. Environ.. A.. and Sloan. M. Fish. acute toxicity test. Ecotoxicol.C. Lacerda. J. (1996).M. Ribeiro. L. A. 81–92. 41. 44. N. A. Effects of non-ortho-substituted polychlorinated biphenyls (congeners 77 and 126) on cytochrome P4501A and conjugation activities in rainbow trout (Onchorhynchus mykiss). Biochem.M. and DeLonay. Kumar. and Pande.680 Castro et al. 35. Toxicol. W. Aquat. 8.

Aquat. M. J..E. E.. Vassault. and Parrot. 69–80.. caught in four harbours on the southwest coast of Iceland. Biochem. Jokinen. Guilhermino. Toxicol. Van der Oost. and Forlin. L. Svavarsson. M. 118–126.P. Sturm. Cieˆncia Biolo temat. A. Saf. Freshwater Res. Water Res. A. Int. Pereira. Biomarker responses in whitefish (Coregonus lavaretus L. Wall. USA. J. Vol. Toxicol. and Lafaurie.J. J.W.M.. Environ. M. H. The S.. and Soares. Dutil. 339–345.S. Novel bioassay based on acetylcholinesterase and lactate dehydrogenase activities to evaluate the toxicity of chemicals to soil isopods. Environ. (1998). 33–47. G. K. H. and Soares.M.D.. R. A simple. R. M. and Vermeulen. 18.. (1998).E. 24. Biomonitoring of aquatic pollution with feral eel (Anguilla anguilla) – II.D.R.R.F.G. Correia. Vertebrados da zona da Mina de Sao ˜ ´gica – Ecol. Costa.. E.M. Phipps. M. Goksøyr. Goncalves. Different sensitivity to organophosphates of acetylcholinesterase and butyrylcholinesterase from three-spined stickleback (Gasterosteus aculeatus): application in biomonitoring. 34. Karels.M. P.M.D.G. III. Serafini.M. (1999). Environ.) experimentally exposed in a large lake receiving effluents from pulp and paper industry.L. Pereira.M.. and Soares. Exp. Mar. ¸ A. P. and Crivello. 46.. 189–222.V. Ribeiro. G. 198.C. D. . Sibley. Mahony. 287– 293. s. Surface binding of contaminants by algae: consequences for lethal toxicity and feeding to Daphnia magna Straus. Sousa. (1984). I. Lactate dehydrogenase: UV-method with pyruvate and NADH. R.A. A. Ribeiro.. and Ankley. R.. Ribeiro. ´ Stien. 1493–1497.. O. 36. 2325–2336. Moura. J.. (2000). H.. J. 44. E. Thomann. Chem. T. N. A.A..J. M.. J. Hoke. 18. S. and Guderley. I. Martins.. Transferases.). Benoit. P. 17. (1995).K.. (1995). Lopes. and Ristola. A. Toxicol. 145–151. 2504–2511. 19(6) 1607–1615. X. Goncalves. Salmo gairdneri. and Soares. Aquat. J. (1995). (2000).V. (1978). Chem. J. I.. Waiwood.T. Oikari.. (2000). (1996).V.. The effect of copper.W. Biol. Oikari. Environ. M.M. A. G. 31..V. Pereira. Methods of Enzymatic Analysis.. 27–38. Environ. Wilde. 18. R. F. (1998). Romeo.H.M. 99. Baird. J.. Balcer. New York: Academic Press. A. M. A. Fish Biol. hardness and pH on the growth of rainbow trout. Toxicol. and Ribe¸ iro. G. J.U.V..M.. Soares. West. Soimasuo.M.H. Santti.. Biomarkers: pollution–induced biochemical responses. Bergmeyer (ed. 13. Assessment of various biomarkers in winter flounder from coastal Massachusetts. Celander. Ecotoxicol. R. M. A. Vol..A. F. Toxicol. Sturve. and Romeu. Canteiro. SysDomingos (Baixo Alentejo).. ¸ A. 15. Environ. 329–345. Test chambers and test procedures for in situ toxicity testing with zooplankton. Segner. K. 591–598.M. Toxicol.T. (1998). Pollut.. M. 783–785.M... Wogram.. Aquat. In situ bioassay chamber for assessment of sediment toxicity and bioaccumulation using benthic invertebrates. K. A. (1983).J. A.. Percic. J. Gnassia-Barelli. Cu(II) and Cd(II) inhibition of rat–liver glutathione-S-transferase – a steady-state kinetic study. (1995). Environ. Stephensen.L. R.M. Chem. Taylor.l.. 47. and Liess. inexpensive in situ method for assessing acute toxicity of effluents to fish. Domingos Mine: a study of heavy metals contamination in the water column and sediments of the Chanca river basin by discharge from an ancient ¸ cupriferous pyrite mine (Portugal)..B. D. Pereira. Jessen-Eller. 3rd edn. (1991). and Oikari. Evaluation of biomarkers in caged fishes and mussels to assess the quality of waters in a bay of the NW Mediterranean Sea. In H.O. Goncalves. Blier. D. Ecotoxicol. L. 412–419. Soimasuo.In Situ Bioassay With Small Fish 681 Pelletier.O. pp. 17. A.. A. Chem. W. and Beamish.. 1956–1964. Arch.R. Water-column.. Environ. F.F. A. Toxicol. Toxicol. 431–442.M (1999). Biochemical ¨ indicators of pollution exposure in shorthorn sculpin (Myoxocephalus scorpius). How should enzyme activities be used in fish growth studies? J. Contam. sediment and in situ chronic bioassays with cladocerans. Toxicol. Heida. (1999). 48.A. Environ. M.. III Enzymes: Oxireductases. Chem. Biomarker responses along a pollution gradient: effects of pulp and paper mill effluent on caged whitefish. Ericson. Kukkonen. Saf.P. A...V. Adolfsson-Erici. C. 497–505.