Avoidance responses to aluminium in the freshwater bivalve Anodonta cygnea

Aquatic Toxicology 55 (2001) 137– 148

www.elsevier.com/locate/aquatox

Avoidance responses to aluminium in the freshwater bivalve Anodonta cygnea Eniko˜ Ka´da´r a, Ja´nos Sala´nki b, Ravin Jugdaohsingh c, Jonathan J. Powell c, Catherine R. McCrohan a,*, Keith N. White a b

a School of Biological Sciences, Uni6ersity of Manchester, Manchester M13 9PT, UK Balaton Limnological Research Institute of the Hungarian Academy of Sciences, 8237, Tihany, Hungary c Gastrointestinal Laboratory, The Rayne Institute, St Thomas’ Hospital, London SE1 7EH, UK

Received 30 August 2000; received in revised form 5 March 2001; accepted 7 March 2001

Abstract This study examined the effect of aluminium (Al) on the filtering behaviour (shell opening or gape) of the freshwater bivalve Anodonta cygnea L in neutral fresh water. Parallel measurements of Al concentration in the soft tissues were made to examine the relationship between changes in behaviour and accumulation of Al. The number of lysosomal granules in the gill, kidney and digestive gland were counted, as lysosomes are known to be involved in the excretion and detoxification of trace metals. The bivalves were exposed to two environmentally relevant concentrations of added Al i.e. 250 and 500 mg l − 1 (9.25 and 18.5 mM l − 1) at neutral pH for 15 days and shell movement monitored continuously. Aluminium affected the mussels’ filtering activity, producing an avoidance reaction whose magnitude was concentration-dependent; 250 mg l − 1 added Al produced no detectable change, while 500 mg l − 1 Al reduced mean duration of shell opening by 50%. This effect was irreversible over a 15 day recovery period. Tissue levels of Al after 15 days exposure were an order of magnitude higher in animals exposed to 250 mg l − 1 added Al than in those exposed to 500 mg l − 1. This was consistent with the inhibition of filtering activity due to valve closure at the higher concentration, which may have prevented uptake of Al. In addition, probable different chemical speciation of Al in the water column (soluble for 250 and colloidal for 500 mg l − 1) may lead to marked differences in tissue uptake. The kidney and digestive gland were the main sites of accumulation of Al and concentrations remained significantly elevated 15 days after transfer of animals to clean water. It is suggested that mucus plays a role in the exclusion of Al as elevated concentrations were measured in the pseudofaeces of animals during and after exposure. Lysosomal granules may be involved in the intracellular handling and detoxification of Al as numbers increased significantly in all organs during exposure and continued to increase after the animals were transferred to clean water. The present study provides evidence for the bioavailability and toxicity of Al to mussels at neutral pH and at concentrations which are known to enter neutral freshwaters when mobilised by natural or anthropogenic acidity. The changes in behaviour and uptake of Al in the mussel observed in this investigation are, therefore, likely

* Corresponding author. Tel.: + 44-161-2755375; fax: + 44-161-2755363. E-mail address: [email protected] (C.R. McCrohan). 0166-445X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 4 4 5 X ( 0 1 ) 0 0 1 8 3 - 7

138

E. Ka´da´r et al. / Aquatic Toxicology 55 (2001) 137–148

to be reflected in the natural environment and the degree to which Al affects the ‘fitness’ of the mussel populations and the transfer of Al through the food chain merit investigation. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Aluminium; Behaviour; Avoidance response; Pseudofaeces; Bivalve; Lysosomal granule; Anodonta cygnea

1. Introduction Aluminium is highly insoluble at neutral pH, suggesting limited bioavailability in the freshwater environment (Driscoll and Schecher, 1989). However, some aquatic organisms, such as the snail Lymnaea stagnalis, can accumulate significant quantities of Al in their soft tissues (Elangovan et al., 1997) probably by ingesting polymeric Al complexes trapped in the mucus on which they graze (Jugdaohsingh et al., 1998). Concentration factors comparable to those observed for essential trace metals such as copper in the amphipod Choropium 6olutator exposed to concentrations in the water ranging between 100 and 5000 mg l − 1 (Langston and Spence, 1995), and zinc in the tissues of prawns, Palemonetes 6arians and Palemon elegans exposed to 100– 500 mg l − 1 Zn (Nugegoda and Rainbow, 1989), were observed by Elangovan et al. (1997) for Al in L. stagnalis following 30 days exposure to environmentally relevant concentrations of the metal (100– 500 mg l − 1). Exposure to Al also resulted in behavioural perturbations in L. stagnalis, such as changes in locomotory activity (Truscott et al., 1995) and depression of feeding behaviour (Campbell et al., 2000), which are likely to affect the fitness of the snail. Filter feeding organisms, such as bivalve molluscs, are potentially exposed to more Al than grazing gastropods as they come into contact with both particulate (colloidal or ligand-bound) and dissolved metal whilst feeding. The present study combines the analysis of behaviour patterns, tissue concentrations and sub-cellular changes to assess the response of the swan mussel, Anodonta cygnea to sub-lethal Al exposure. Lake water, maintained at near-neutral pH (7– 7.5), was used in these experiments to reflect the chemical speciation of Al that can occur in the natural environment, although direct comparison to the Al speciation in the lake was not possible.

Under optimal conditions of feeding, the adductor muscle of A. cygnea is relaxed and the valves are open. Extended periods of shell gape are interrupted by rest periods, giving a continuous pattern of periodicity. Such rhythmic behaviour may be perturbed by changes in environmental conditions (Sala´ nki, 1992; Kontreczky et al., 1997), thus providing a measure of the impact of pollutants such as Al. The average duration of shell opening as a proportion of time engaged in filtering was monitored in A. cygnea following exposure to concentrations (250 and 500 mg l − 1 added Al) commonly present in UK natural surface waters (Dixon and Gardner, 1998). Accumulation of Al by the soft tissues of the mussel was measured in parallel with behavioural changes, as metal body burden in aquatic invertebrates may be predictive of bioavailability and hence toxicity, and ultimately, the ecological impact of a metal (Depledge et al., 1995). However, the relationship between environmental metal concentration and toxicity is complex because of the degree to which accumulated metal is detoxified within the body (Birge et al., 2000). An important detoxification mechanism is thought to be the sequestration and storage of metals in intracellular lysosomal structures or ‘granules’ (George and Pirie, 1980; Duquesne and Coll, 1995). Proliferation of granules has been observed in the digestive gland of marine mussels in response to both xenobiotics and trace metals (Cajaraville et al., 1992). Such changes may reflect the balance between uptake and depuration and so provide information on the half-life of the toxin. The number of lysosomal granules was, therefore, counted in selected tissues of A. cygnea using light microscopy as an indication of relative changes in the storage and excretion of Al prior to, during and after exposure to the added metal.

E. Ka´ da´ r et al. / Aquatic Toxicology 55 (2001) 137–148

2. Materials and methods

2.1. Animal maintenance and experimental conditions Adult A. cygnea (9 – 11 cm shell length) were collected from a pollution free fishpond close to Lake Balaton, Hungary (46°90.22% N, 18°04.91% W), and acclimated to laboratory conditions for 2 weeks prior to experimentation. The animals were kept in a 100-l volume tank containing sandfiltered Lake Balaton water, which was continuously replenished from a mixing tank (100 l) by a pump at a rate of 4 l h − 1, to provide a natural food supply (Al concentration in the lake ranged between 20 and 40 mg l − 1, as confirmed by ICPOES measurements). Aluminium was added daily to the mixing tank as Al3 + from an acidic stock solution (pH 1.7) of 500 mg l − 1 prepared from Al(NO)3 · 9H2O (Sigma Chemical Co). The final added Al concentration in the exposure tank was either 250 or 500 mg l − 1. The calculated residence time of water containing Al in the exposure tank was 25 h. Near neutral pH (7– 7.5) was maintained by adding nitric acid (Merck Ltd) to the mixing tank. Experiments were carried out at room temperature (19– 22°C) in the laboratory where mussels were subject to a natural dark:light regime of 8–10 h dark: 14– 16 h light. Water samples were taken daily from both mixing and exposure tanks by removing 10 ml aliquots from 10 cm below the surface at the centre of the tanks, and analysed for total Al using inductively coupled plasma optical emission spectroscopy (ICPOES). Average concentrations of three samples taken from different distances from the inlet were analysed at each time.

2.2. Experimental design The experiments using the two exposure levels (500 and 250 mg l − 1 added Al) were carried out between November 1998 and January 1999 and May and July 1999, respectively. About 35 animals were exposed to Al in each experiment. In each treatment, eight animals were permanently attached to a mussel actograph for shell gape monitoring (Vero´ and Sala´ nki, 1969). This

139

device permits continuous monitoring of adductor muscle activity without disturbing the free movement of the mussels. Duration of shell opening and closing was measured continuously over 45 days — 15 days under control conditions with no added Al, followed by 15 days exposure to Al and a further 15 days of recovery with no added Al. Actographs were obtained using signal analysis system Signalview (Real Time Devices USA, Inc.), which permitted the calculation of average duration of shell gape from the sum of the consecutive shell gapes averaged for eight animals. The experimental design for the behavioural response (shell gape) study was, therefore, factorial 8× 3, involving eight animals and three treatments. A further 15 animals were sampled for measurement of Al accumulated in the digestive gland, kidney and gill, following dissection of three animals on day 0 (control), day 8 and 15 of the exposure period and day 8 and 15 of the recovery period. Pseudofaecal samples were collected daily, pooled for the periods between each dissection, and their Al concentration determined. Aluminium concentrations in the soft tissues and pseudofaeces were measured by ICPOES (JobinYvon, JY24 Instrument SA, France). The analytical wavelength was 396.15 nm and sample-based standards were used as earlier described (Jugdaohsingh et al., 1998). Tissue and pseudofaeces were digested prior to analysis as earlier described (Jugdaohsingh et al., 1998). Briefly, 0.5–1 g of (oven) dried tissue was digested with equal volumes (5 ml) of Aristar grade concentrated (65% v/v; Merck Ltd) nitric acid and 30% (v/v) hydrogen peroxide (Merck Ltd). Digested samples were diluted 1+ 3 with ultra-high purity deionized water prior to analysis. Further tissue samples (1 mm3) of digestive gland, gill and kidney were embedded in Spurr resin following primary fixation in 2.5% glutaraldehyde solution prepared in the mussel’s haemolymph as a medium iso-osmolar (46 mOsm l − 1) to the tissue fluid (Hayat, 1981). Thick (0.5 mm) sections were cut using a glass knife and a Reichert Jung IV ultramicrotome and stained with toluidine blue (1% w/v) in borax (1% w/v). Triplicate sections for each organ from two animals at each time interval were observed under a

140

E. Ka´ da´ r et al. / Aquatic Toxicology 55 (2001) 137–148

light microscope (Leitz-Dialux, Wetzlar). The total number of lysosomes was counted in randomly chosen 100 mm2 areas using a graduated bar graticule. This experimental design was factorial 2× 3×5 involving two animals, three organs (gill, kidney and digestive gland) and five treatments (control, day 8 and 15 of exposure and day 8 and 15 of recovery). Animals removed from the tank for dissection were replaced by fresh individuals each time (total 12 animals) in order to maintain approximately the same rate of removal of Al associated with mussels during the experiment.

2.3. Statistical analysis Changes in the behavioural response, as well as changes in the number of lysosomes were tested using factorial analysis of variance. One-way analysis of variance (ANOVA) was used for determining the level of significance of differences in lysosomal numbers between control and different treatments within each organ. Specific differences between treatments were examined a posteriori using the S-N-K multiple range test (Steel and Torrie, 1980). The significance of the results from tissue partitioning measurements was tested using a Student’s t-test (assuming equal variances). Calculations were made using SPSS (SPSS, Inc). The level of significance was taken as P B0.05.

3. Results

3.1. Aluminium in the water column Average daily amounts of Al in the water column are shown in Fig. 1 for both experiments. The daily concentrations of Al measured in the mixing and exposure tanks over the 15-day exposure were 525.3996.7 and 272.4920.6 mg l − 1 (19.4 and 10 mM l − 1) for the mixing tank, and 516.39 12.4 and 241.39 24.8 mg l − 1 (19.1 and 8.9 mM l − 1) for the experimental tank, respectively. Water quality parameters are shown in Table 1 and represent values measured daily or taken from the Balaton Limnological Research Institute’s database.

3.2. The effect of Al on duration of shell gape Mean duration of shell gape for the control periods were markedly different as the two experiments were carried out at different times of the year (Controls in Fig. 2). Animals monitored during the spring months (May–July) showed approximately twice the mean shell gape duration compared with winter animals. Despite this difference, exposure to 500 mg l − 1 Al over 15 days caused a significant (PB0.05) decrease in mean duration of shell opening, which fell by 50% (Fig. 2); in contrast, animals exposed to 250 mg l − 1 added Al showed no significant (P= 0.263) change in behaviour (Fig. 2). Shell gape of animals exposed to 500 mg l − 1 Al did not recover to pre exposure levels following transfer to untreated lake water for 15 days.

3.3. Al partitioning in the soft tissue and pseudofaeces Metal concentrations in the kidney and digestive gland of exposed animals were significantly higher (PB 0.05) than in controls by day 8 of exposure, and remained elevated at day 15. Animals exposed to 250 mg l − 1 added Al showed an order of magnitude higher concentration of the metal in the soft tissues than those exposed to 500 mg l − 1 added Al (Fig. 3). Following transfer to untreated lake water a rapid loss of Al was observed but concentrations remained significantly elevated (P=0.015) in the kidney of animals exposed to 500 mg l − 1 added Al. Gill Al content was lower compared with other organs, but was significantly higher (P= 0.04) than in controls on day 8 of the 250 mg l − 1 Al exposure and day 15 to 500 mg l − 1exposure. Concentration factors (Table 2) ranged between 7 and 12×103, and reached highest values in the kidney and digestive gland for both exposure concentrations. Concentration factors of 4.9× 103 in the kidney and 9.9× 103 in the digestive gland were recorded on day 15 of exposure to 250 mg l − 1 added Al. Measurements were made of the Al content of pseudofaeces pooled for the specific time intervals; hence no statistical tests were carried out. The Al concentration in pseudofaeces produced

E. Ka´ da´ r et al. / Aquatic Toxicology 55 (2001) 137–148

by animals exposed to 250 mg l − 1 added Al was greater than in controls throughout the exposure and higher than that produced by animals exposed to 500 mg l − 1 added Al (Fig. 4.). Alu-

141

minium concentration in pseudofaeces continued to increase up to day 8 of the recovery period, but then fell to control levels over days 8–15 of being transferred to untreated lake water.

Fig. 1. Mean concentration of Al in the water column 9 S.E. (n = 3) in the mixing () and experimental ( ) tanks during 15-days exposure to (A) 250 and (B) 500 mg l − 1 added Al followed by 15 days recovery in clean, untreated lake water.

Table 1 Water quality parameters measured in the sand-filtered lake water during the two experiments (parameters marked the untreated lake water and were taken from Balogh and Voros (1997)) Water quality parameter Conductivity Dissolved organic carbona Chlorophyll aa Bacterial countsa a

are given for

During exposure to 250 mg l−1

During exposure to 500 mg l−1

6409 29 mS

720 931 mS

2.5×106 cells

40×106 cells

20–40 mg l−1 5–10 mg l−1

According to Balogh and Voros, 1997.

a

142

E. Ka´ da´ r et al. / Aquatic Toxicology 55 (2001) 137–148

Fig. 2. The effect of Al exposure on the duration of shell gape of A. cygnea measured over three 15 day periods (respectively, control, exposure, recovery) at an exposure concentration of 250 and 500 mg l − 1 added Al. Values are mean 9 S.E. (n= 24). Columns with the same letter are not significantly different according to S – N – K multiple range test (PB 0.05).

3.4. The effect of Al on lysosomal granule number The number of lysosomes increased in all organs of animals exposed to Al at both concentrations (Fig. 5). Analysis of variance (Table 3) showed a significant (P B0.001) effect of treatment and organ type on the number of lysosomes. Following Al exposure the number of granules increased significantly, but the change in the kidney, gill and digestive gland differed significantly. First order interactions were also significant between treatments and organs, indicating inter-organ differences in the pattern of change in number of lysosomes throughout the experiment. The largest increase in the number of lysosomes was

observed in the kidney where numbers doubled by the end of the 15 day exposure to 250 mg l − 1 added Al, and continued to increase to more than triple that of controls after the animals were transferred to untreated lake water. A more than two-fold increase in number of lysosomes was observed in the gills, but following transfer to clean water for 8 days a reduction was observed to control values. The digestive gland contained the highest density of lysosomes under control conditions and numbers increased in mussels exposed to 250 mg l − 1 added Al by day 15 of exposure and which continued following transfer to untreated lake water (Fig. 5). Exposure to 500 mg l − 1 added Al resulted in a less pronounced increase in the number of lysosomes in the kidney and the increase was delayed compared with that recorded following exposure to 250 mg l − 1, becoming significant only on day 15 and after transfer of animals to untreated water (Fig. 5). The number of lysosomes in the digestive gland was greatest at the higher level of Al exposure and a further increase was observed following transfer of mussels to clean water. A two-fold increase in the number of lysosomes occurred in the gill during exposure to Al, followed by a fall to the pre-exposure levels as in mussels exposed to 250 mg l − 1 added Al.

4. Discussion These results demonstrate a clear effect of Al on the duration of shell gape, which is indicative of the time engaged in filtration. This suggests that A. cygnea is able to detect and respond to increased Al concentrations in natural lake water by closing its shell and thus reducing uptake. However, after prolonged closure, which is facilitated by the ability to respire anaerobically (Willson and Burnett, 2000), the build up of lactic acid eventually necessitates a resumption of filtration, through which the metal may enter the body. Accumulation of Al by the tissues, therefore, did occur in mussels exposed to 500 mg l − 1 added Al, but to a much lesser degree than those exposed to 250 mg l − 1 Al, a concentration which did not significantly affect filtration. Decreased shell gape

E. Ka´ da´ r et al. / Aquatic Toxicology 55 (2001) 137–148

may have caused a reduction in bioaccumulation due to reduced filtration, although this is unlikely to account for all of the marked difference between mussels exposed to 250 and 500 mg l − 1 Al. Another possible explanation for the difference in Al accumulation in the two experiments is related to the chemistry of Al, which is markedly different at the two concentrations used in our experiments (Jugdaohsingh et al., 1998). Clearly Al was maintained in the water column at both concentrations, as shown in Fig. 1, but it is likely

143

that the chemical species were quite different. Although in this work, we did not analyse the chemical speciation of Al in the water column, it is likely that at pH close to 7 and at the lower concentration (250 mg l − 1), Al was maintained in solution in a ‘dissolved’ form (Al(OH)− 4 ion and small hydroxypolymers) (Jugdaohsingh et al., 1998), that would be expected to be more bioavailable to the mussel than larger hydroxypolymers probably formed at the higher Al concentration (500 mg l − 1). At concentrations around

Fig. 3. Aluminium partitioning in the soft tissues of A. cygnea exposed to 250 and 500 mg l − 1 added Al. Note the different scales of the Y-axes. Vertical bars represent means + S.E. (n = 3) and statistical differences between control and treatment values, tested by the Student’s t-test at P= 0.05, are marked by asterisks.

E. Ka´ da´ r et al. / Aquatic Toxicology 000 (2001) 000–000

144

Table 2 Concentration factors measured in the gill, kidney and digestive gland of A. cygnea exposed to 250 and 500 mg l−1 added Al for 15 days followed by 15 days recovery in clean water with no Al added Concentration factor (CF)a

Treatment

250 mg l−1 added Al

Exposure Exposure Recovery Recovery a

day day day day

8 15 8 15

500 mg l−1 added Al

Gill

Digestive gland

Kidney

Gill

Digestive gland

Kidney

618 286 81 1120

8893 9889 2420 3866

12015 4884 5226 1737

7 521 112 99

164 139 12 30

954 978 491 485

CF, mean Al concentration in the tissue (mg kg−1 dry weight)/mean Al concentration in the water (mg l−1) during exposure.

Fig. 4. Aluminium associated with pseudofaeces produced by A. cygnea exposed to two concentrations of Al (250 and 500 mg l − 1 added Al) for 15 days followed by 15 days recovery in untreated lake water.

500 mg l − 1 Al, the dominant species are large, colloidal hydroxy polymers which are less easily transported through biological membranes (Berthon, 1996), but still may have inhibited filtration as suggested by our earlier direct measurements on filtration activity (Ka´ da´ r et al., 2001). Thus it is possible that tissue burden is more readily observed with soluble (250 mg l − 1), while effects on shell gape are more likely to be caused by colloidal Al (500 mg l − 1). The different timing of the two experiments could also have influenced behaviour owing to seasonal variations in temperature and food

availability of Balaton Lake water (Vo¨ ro¨ s et al., 1996) influencing metabolic activity and tissue weight (Serra et al., 1999). Furthermore, Al speciation may also be subject to seasonal changes, since the bacterial and algal bloom peaks in November and April (Vo¨ ro¨ s et al., 1996) and this would have resulted in increased concentrations of extra cellular polysaccharides (EPS) and dissolved organic carbon (DOC) that influence metal ion chemistry (Conrad et al., 1991). However, data on seasonal changes in Lake Balaton do not suggest significant increase DOC levels in spring (Table 1). Aluminium concentrations during wash

E. Ka´ da´ r et al. / Aquatic Toxicology 55 (2001) 137–148

out, presumably also reflecting background levels in the lake, were higher in the experiment conducted in May–July. Conductivity, which also showed seasonal differences, is another major influencing factor of Al chemistry in that hydroxy precipitation is facilitated by high conductivity/ ionic strength solutions (Letterman and Vanderbrook, 1983) resulting in lower levels of bioavailable Al. It may be difficult, therefore, to separate concentration-dependent effects of Al from seasonal effects, although controls were carefully run on every occasion to allow for this. Thus, the inhibitory effect of 500 mg l − 1 Al exposure on shell gape, and hence bioaccumulation, was significant in spite of the differences observed in controls. Moreover, our results are

145

consistent with data from earlier studies on the filtering behaviour of A. cygnea which showed more accentuated decrease in average duration of filtration at lower Al concentrations compared with the high Al levels (500 mg l − 1) under similar experimental conditions (Ka´ da´ r et al., 2001). Filtration activity affects metabolic activity and hence mucus production including, possibly, its binding capacity for Al. The gill and associated ctenidial mucus is exposed to large volumes of water (MacGinitie, 1941; Ballantine and Morton, 1956; Fretter and Graham, 1976). Therefore, it is likely that mucus played an important role in trapping Al in this study. Under normal filtering conditions, the mucus layer is continuously renewed, providing new binding sites in the form of

Fig. 5. The effect of Al exposure on the number of lysosomal granules counted in 100 mm2 area from sections of kidney, digestive gland and gill of A. cygnea exposed for 15 days to 250 and 500 mg l − 1 added Al, followed by transfer to untreated lake water for 15 days. Vertical bars represent means 9S.E. (n= 6). Statistical difference between control and treatment was tested by one way analysis of variance. Columns with the same letters are not significantly different according to S – N – K multiple range test (PB 0.05) within a given organ and given exposure concentration.

146

E. Ka´ da´ r et al. / Aquatic Toxicology 55 (2001) 137–148

Table 3 Factorial analysis of variance testing the effect of Al exposure, the intrinsic variability of the animals and three organs (kidney, digestive gland and gill) on the number of lysosomes of A. cygnea exposed to 250 and 500 mg l−1 added Ala Source of variation

Animals Organs Treatments Animals×organs Animals×treatments Organs×treatments Animals×organs×treatments Error Total

dfb

1 2 4 2 4 8 8 60 89

250 mg l−1

500 mg l−1

MSc

Fd

MSc

Fd

35 13657 4109 143 42 1854 179

0.4 173*** 52*** 2 0.5 23*** 2*

288 25794 5848 401 331 3947 211

3*** 315*** 71*** 4** 4*** 48*** 3**

a

Probability levels for significant effects are: PB0.05 (*), PB0.01 (**) and PB0.001 (***). df, degrees of freedom. c MS, mean squares. d F, F-ratio. b

the mucus glycoproteins that avidly bind trivalent metals (Conrad et al., 1991) including Al (Whitehead et al., 1995; Powell and Thompson, 1993; Exley et al., 1996). This could provide an explanation for the higher concentration of Al in the pseudofaeces produced by the animals with unaltered filtration activity at the lower exposure concentration (250 mg l − 1 added Al) and may be a further defence mechanism to prevent acute toxicity of the metal. Despite the animals’ ability to restrict uptake of Al, significant amounts do enter the body, again, possibly trapped by mucus associated with the food grooves bordering the gills along which strings of food-laden mucus is transported to the mouth. High concentration factors (\ 103) were measured in tissues of animals exposed to 250 mg l − 1 added Al, and were similar to those measured by Elangovan et al. (1997) in Lymnaea stagnalis following exposure to 285 mg l − 1 Al in artificial lake water containing no organic material. However, the two species are likely to handle the metal differently, owing to differences in feeding, digestion and respiration. The marked increase in the density of lysosomal granules observed in exposed animals indicates a close interaction between the lysosomal system

and Al accumulation. This is consistent with the results of other workers who reported a change in lysosomal activity in response to other metals (Moore, 1991; Bebianno and Langston, 1991; Winston et al., 1996). Copper, cadmium and manganese were reported to be accumulated within the lysosomal system in Mytilus gallopro6incialis (Viarengo et al., 1985). This author suggested that the metals are rapidly detoxified and observed that they are eliminated by exocytosis from the cells following transfer of animals to clean water. Our results indicate that lysosomes may play a role in handling Al at the cellular level. Aluminium is a class ‘A’ metal and will bind to ligands with oxygen as the donor atom such as phosphate (Nieborer and Richardson, 1980). This could explain the high concentrations of Al measured in the kidney, where granules containing heavy metals (i.e. Zn, Pb) have been found to proliferate during exposure to such metals (Regoli and Orlando, 1992). The continuous increase in the number of granules after transfer of the mussels to untreated lake water suggests that the lysosomal system may be involved not only in the sequestration but also in the excretion of Al and this aspect deserves further investigation.

E. Ka´ da´ r et al. / Aquatic Toxicology 55 (2001) 137–148

5. Conclusion This work demonstrates the bioavailability of Al to the mussel A. cygnea in the environmentally relevant range of 250 and 500 mg l − 1 added to neutral lake water. The kidney and digestive gland are likely to be medium- and long-term storage sites and their lysosomal systems are probably involved in the sequestration and excretion of the metal. We confirm the usefulness of behavioural studies for assessing responses to sub-lethal Al concentrations. Acknowledgements This work was supported by a postgraduate scholarship from the Open Society Institute, Hungary and the Foreign Office, UK. The authors acknowledge Dr R.S. Callow for advice and assistance with statistical analysis. References Balogh, K.V., Voros, L., 1997. High bacterial production in hypertrophic shallow reservoirs rich in humic substances. Hydrobiologia 342, 63 –70. Ballantine, D., Morton, J.E., 1956. Filtering, feeding and digestion in the lamellibranch Lasaea rubra. J. Mar. Biol. Assoc. UK 35, 241 –274. Bebianno, M., Langston, W., 1991. Metallothionein induction in Mytilus edulis exposed to cadmium. Mar. Biol. 108, 91– 96. Berthon, G., 1996. Chemical speciation studies in relation to aluminium metabolism and toxicity. Coordination Chem. Rev. 149, 241 – 280. Birge, W.J., Price, D.J., Shaw, J.R., Spromberg, J.A., Wigginton, A.J., Hogstrand, C., 2000. Metal body burden and biological sensors as ecological indicators. Environ. Toxicol. Chem. 19, 1199 –1212. Cajaraville, M.P., Volkl, A., Fahimi, H.D., 1992. Peroxizomes in the digestive gland cells of the mussel Mytilus gallopro6incinalis LMK biochemical, ultrastructural and immuno-cytochemical characterisation. Eur. J. Cell Biol. 59, 255 – 264. Campbell, M.M., Jugdaohsingh, R., White, K.N., Powell, J.J., McCrohan, C.R., 2000. Aluminium toxicity in a molluscan neuron: effects of counterions. J. Toxicol. Environ. Health Part A 59, 253 – 270. Conrad, M.E., Umbreit, J.N., Moore, E.G., 1991. A role for mucin in the absorption of inorganic iron and other metals. Gastroenerology 100, 129 –136.

147

Depledge, M.H., Aagaard, A., Gyorkos, P., 1995. Assessment of trace metal toxicity using molecular, physiological and behavioural biomarkers. Mar. Pollut. Bull. 31, 19 – 27. Driscoll, T.C., Schecher, W.D., 1989. Aluminium in the environment. In: Metal Ions in Biological Systems, vol. 24. Marcel Dekker, New York, pp. 74 – 104. Dixon, E., Gardner, M., 1998. Reactive aluminium in UK surface waters. Chem. Speciation Bioavailability 10, 11 – 17. Duquesne, S., Coll, J., 1995. Metal accumulation in the clam Tridacna crocea under natural and experimental conditions. Aquat. Toxicol. 32, 239 – 253. Elangovan, R., White, K.N., McCrohan, C.R., 1997. Bioaccumulation of aluminium in the freshwater snail Lymnaea stagnalis at neutral pH. Environ. Pollut. 96, 29 – 33. Exley, C., Wicks, A.J., Hubert, R.B., Birchall, J.D., 1996. Kinetic constraints in acute aluminium toxicity in rainbow trout (Onkorhyncus mykiss). J. Theor. Biol. 179, 25 – 31. Fretter, V., Graham, A., 1976. Functional Anatomy of Molluscs. Academic Press, London. George, S., Pirie, B., 1980. Metabolism of zinc in the mussel Mytilus edulis (L): a combined ultrastructural and biochemical study. J. Mar. Biol. 60, 575 – 590. Hayat, M.A., 1981. Fixation for Electron Microscopy. Academic Press, London. Jugdaohsingh, R., Campbell, M.M., Thompson, R.P.H., McCrohan, C.R., White, K.N., Powell, J.J., 1998. Mucus secretion by the freshwater snail Lymnaea stagnalis limits aluminium concentration of the aqueous environment. Environ. Sci. Technol. 32, 2591 – 2595. Ka´ da´ r, E., Sala´ nki, J., Powell, J.J., White, K.N., McCrohan C.R., Effect of aluminium on the filtration activity of the freshwater mussel Anodonta cygnea L. in press. Kontreczky, Cs., Farkas, A., Nemcso´ k, J., Sala´ nki, J., 1997. Short and long term effects of deltamethrin on filtering activity of the freshwater mussel (Anodonta cygnea L.). Ecotoxicol. Environ. Saf. 38, 195 – 199. Langston, W.J., Spence, S.K., 1995. Biological factors involved in metal concentrations observed in aquatic organisms. In: Tessier, A., Turner, D.R. (Eds.), Metal Speciation and Bioavailability in Aquatic Systems, vol. 3. Wiley, Chichester, pp. 407 – 467. Letterman, R.D., Vanderbrook, S.G., 1983. Effect of solution chemistry on coagulation with hydrolysed Al(III)-significance of sulphate ion and pH. Water Res. 17, 195 – 204. MacGinitie, V.L., 1941. On the method of feeding of four pelecypods. Biol. Bull. 80, 18 – 25. Moore, M., 1991. Lysosomal changes in the response of molluscan hepatopancreatic cells to extracellular signals. Histochem. J. 23, 495 – 500. Nieborer, E., Richardson, D.H.S., 1980. The replacement of the nondescript term ‘heavy metals’ by a biologically and chemically significant classification of metal ions. Environ. Pollut. 1, 3 – 26. Nugegoda, D., Rainbow, P.S., 1989. Effects of salinity changes on zinc uptake and regulation by the decapod crustaceans Palemon elegans and Palaemonetes 6arians. Mar. Ecol. Prog. Ser. 51, 57 – 63.

148

E. Ka´ da´ r et al. / Aquatic Toxicology 55 (2001) 137–148

Powell, J.J., Thompson, R.P.H., 1993. The chemistry of aluminium in the gastrointestinal lumen and its uptake and absorption. Proc. Nutr. Soc. 5, 241 –253. Regoli, F., Orlando, E., 1992. Lysosomal responses as sensitive stress index in biomonitoring heavy metal pollution. Mar. Ecol. Prog. Ser. 84, 63 –69. Sala´ nki, J., 1992. Heavy metal induced behaviour modulation in mussels: possible neural correlates. Acta Biol. Hungarica 43, 375 – 386. Serra, R., Isani, G., Tramontano, G., Carpene, E., 1999. Seasonal dependence of cadmium accumulation and Cdbinding proteins in Mytilus gallopro6incialis exposed to cadmium. Toxicol. Endocrinol. 123, 165 – 174. Steel, R.D.G., Torrie, J.H., 1980. Principles and Procedures of Statistics — A biometrical approach. McGraw-Hill Inc, London. Truscott, R., McCrohan, C.R., Bailey, S.E.R., White, K.N., 1995. Effect of aluminum and lead on activity in the fresh-water pond snail Lymnaea stagnalis. Can. J. Fisheries Aquat. Sci. 52, 1623 –1629. Vero´ , M., Sala´ nki, J., 1969. Inductive attenuator for continuous registration of rhythmic and periodic activity of mus-

sels in their natural environment. Med. Biol. Eng. 7, 235 – 237. Viarengo, A., Moore, M.N., Pertica, M., Mancinelli, G., Zinicchi, G., Pipe, R.K., 1985. Detoxification of copper in cells of the digestive gland of mussel: the role of lysosomes and thioneins. Sci. Total Environ. 42, 135 – 145. Vo¨ ro¨ s, L., Balogh, V.K., Herodek, S., 1996. Microbial food web in a large shallow lake (Lake Balaton, Hungary). Hydrobiologia 339, 57 – 65. Whitehead, M., Powell, J.J., and Thompson, R.P.H., 1995. The gut mucus layer regulates metal absorption. Gut: A6 (abstr.). Willson, L.L., Burnett, L.E., 2000. Whole animal and gill tissue oxygen uptake in the Eastern oyster, Crassostrea 6irginica: effects of hypoxia, hypercapnia, air exposure, and infection with the protozoan parasite Perkinsus marinus. J. Exp. Mar. Biol. Ecol. 246, 223 – 240. Winston, G.W., Moore, M.N., Kirchin, M.A., Soverchia, C., 1996. Production of reactive oxygen species by haemocytes from the marine mussel, Mytilus edulis: lysosomal localisation and effect of xenobiotics. Comp. Biochem. Physiol. 113C, 221 – 229.