- Email: [email protected]
Development of an embryo toxicity test with the pond snail Lymnaea stagnalis using the model substance tributyltin and common solvents
Science of the Total Environment 435–436 (2012) 90–95
Contents lists available at SciVerse ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Development of an embryo toxicity test with the pond snail Lymnaea stagnalis using the model substance tributyltin and common solvents☆ Cornelia Bandow a, b,⁎, Lennart Weltje c a b c
University of Applied Sciences Zittau/Görlitz, Theodor-Körner-Allee 16, D-02763 Zittau, Germany LOEWE Biodiversity and Climate Research Centre (BiKF), Senckenberganlage 25, D‐60325 Frankfurt am Main, Germany BASF SE, Crop Protection, Ecotoxicology, Speyerer Strasse 2, D-67117 Limburgerhof, Germany
H I G H L I G H T S ► ► ► ►
The pond snail Lymnaea stagnalis is a suitable organism for ecotoxicological testing. A developmental toxicity test using L. stagnalis embryos is completed within 21 days. Sensitive endpoints are easily assessed. Low effort regarding human and financial resources.
a r t i c l e
i n f o
Article history: Received 19 March 2012 Received in revised form 2 July 2012 Accepted 2 July 2012 Available online 28 July 2012 Keywords: Mollusc Early life stages Organotin Pulmonate Embryonic growth Test guideline
a b s t r a c t The development of a chronic mollusc toxicity test is a current work item on the agenda of the OECD. The freshwater pond snail Lymnaea stagnalis is one of the candidate snail species for such a test. This paper presents a 21-day chronic toxicity test with L. stagnalis, focussing on embryonic development. Eggs were collected from freshly laid egg masses and exposed individually until hatching. The endpoints were hatching success and mean hatching time. Tributyltin (TBT), added as TBT-chloride, was chosen as model substance. The selected exposure concentrations ranged from 0.03 to 10 μg TBT/L (all as nominal values) and induced the full range of responses. The embryos were sensitive to TBT (the NOEC for mean hatching time was 0.03 μg TBT/L and the NOEC for hatching success was 0.1 μg TBT/L). In addition, data on maximum limit concentrations of seven common solvents, recommended in OECD aquatic toxicity testing guidelines, are presented. Among the results, further findings as average embryonic growth and mean hatching time of control groups are provided. In conclusion, the test presented here could easily be standardised and is considered useful as a potential trigger to judge if further studies, e.g. a (partial) life‐cycle study with molluscs, should be conducted. © 2012 Elsevier B.V. All rights reserved.
1. Introduction In 2010 a detailed review paper was published by the OECD, exploring the possibilities for mollusc life cycle testing with a focus on endocrine disruptors and other chemicals (Organisation for Economic Cooperation and Development, 2010). The need for (chronic) mollusc testing is partly motivated by the well documented effects of tributyltin (TBT) and triphenyltin (TPT) compounds on molluscs. In the past, effects of these compounds, used in antifouling paints on ship hulls, were recognised too late, which led to a threat for many marine mollusc populations. TBT and TPT compounds affect the endocrine system of
☆ This paper is dedicated to the memory of Thomas Knacker. ⁎ Corresponding author at: ECT Oekotoxikologie GmbH, Böttgerstrasse 2‐14, D-65439 Flörsheim, Germany. Tel.: +49 6145 9564-61; fax: +49 6145 9564-99. E-mail addresses: [email protected] (C. Bandow), [email protected] (L. Weltje). 0048-9697/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2012.07.005
molluscs and may cause effects known as imposex or intersex at very low concentrations (Oehlmann and Schulte-Oehlmann, 2003). In addition, Matthiessen (2008) refers to the ecological importance of this large and taxonomically diverse group of invertebrates and their under representation in regulatory aquatic testing. Further, Grosell et al. (2006) concluded that aquatic snails might not be protected under current regulatory guidelines. Their conclusion was derived from chronic toxicity studies on lead with freshwater invertebrates, i.e. rotifers Brachionus calyciflorus, non-biting midges Chironomus tentans and hatchlings of Lymnaea stagnalis, where L. stagnalis showed the highest sensitivity. In this paper, a method for conducting an embryo toxicity test with the pond snail is described. Testing embryos is particularly relevant as adverse effects of test compounds on critical developmental processes such as organogenesis can be studied. Also, due to L. stagnalis' high production rate of egg masses an embryo test can be standardised to a high extent. An embryo test might become a part of the mollusc life cycle test, which is currently under development by the OECD Endocrine Disrupter Testing
C. Bandow, L. Weltje / Science of the Total Environment 435–436 (2012) 90–95
and Assessment group (Organisation for Economic Co-operation and Development, 2010). In the detailed review paper on mollusc lifecycle toxicity testing, the pond snail L. stagnalis is proposed as a possible test species for test development, next to the prosobranch mollusc Potamopyrgus antipodarum and the marine bivalve Crassostrea gigas. L. stagnalis is well suited as a standard test organism: it is distributed holarctic and because of its halotolerance of up to 0.7% it is also found in the Baltic Sea (Glöer et al., 1992; Schwab, 2002). Further, breeding in the laboratory is relatively easy. Due to its considerable size, i.e. the maximum shell height is 7 cm; histopathological and biochemical analyses (e.g. on the haemolymph) could even be conducted. Maturity occurs after approximately 10 weeks at 20 °C. Furthermore, development of the embryos occurs in eggs outside the adult snail contrary to P. antipodarum, whose embryos stay in the brood pouch, or C. gigas, which produces planktonic larval stages. Because of the transparent egg capsule embryo development can be observed at any time. Being a simultaneous hermaphrodite, parental sex ratio in starting reproduction tests is not an issue. Due to the characteristics outlined above, L. stagnalis offers ample possibilities for mollusc tests, including a full life-cycle test (Leung et al., 2007), which would cover potential endocrine mediated effects, as well as the embryo toxicity test described in this paper, which may serve as a screening tool for assessing chronic toxic potential against molluscs within a short time.
2. Material and methods
91
In the eggs, embryos go through four different stages (Canton and Slooff, 1977; Lalah et al., 2007), compare Fig. 3: a) Morula: days 0 to 3, yellowish appearance, slow movement. b) Trochophora: days 3 to 5, grainy appearance, uninterrupted movement. c) Veliger: days 5 to 7, the embryo loses its round appearance, foot and shell can be distinguished. d) Hippo: from day 7 the embryo is fully developed; eyes, shell and foot are differentiated, heart beat is visible. Depending on the temperature, snails hatch between day 10 and day 20. At 20±2 °C the mean hatching time in our laboratory was between 12 and 14 days. 2.2. Culturing Snails were held in aquaria of different sizes (from 6 up to 36 L) under constant aeration. As a culture medium the M4-medium (Elendt and Bias, 1990; Organisation for Economic Co-operation and Development, 2004a) was used. The physical and chemical parameters of the water fulfilled the requirements of the OECD test guideline 202 (Organisation for Economic Co-operation and Development, 2004a). The culture was maintained in a climatised room at 20±2 °C and a 16:8 h light–dark-rhythm. L. stagnalis was fed ad libitum with different kinds of vegetables, e.g. cucumber, lettuce, carrot or courgette. Furthermore, cuttlebones were given in each aquarium to support shell growth. Faeces and remaining food was removed 5 to 6 times a week. In addition, periphyton on the glass walls was removed every week.
2.1. Test organism 2.3. Embryonic growth With its shell height of up to 7 cm, the great pond snail Lymnaea stagnalis is the largest aquatic snail domestic in Europe and Northern America. It is found in freshwater as well as in brackish water. The pond snail is a detritivore and feeds on plants, carrion, algae and microorganisms (Glöer et al., 1992; Schwab, 2002). The development of a newly hatched juvenile to an adult snail takes about 10 weeks at 20 °C (Schwab, 2002), after which a shell height of circa 2 cm is reached (de Lange et al., 1994; Weltje et al., 2003). L. stagnalis is a simultaneous hermaphrodite, which is able to self fertilise, but sexual reproduction is more common (Van Duivenboden, 1983). During mating only the snail taking the female role is fertilised. The gelatinous egg masses (see Fig. 1) are fixed on stones, water plants or on the shells of other pond snails. Between 2 and 109 eggs per egg mass were counted, whereby the egg number correlates positively with the size of the parent (see Fig. 2).
Fig. 1. Egg mass of Lymnaea stagnalis; black circles indicate two double-fertilised eggs.
For evaluation of the embryonic growth a single egg from a few minutes old egg mass was isolated. The embryo was observed in 24 ± 1 h intervals (see Fig. 3) under a stereo microscope and its size was measured. To determine its growth, measurements were made with the help of a software programme (AxioCam MRc; Carl Zeiss AG, Oberkochen, Germany). During the first days, the size of the whole embryo was recorded. After the 4th day, the shell diameter was measured (veliger-stage, see Fig. 3C). 2.4. Test conditions For the embryo toxicity test, 5 to 10 freshly laid egg masses were collected from the culture. The egg masses were not older than 24 h. With the help of a scalpel and a disposable pipette, eggs were isolated from their gelatinous matrix. With the scalpel the eggs were transferred randomly into the wells of a thoroughly deionised-water-rinsed 12-multiwell-plate; one egg per well containing
Fig. 2. Daily number of eggs per adult Lymnaea stagnalis against shell size (mm). Mean values of nine control replicates, each containing two to five individuals from four independent experiments (Pearson correlation r = 0.94) (unpublished data).
92
C. Bandow, L. Weltje / Science of the Total Environment 435–436 (2012) 90–95
Fig. 3. Embryonic development of Lymnaea stagnalis with its four stages; A) morula, B) trochophora, C) veliger, D) hippo; photos show daily development from day 0 to day 11.
3 mL M4 test medium. To check for possible abnormalities, for instance multi-fertilisation (see Fig. 1) or a damaged egg capsule, the eggs were examined under a stereo microscope directly after the transfer to the multiwell plate. If considered abnormal, the egg was replaced with a normal (healthy) one. The medium was renewed twice a week on Tuesday and Friday. To that end the old M4 medium was sucked out of the wells with a disposable pipette. Immediately thereafter the wells were filled with freshly prepared test solutions. The test was run under the same conditions as employed for culturing. 2.5. Observations After seven days of exposure, embryos were visually checked if they had reached the hippo-stage. As mentioned above, that is when shell, eyes and foot can clearly be distinguished and the heartbeat can be identified. From days 10 to 21, hatching, respectively mortality, was observed daily. An embryo was considered dead, when no heartbeat was visible. Between days 12 and 14 (i.e. the expected mean hatching time under control conditions) the hatching success was scored twice a day to give the results even more informative value. The observations were performed until day 21. As endpoints NOECs for hatching success and mean hatching time were defined. 2.6. Test substances As model test substance, tributyltin (TBT) was chosen, which was added as TBT chloride (CAS number 1461-22-9, Merck, purity>97%). The following nominal concentrations were tested: 0 (control), 0.03, 0.1, 0.3, 1.0, 3.0 and 10 μg TBT/L. A stock solution was prepared by adding 10 μL TBT-Cl to 1 L M4-medium without the use of a solvent; this stock solution was stirred for at least 10 min. The test solutions were prepared in 50 mL M4-medium by taking appropriate amounts of the stock solution and diluting them further. For each treatment 48 replicates (i.e. 4 plates) were used. All test concentrations mentioned in this paper are based on nominal cationic TBT. Chemical analysis of TBT was not conducted, since losses of the test substance due to degradation were considered negligible (although some sorption to the test vessels may occur). TBT chloride in water is
expected to dissociate to TBT cations (Eng et al., 1986). TBT is susceptible to biodegradation in water with half-lives of between 6 days and 35 weeks reported in water and water–sediment mixtures (Cooney, 1988; Maguire and Tkacz, 1985). Due to the medium renewal twice a week (on Tuesday and Friday), it can be assumed that exposure was maintained. Finally, this study was targeting to develop a new method for chemical testing on mollusc embryos and not to generate data for risk assessment. For aquatic toxicity testing of chemicals with low water solubility it can be necessary to use organic solvents. Therefore, it was tested whether solvents might influence the embryonic development of L. stagnalis. Hence, solvents recommended in various aquatic toxicity testing guidelines were assayed at their maximum allowed concentration, i.e. 100 mg/L respectively 100 μL/L (Organisation for Economic Cooperation and Development, 2000) in addition. Based upon OECD guidelines 202 (Organisation for Economic Co-operation and Development, 2004a), 211 (Organisation for Economic Co-operation and Development, 2008), 218 (Organisation for Economic Co-operation and Development, 2004b) and 219 (Organisation for Economic Co-operation and Development, 2004c) for testing of chemicals and the OECD guidance document no. 23 (Organisation for Economic Co-operation and Development, 2000), the following solvents were tested: methanol, ethanol, acetone, dimethyl sulfoxide (DMSO), triethylene glycol (TEG), ethylene glycol monoethyl ether (EGME) and ethylene glycol dimethyl ether (EGDE). Therefore, 24 embryos were exposed individually in two 12-multi well-plates to the maximum allowed concentration. Depending on the density of the tested solvent either 100 mg/L or 100 μL/L was chosen, whichever was the lowest. Hence, DMSO and TEG, whose densities exceed 1 g/cm³, were tested at 100 mg/L. The other solvents were tested at a concentration of 100 μL/L. A water control was included for comparison. 2.7. Statistical analysis All statistical computations were performed using GraphPad Prism, Version 5.01 for Windows (GraphPad Software, San Diego, California, USA). The data were analysed for normality with D'Agostino and Pearson omnibus normality test (α = 0.05). Since the control group did not pass the test for homogeneity of variance, the non-parametric
C. Bandow, L. Weltje / Science of the Total Environment 435–436 (2012) 90–95
93
Kruskal–Wallis test followed by Dunn's multiple comparison test was used to compare the differences among all treatment groups and so determine the NOEC for mean hatching time. The NOEC concerning hatching success was determined with Fisher's exact test. The concentration-response-curve was modelled with a cumulative logistic curve to establish the LC50. The embryonic growth was described with an exponential growth equation. The latter two calculations were done by non-linear fitting procedures using the Levenberg–Marquardt algorithm. 3. Results 3.1. Growth and hatching of embryos in the control To validate the test method, the mean hatching times of six control groups of independent experiments were compared (Fig. 4). On average the mean hatching time of the controls was 13.3 days. The comparison of these independent experiments shows some variation. However, the minimum and maximum value of the arithmetic mean hatching time differs circa 24 h. This maximum observed difference can at least partly be explained by the varying age of the egg masses at the beginning of the experiment. In fact, it is not distinguishable, whether an egg mass has an age of almost one day or whether it is freshly laid. Due to the above-mentioned hatching time in the controls 21 days as the total test duration seem to be adequate. Further, the size of a control embryo was recorded daily. The data showed an exponential curve for embryonic growth (see Fig. 5). It reveals that the embryonic diameter doubles within 3.2 days. The growth rate was 0.21 per day. On day 0, the embryo had a size of 135 μm. After hatching on the 12th day it measured 1736 μm. 3.2. Embryo toxicity test with TBT
Fig. 5. Embryo size (μm) until hatching versus time (d) under control conditions; dashed line: exponential growth equation, r2 ≥ 0.97.
4. Discussion 4.1. Feasibility of the embryo toxicity test Exposing isolated Lymnaea stagnalis eggs individually proves to be a suitable method for generating chronic toxicity endpoints on snail development. Embryo toxicity tests with L. stagnalis have previously been conducted by Boon-Niermeijer et al. (2000) and Canton and Slooff (1977) but in a design that was less suitable for standardisation e.g. because egg masses were used. But nevertheless, working with isolated eggs offers various advantages. First, abnormal (e.g. damaged, multi-fertilised or infertile) eggs can be excluded from the experiment immediately at the start. As mentioned above, eggs can be multi-fertilised (see Fig. 1) and up to 8 embryos in one egg have been reported by Lanzer (1999). Also, if a limited number of egg
The cumulated hatch shows a clear relationship (Fig. 6A). The three highest concentrations, namely 1.0, 3.0 and 10 μg TBT/L, lead to 100% mortality. Even at 0.3 μg TBT/L hatching was significantly reduced. Consequently the NOEC for hatching success was 0.1 μg TBT/L. The calculated LC50 was 0.36 μg TBT/L (95% confidence interval 0.28–0.46 μg TBT/L). Besides this, there was a significant delay in hatching in the treatments of 0.1 and 0.3 μg TBT/L, which is shown in Fig. 6B. Hence, the NOEC for mean hatching time was 0.03 μg TBT/L. 3.3. Toxicity of organic solvents in a limit test Considering the hatching success, no significant differences were observed between the treatments and the control. However, the mean hatching time showed a significant reduction (i.e. faster development) for all glycol containing solvents: TEG, EGME and EGDE (see Fig. 7).
Fig. 4. Mean hatching time with the standard deviation [d] of control treatments of six independent experiments (12 b n b 48); dashed lines are the range of minimum and maximum average values.
Fig. 6. A) Cumulative hatch of Lymnaea stagnalis embryos exposed to TBT (μg TBT/L nominal values) in ovo against time (days); dashed lines are cumulative logistic curves, r² ≥ 0.99, B) Mean hatching time with standard deviation [d] versus nominal TBT-concentration (μg TBT/L); asterisks indicate significant difference from control using Kruskal–Wallis test followed by Dunn's multiple comparison test (* = p b 0.05; *** = pb 0.001) (31 b n b 46).
94
C. Bandow, L. Weltje / Science of the Total Environment 435–436 (2012) 90–95
Fig. 7. Mean hatching time with standard deviation [d] of Lymnaea stagnalis embryos exposed in ovo to various solvents at nominal 100 mg/L L and 100 μL/L, respectively; asterisks indicate significant difference from control using Kruskal–Wallis test followed by Dunn's multiple comparison test (* = p b 0.05; ** = p b 0.01; n = 24; exception: methanol, n = 12).
masses are used, genetic variation is strongly reduced, as proposed by Sawasdee and Köhler (2009) who used a similar method of studying individual eggs of the ramshorn snail Marisa cornuarietis. Finally, an influence of the position of the egg within the egg mass can be avoided. The egg mass consists of gelatine, which protects the eggs. Depending on the location of the egg in the egg mass the protection level offered by the exterior gelatine matrix could vary (Gomot, 1998). Obviously eggs are not isolated under natural exposure conditions, so this test design may simulate the worst case for chemical exposure, possibly leading to an overestimation of toxicity in comparison to a more natural exposure under higher densities (Leung et al., 2004). Otherwise, the protective function of the gelatine will also depend on the chemical. According to physical and chemical characteristics of the substance (for instance logPow or pKS) the penetrating power of the substance into the egg mass could be intensified or inhibited. All in all it cannot be assumed that the protective function of the gelatine is constantly equal. Whether the embryo is protected by the egg capsule itself and how could potentially be investigated in tests where capsules are removed so that the embryo is exposed bare. Similar studies on zebrafish Danio rerio eggs and embryos are already available (Braunbeck et al., 2005; Scholz et al., 2008). The determined growth rate of embryonic development was 0.21 per day. Grosell and Brix (2009) studied the growth rate of 48 h aged L. stagnalis for 38 days. They found a similar growth rate of 0.28 per day. Also for hatching time confirmatory data could be found in the literature. Taylor (1977) reported a hatching time of about 11 days at 25 °C, which is obviously shorter due to the higher temperature, but still fits very well to the data obtained in this work at 20 °C. The control mortality in this work was less than 5%, which is well below the typical validity criteria for a chronic test. Sawasdee and Köhler (2009) regarded a mortality in the controls of less than 15% as valid in their embryo toxicity test with M. cornuarietis.
4.2. Effects of TBT and organic solvents on Lymnaea stagnalis Leung et al. (2007) tested TBT-oxide in a life cycle test with L. stagnalis. Whole egg masses were used and exposed to 0.01, 1.0 and 10 μg TBT/L. In the highest concentration, which equates to the maximum concentration in our experiments, the mortality was 100%. In the case of 1.0 μg TBT/L Leung et al. (2007) assessed a non‐significant mortality of circa 30%, however at a comparatively high snail density. In the present study the mortality was still 100% at this concentration. The maximum concentration of 10 μg TBTO/L was also tested by Morley et al. (2004) in an investigation on adults of another pond snail,
Lymnaea peregra. The adults of this species react as sensitively as the embryos of L. stagnalis, with 100% mortality. In a study by Mathijssen-Spiekman et al. (1989) chronic toxicity of TBT-oxide in L. stagnalis was examined. The NOEC of hatching success was 0.32 μg TBT/L, which is very similar to the NOEC of the present research, i.e. 0.1 μg TBT/L. In the 0.3 μg TBT/L treatment the embryos were significantly delayed in development, so that hatching was not finalised after 21 days. It is probable, that an extended test duration would have led to further hatching success, which would increase the NOEC for this endpoint. Nevertheless, the NOEC of the mean hatching time of the present study (0.03 μg TBT/L) was lower than the one reported by Mathijssen-Spiekman et al. (1989), who assessed 17% of the effects after 9 days exposure of embryos in 0.1 μg TBT/L. Information on the toxicity of TBT to other gastropod molluscs can be found in Leung et al. (2004) and Duft et al. (2005). Exposing egg masses of the pulmonate snail Physa fontinalis, Leung et al. (2004) determined a NOEC and LOEC of 0.01 and 1.0 μg TBT/L, respectively. Duft et al. (2005) developed a sediment toxicity test with the parthenogenetic ovoviviparous snail Potamopyrgus antipodarum. The LOEC for juvenile production after 4 weeks was 24.4 μg TBT/kg dw, which was the lowest concentration. Hence, a NOEC could not be determined. The EC10 was calculated as 2.39 μg TBT/kg dw. Due to the differences in employed test designs, e.g. a spacing factor of 10 (Leung et al., 2004) and sediment exposure (Duft et al., 2005), a comparison of species sensitivities is hampered. Regarding the different solvents, it seems that there could be a stimulating influence of some of these substances (i.e. those containing glycol) on embryo development, albeit at relatively high concentrations. Yang et al. (2008) found such an effect on larvae of the mussel Mytilus galloprovincialis when exposed to various solvents, including ethylene glycol. At present it is not understood if the cause for these effects is of a direct or indirect nature. Comprehensive data regarding solvent effects on aquatic organisms, resulting both in stimulating and inhibiting effects, have been reviewed by Hutchinson et al. (2006). Nonetheless studies on molluscs, especially snails, are rare. In further studies with more replicates and lower concentrations, the influence of these solvents should be confirmed and assessed clearly.
4.3. Embryo testing with other mollusc species There are two acute standard toxicity tests with mollusc embryos, employing marine bivalve species of commercial interest. One test was described by Thain (1991) for the International Council for the Exploration of the Sea (ICES). In this 24-h test the development of embryos of the oyster Crassostrea gigas to the shelled larval stage is examined. The American Society for Testing and Materials (ASTM, 2004) also published a standard test describing a 48-h test on embryos and the resulting larvae of four species of saltwater bivalve molluscs (Pacific oyster, Crassostrea gigas; eastern oyster, Crassostrea virginica; hard clam, Mercenaria mercenaria; and blue mussel, Mytilus edulis). In contrast to L. stagnalis, which produces egg masses, the gametes of bivalves have to be collected and then be fertilised extra corporal. This biological attribute complicates the test, despite its short exposure duration. Embryos of other pulmonate mollusc species have been employed in toxicity tests as well, e.g. Physa acuta (Cheung and Lam, 1998). These authors demonstrated that embryos were slightly more sensitive to cadmium than juvenile snails and suggested also that there was a role for early life stage testing of pulmonate snails in ecotoxicological testing. Obviously, there is a specific interest to develop tests with eggs/ embryos of fish as these do not qualify as vertebrate tests before the fish start eating independently (Braunbeck et al., 2005; OECD 212, 1998; Scholz et al., 2008). When further data for molluscs become available, it would be interesting to systematically study the sensitivity differences between fish and molluscs. A comparison of test results between fish and Marisa cornuarietis embryo toxicity tests showed a comparable
C. Bandow, L. Weltje / Science of the Total Environment 435–436 (2012) 90–95
sensitivity for lithium and copper, while fish were more sensitive to lead and palladium (Sawasdee and Köhler, 2010). 5. Conclusion Considering all data it seems that this method of chemical testing is rather simple for assessing embryo toxicity. For a single 12 well plate a volume of only 36 mL of medium is required at each renewal. Hence, costs for chemicals and laboratory materials are low. Furthermore the multiwell plates need little space and are easily handled. With 21 days the test has a comparatively short duration to achieve explicit and sensitive endpoints as was demonstrated by means of the reference toxicant TBT. In short, this test appears to be adequate to become a part of the aspired guideline for mollusc life cycle toxicity test within the OECD or alternately to serve as a trigger to decide upon the necessity of a (partial) life-cycle test. Acknowledgements The authors wish to thank Jörg Oehlmann (Goethe-University Frankfurt/Main, Germany) and three anonymous reviewers for their helpful advice and comments on a previous version of this paper. This work originates from the diploma thesis of C.B., under supervision of Dieter Greif (University of Applied Sciences Zittau/Görlitz). References ASTM Standard E724 - 98. Standard Guide for Conducting Static Acute Toxicity Tests Starting with Embryos of Four Species of Saltwater Bivalve Molluscs. West Conshohocken, USA: ASTM International; 2004. Available from: www.astm.org. Boon-Niermeijer EK, van den Berg A, Wikman G, Wiegant FAC. Phyto-adaptogens protect against environmental stress-induced death of embryos from the freshwater snail Lymnaea stagnalis. Phytomedicine 2000;7(5):389–99. Braunbeck T, Böttcher M, Hollert H, Kosmehl T, Lammer E, Leist E, et al. Towards an alternative for the acute fish LC50 test in chemical assessment: the fish embryo toxicity test goes multi-species—an update. ALTEX 2005;22(2):87-102. Canton JH, Slooff W. The usefulness of Lymnaea stagnalis L. as a biological indicator in toxicological bio-assays (model substance α-HCH). Water Res 1977;11(1):117–21. Cheung CCC, Lam PKS. Effect of cadmium on the embryos and juveniles of a tropical freshwater snail, Physa acuta (Draparnaud, 1805). Water Sci Technol 1998;38(7): 263–70. Cooney JJ. Microbial transformations of tin and tin compounds. J Ind Microbiol 1988;3(4): 195–204. [as citied in Hazardous Substances Data Bank [2012 May 30]. Available from: http://toxnet.nlm.nih.gov/cgi-bin/sis/search/f?./temp/~8tUXHU:1]. De Lange RPJ, van Minnen J, Boer HH. Expression and translation of the egg-laying neuropeptide hormone genes during post-embryonic development of the pond snail Lymnaea stagnalis. Cell Tissue Res 1994;275(2):369–75. Duft M, Schulte-Oehlmann U, Tillmann M, Weltje L, Oehlmann J. Biological impact of organotin compounds on molluscs in marine and freshwater ecosystems. Coast Mar Sci 2005;29(2):95-110. Elendt B-P, Bias W-R. Trace nutrient deficiency in Daphnia magna cultured in standard medium for toxicity testing. Effects of the optimization of culture conditions on life history parameters of D. magna. Water Res 1990;24(9):1157–67. Eng G, Bathersfield O, May L. Mössbauer studies of the speciation of tributyltin compounds in seawater and sediment samples. Water Air Soil Pollut 1986;27(1–2): 191–7. [as citied in Hazardous Substances Data Bank [2012 May 30]. Available from: http://toxnet.nlm.nih.gov/cgi-bin/sis/search/f?./temp/~8tUXHU:1]. Glöer P, Meier-Brook C, Ostermann O. Süßwassermollusken, ein Bestimmungsschlüssel für die Bundesrepublik Deutschland (Freshwater molluscs, a determination key for the Federal Republic of Germany). 10th ed. Hamburg, Germany: Deutscher Jugendbund für Naturbeobachtung; 1992 [In German]. Gomot A. Toxic effects of cadmium on reproduction, development, and hatching in the freshwater snail Lymnaea stagnalis for water quality monitoring. Ecotoxicol Environ Saf 1998;41(3):288–97. Grosell M, Brix KV. High net calcium uptake explains the hypersensitivity of the freshwater pulmonate snail, Lymnaea stagnalis, to chronic lead exposure. Aquat Toxicol 2009;91(4):302–11. Grosell M, Gerdes RM, Brix KV. Chronic toxicity of lead to three freshwater invertebrates—Brachionus calyciflorus, Chironomus tentans and Lymnaea stagnalis. Environ Toxicol Chem 2006;25(1):97-104.
95
Hutchinson TH, Shillabeer N, Winter MJ, Pickford DB. Acute and chronic effects of carrier solvents in aquatic organisms: a critical review. Aquat Toxicol 2006;76(1): 69–92. Lalah JO, Severin GF, Schramm KW, Lenoir D, Behechti A, Guenther K, et al. Effects of a branched p-nonylphenol isomer (4(3′,6′-dimethyl-3′-heptyl)-phenol) on embryogenesis in Lymnaea stagnalis L. Arch Environ Contam Toxicol 2007;52(1):104–12. Lanzer R. GSF—National Research Center for Environment and Health. Toxicity of nonylphenol during embryo growth in Lymnaea stagnalis L. GSF Report. Munich, Germany; 1999. Leung KMY, Morley NJ, Grist EPM, Morritt D, Crane M. Chronic toxicity of tributyltin on development and reproduction of the hermaphroditic snail Physa fontinalis: influence of population density. Mar Environ Res 2004;58(2–5):157–62. Leung KMY, Grist EPM, Morley NJ, Morritt D, Crane M. Chronic toxicity of tributyltin to development and reproduction of the European freshwater snail Lymnaea stagnalis (L.). Chemosphere 2007;66(7):1358–66. Maguire RJ, Tkacz RJ. Degradation of the tri-n-butyltin species in water and sediment from Toronto Harbor. J Agric Food Chem 1985;33(5):947–53. [as citied in Hazardous Substances Data Bank [2012 May 30]. Available from: http://toxnet.nlm.nih. gov/cgi-bin/sis/search/f?./temp/~8tUXHU:1]. Mathijssen-Spiekman EAM, Canton JH, Roghair CJ. Dutch National Institute for Public Health and the Environment. Onderzoek naar de toxiciteit van TBTO voor een aantal zoetwaterorganismen (Research into the toxicity of TBTO for a number of freshwater organisms). Bilthoven, The Netherlands: RIVM; 1989 [Report No.: 668118 001. In Dutch]. Matthiessen P. An assessment of endocrine disruption in mollusks and the potential for developing internationally standardized mollusk life cycle test guidelines. Integr Environ Assess Manag 2008;4(3):274–84. Morley NJ, Leung KMY, Morritt D, Crane M. Toxicity of anti-fouling biocides to encysted metacercariae of Echinoparyphium recurvatum (Digenea: Echinostomatidae) and their snail hosts. Chemosphere 2004;56(4):353–8. Oehlmann J, Schulte-Oehlmann U. Endocrine disruption in invertebrates. Pure Appl Chem 2003;75(11):2207–18. Organisation for Economic Co-operation and Development. OECD guideline for testing of chemicals. OECD 212 fish, short-term toxicity test on embryo and sac-fry stages. Paris, France; 1998. Organisation for Economic Co-operation and Development. OECD series on testing and assessment, number 23. Guidance document on aquatic toxicity testing of difficult substances and mixtures. Paris, France; 2000. Organisation for Economic Co-operation and Development. OECD (2004a): OECD guidelines for the testing of chemicals. OECD 202 Daphnia sp., Acute immobilisation test. Paris, France; 2004a. Organisation for Economic Co-operation and Development. OECD Guidelines for the testing of chemicals. OECD 218 Sediment-water chironomid toxicity test using spiked sediment. Paris, France; 2004b. Organisation for Economic Co-operation and Development. OECD Guidelines for the testing of chemicals. OECD 219 Sediment-water chironomid toxicity test using spiked water. Paris, France; 2004c. Organisation for Economic Co-operation and Development. OECD Guidelines for the testing of chemicals. OECD 211 Daphnia magna Reproduction test. Paris, France; 2008. Organisation for Economic Co-operation and Development. Detailed Review Paper on Mollusc Life-Cycle Toxicity Testing. Paris, France; 2010. Sawasdee B, Köhler H-R. Embryo toxicity of pesticides and heavy metals to the ramshorn snail, Marisa cornuarietis (Prosobranchia). Chemosphere 2009;75(11):1539–47. Sawasdee B, Köhler H-R. Metal sensitivity of the embryonic development of the ramshorn snail Marisa cornuarietis (Prosobranchia). Ecotoxicology 2010;19(8):1487–95. Scholz S, Fischer S, Gündel U, Küster E, Luckenbach T, Voelker D. The zebrafish embryo model in environmental risk assessment – applications beyond acute toxicity testing. Environ Sci Pollut Res Int 2008;15(5):394–404. Schwab H. Süßwassertiere, ein ökologisches Bestimmungsbuch (Freshwater organisms, an ecological determination guide). 1st ed. Stuttgart, Germany: Ernst Klett Verlag; 2002 [In German]. Taylor HH. The ionic and water relations of embryos of Lymnaea stagnalis, a freshwater pulmonate mollusc. J Exp Biol 1977;69:143–72. Thain JE. International Council for the Exploration of the Sea. Biological effects of contaminants: Oyster (Crassostrea gigas) embryo bioassay, 11. Techniques in marine environmental sciences; 1991.. [Copenhagen, Denmark]. Van Duivenboden YA. Transfer of semen accelerates the onset of egg-laying in female copulants of the hermaphrodite freshwater snail, Lymnaea stagnalis. Int J Invert Reprod 1983;6(4):249–57. Weltje L, Scholz C, Stark C, Ziebart S, van Doornmalen J, Oehlmann J. Endocrine disruption in the freshwater pondsnail Lymnaea stagnalis L. Poster presentation of the 13th annual meeting of SETAC Europe; 2003 Apr 27-May 1; Hamburg, Germany; 2003. Yang J-L, Satuito CG, Bao W-Y, Kitamura H. Induction of Metamorphosis of Pediveliger Larvae of the Mussel Mytilus galloprovincialis Lamarck, 1819 using Neuroactive Compounds, KCl, NH4Cl and Organic Solvents. Biofouling 2008;24(6):461–70.
Contents lists available at SciVerse ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Development of an embryo toxicity test with the pond snail Lymnaea stagnalis using the model substance tributyltin and common solvents☆ Cornelia Bandow a, b,⁎, Lennart Weltje c a b c
University of Applied Sciences Zittau/Görlitz, Theodor-Körner-Allee 16, D-02763 Zittau, Germany LOEWE Biodiversity and Climate Research Centre (BiKF), Senckenberganlage 25, D‐60325 Frankfurt am Main, Germany BASF SE, Crop Protection, Ecotoxicology, Speyerer Strasse 2, D-67117 Limburgerhof, Germany
H I G H L I G H T S ► ► ► ►
The pond snail Lymnaea stagnalis is a suitable organism for ecotoxicological testing. A developmental toxicity test using L. stagnalis embryos is completed within 21 days. Sensitive endpoints are easily assessed. Low effort regarding human and financial resources.
a r t i c l e
i n f o
Article history: Received 19 March 2012 Received in revised form 2 July 2012 Accepted 2 July 2012 Available online 28 July 2012 Keywords: Mollusc Early life stages Organotin Pulmonate Embryonic growth Test guideline
a b s t r a c t The development of a chronic mollusc toxicity test is a current work item on the agenda of the OECD. The freshwater pond snail Lymnaea stagnalis is one of the candidate snail species for such a test. This paper presents a 21-day chronic toxicity test with L. stagnalis, focussing on embryonic development. Eggs were collected from freshly laid egg masses and exposed individually until hatching. The endpoints were hatching success and mean hatching time. Tributyltin (TBT), added as TBT-chloride, was chosen as model substance. The selected exposure concentrations ranged from 0.03 to 10 μg TBT/L (all as nominal values) and induced the full range of responses. The embryos were sensitive to TBT (the NOEC for mean hatching time was 0.03 μg TBT/L and the NOEC for hatching success was 0.1 μg TBT/L). In addition, data on maximum limit concentrations of seven common solvents, recommended in OECD aquatic toxicity testing guidelines, are presented. Among the results, further findings as average embryonic growth and mean hatching time of control groups are provided. In conclusion, the test presented here could easily be standardised and is considered useful as a potential trigger to judge if further studies, e.g. a (partial) life‐cycle study with molluscs, should be conducted. © 2012 Elsevier B.V. All rights reserved.
1. Introduction In 2010 a detailed review paper was published by the OECD, exploring the possibilities for mollusc life cycle testing with a focus on endocrine disruptors and other chemicals (Organisation for Economic Cooperation and Development, 2010). The need for (chronic) mollusc testing is partly motivated by the well documented effects of tributyltin (TBT) and triphenyltin (TPT) compounds on molluscs. In the past, effects of these compounds, used in antifouling paints on ship hulls, were recognised too late, which led to a threat for many marine mollusc populations. TBT and TPT compounds affect the endocrine system of
☆ This paper is dedicated to the memory of Thomas Knacker. ⁎ Corresponding author at: ECT Oekotoxikologie GmbH, Böttgerstrasse 2‐14, D-65439 Flörsheim, Germany. Tel.: +49 6145 9564-61; fax: +49 6145 9564-99. E-mail addresses: [email protected] (C. Bandow), [email protected] (L. Weltje). 0048-9697/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2012.07.005
molluscs and may cause effects known as imposex or intersex at very low concentrations (Oehlmann and Schulte-Oehlmann, 2003). In addition, Matthiessen (2008) refers to the ecological importance of this large and taxonomically diverse group of invertebrates and their under representation in regulatory aquatic testing. Further, Grosell et al. (2006) concluded that aquatic snails might not be protected under current regulatory guidelines. Their conclusion was derived from chronic toxicity studies on lead with freshwater invertebrates, i.e. rotifers Brachionus calyciflorus, non-biting midges Chironomus tentans and hatchlings of Lymnaea stagnalis, where L. stagnalis showed the highest sensitivity. In this paper, a method for conducting an embryo toxicity test with the pond snail is described. Testing embryos is particularly relevant as adverse effects of test compounds on critical developmental processes such as organogenesis can be studied. Also, due to L. stagnalis' high production rate of egg masses an embryo test can be standardised to a high extent. An embryo test might become a part of the mollusc life cycle test, which is currently under development by the OECD Endocrine Disrupter Testing
C. Bandow, L. Weltje / Science of the Total Environment 435–436 (2012) 90–95
and Assessment group (Organisation for Economic Co-operation and Development, 2010). In the detailed review paper on mollusc lifecycle toxicity testing, the pond snail L. stagnalis is proposed as a possible test species for test development, next to the prosobranch mollusc Potamopyrgus antipodarum and the marine bivalve Crassostrea gigas. L. stagnalis is well suited as a standard test organism: it is distributed holarctic and because of its halotolerance of up to 0.7% it is also found in the Baltic Sea (Glöer et al., 1992; Schwab, 2002). Further, breeding in the laboratory is relatively easy. Due to its considerable size, i.e. the maximum shell height is 7 cm; histopathological and biochemical analyses (e.g. on the haemolymph) could even be conducted. Maturity occurs after approximately 10 weeks at 20 °C. Furthermore, development of the embryos occurs in eggs outside the adult snail contrary to P. antipodarum, whose embryos stay in the brood pouch, or C. gigas, which produces planktonic larval stages. Because of the transparent egg capsule embryo development can be observed at any time. Being a simultaneous hermaphrodite, parental sex ratio in starting reproduction tests is not an issue. Due to the characteristics outlined above, L. stagnalis offers ample possibilities for mollusc tests, including a full life-cycle test (Leung et al., 2007), which would cover potential endocrine mediated effects, as well as the embryo toxicity test described in this paper, which may serve as a screening tool for assessing chronic toxic potential against molluscs within a short time.
2. Material and methods
91
In the eggs, embryos go through four different stages (Canton and Slooff, 1977; Lalah et al., 2007), compare Fig. 3: a) Morula: days 0 to 3, yellowish appearance, slow movement. b) Trochophora: days 3 to 5, grainy appearance, uninterrupted movement. c) Veliger: days 5 to 7, the embryo loses its round appearance, foot and shell can be distinguished. d) Hippo: from day 7 the embryo is fully developed; eyes, shell and foot are differentiated, heart beat is visible. Depending on the temperature, snails hatch between day 10 and day 20. At 20±2 °C the mean hatching time in our laboratory was between 12 and 14 days. 2.2. Culturing Snails were held in aquaria of different sizes (from 6 up to 36 L) under constant aeration. As a culture medium the M4-medium (Elendt and Bias, 1990; Organisation for Economic Co-operation and Development, 2004a) was used. The physical and chemical parameters of the water fulfilled the requirements of the OECD test guideline 202 (Organisation for Economic Co-operation and Development, 2004a). The culture was maintained in a climatised room at 20±2 °C and a 16:8 h light–dark-rhythm. L. stagnalis was fed ad libitum with different kinds of vegetables, e.g. cucumber, lettuce, carrot or courgette. Furthermore, cuttlebones were given in each aquarium to support shell growth. Faeces and remaining food was removed 5 to 6 times a week. In addition, periphyton on the glass walls was removed every week.
2.1. Test organism 2.3. Embryonic growth With its shell height of up to 7 cm, the great pond snail Lymnaea stagnalis is the largest aquatic snail domestic in Europe and Northern America. It is found in freshwater as well as in brackish water. The pond snail is a detritivore and feeds on plants, carrion, algae and microorganisms (Glöer et al., 1992; Schwab, 2002). The development of a newly hatched juvenile to an adult snail takes about 10 weeks at 20 °C (Schwab, 2002), after which a shell height of circa 2 cm is reached (de Lange et al., 1994; Weltje et al., 2003). L. stagnalis is a simultaneous hermaphrodite, which is able to self fertilise, but sexual reproduction is more common (Van Duivenboden, 1983). During mating only the snail taking the female role is fertilised. The gelatinous egg masses (see Fig. 1) are fixed on stones, water plants or on the shells of other pond snails. Between 2 and 109 eggs per egg mass were counted, whereby the egg number correlates positively with the size of the parent (see Fig. 2).
Fig. 1. Egg mass of Lymnaea stagnalis; black circles indicate two double-fertilised eggs.
For evaluation of the embryonic growth a single egg from a few minutes old egg mass was isolated. The embryo was observed in 24 ± 1 h intervals (see Fig. 3) under a stereo microscope and its size was measured. To determine its growth, measurements were made with the help of a software programme (AxioCam MRc; Carl Zeiss AG, Oberkochen, Germany). During the first days, the size of the whole embryo was recorded. After the 4th day, the shell diameter was measured (veliger-stage, see Fig. 3C). 2.4. Test conditions For the embryo toxicity test, 5 to 10 freshly laid egg masses were collected from the culture. The egg masses were not older than 24 h. With the help of a scalpel and a disposable pipette, eggs were isolated from their gelatinous matrix. With the scalpel the eggs were transferred randomly into the wells of a thoroughly deionised-water-rinsed 12-multiwell-plate; one egg per well containing
Fig. 2. Daily number of eggs per adult Lymnaea stagnalis against shell size (mm). Mean values of nine control replicates, each containing two to five individuals from four independent experiments (Pearson correlation r = 0.94) (unpublished data).
92
C. Bandow, L. Weltje / Science of the Total Environment 435–436 (2012) 90–95
Fig. 3. Embryonic development of Lymnaea stagnalis with its four stages; A) morula, B) trochophora, C) veliger, D) hippo; photos show daily development from day 0 to day 11.
3 mL M4 test medium. To check for possible abnormalities, for instance multi-fertilisation (see Fig. 1) or a damaged egg capsule, the eggs were examined under a stereo microscope directly after the transfer to the multiwell plate. If considered abnormal, the egg was replaced with a normal (healthy) one. The medium was renewed twice a week on Tuesday and Friday. To that end the old M4 medium was sucked out of the wells with a disposable pipette. Immediately thereafter the wells were filled with freshly prepared test solutions. The test was run under the same conditions as employed for culturing. 2.5. Observations After seven days of exposure, embryos were visually checked if they had reached the hippo-stage. As mentioned above, that is when shell, eyes and foot can clearly be distinguished and the heartbeat can be identified. From days 10 to 21, hatching, respectively mortality, was observed daily. An embryo was considered dead, when no heartbeat was visible. Between days 12 and 14 (i.e. the expected mean hatching time under control conditions) the hatching success was scored twice a day to give the results even more informative value. The observations were performed until day 21. As endpoints NOECs for hatching success and mean hatching time were defined. 2.6. Test substances As model test substance, tributyltin (TBT) was chosen, which was added as TBT chloride (CAS number 1461-22-9, Merck, purity>97%). The following nominal concentrations were tested: 0 (control), 0.03, 0.1, 0.3, 1.0, 3.0 and 10 μg TBT/L. A stock solution was prepared by adding 10 μL TBT-Cl to 1 L M4-medium without the use of a solvent; this stock solution was stirred for at least 10 min. The test solutions were prepared in 50 mL M4-medium by taking appropriate amounts of the stock solution and diluting them further. For each treatment 48 replicates (i.e. 4 plates) were used. All test concentrations mentioned in this paper are based on nominal cationic TBT. Chemical analysis of TBT was not conducted, since losses of the test substance due to degradation were considered negligible (although some sorption to the test vessels may occur). TBT chloride in water is
expected to dissociate to TBT cations (Eng et al., 1986). TBT is susceptible to biodegradation in water with half-lives of between 6 days and 35 weeks reported in water and water–sediment mixtures (Cooney, 1988; Maguire and Tkacz, 1985). Due to the medium renewal twice a week (on Tuesday and Friday), it can be assumed that exposure was maintained. Finally, this study was targeting to develop a new method for chemical testing on mollusc embryos and not to generate data for risk assessment. For aquatic toxicity testing of chemicals with low water solubility it can be necessary to use organic solvents. Therefore, it was tested whether solvents might influence the embryonic development of L. stagnalis. Hence, solvents recommended in various aquatic toxicity testing guidelines were assayed at their maximum allowed concentration, i.e. 100 mg/L respectively 100 μL/L (Organisation for Economic Cooperation and Development, 2000) in addition. Based upon OECD guidelines 202 (Organisation for Economic Co-operation and Development, 2004a), 211 (Organisation for Economic Co-operation and Development, 2008), 218 (Organisation for Economic Co-operation and Development, 2004b) and 219 (Organisation for Economic Co-operation and Development, 2004c) for testing of chemicals and the OECD guidance document no. 23 (Organisation for Economic Co-operation and Development, 2000), the following solvents were tested: methanol, ethanol, acetone, dimethyl sulfoxide (DMSO), triethylene glycol (TEG), ethylene glycol monoethyl ether (EGME) and ethylene glycol dimethyl ether (EGDE). Therefore, 24 embryos were exposed individually in two 12-multi well-plates to the maximum allowed concentration. Depending on the density of the tested solvent either 100 mg/L or 100 μL/L was chosen, whichever was the lowest. Hence, DMSO and TEG, whose densities exceed 1 g/cm³, were tested at 100 mg/L. The other solvents were tested at a concentration of 100 μL/L. A water control was included for comparison. 2.7. Statistical analysis All statistical computations were performed using GraphPad Prism, Version 5.01 for Windows (GraphPad Software, San Diego, California, USA). The data were analysed for normality with D'Agostino and Pearson omnibus normality test (α = 0.05). Since the control group did not pass the test for homogeneity of variance, the non-parametric
C. Bandow, L. Weltje / Science of the Total Environment 435–436 (2012) 90–95
93
Kruskal–Wallis test followed by Dunn's multiple comparison test was used to compare the differences among all treatment groups and so determine the NOEC for mean hatching time. The NOEC concerning hatching success was determined with Fisher's exact test. The concentration-response-curve was modelled with a cumulative logistic curve to establish the LC50. The embryonic growth was described with an exponential growth equation. The latter two calculations were done by non-linear fitting procedures using the Levenberg–Marquardt algorithm. 3. Results 3.1. Growth and hatching of embryos in the control To validate the test method, the mean hatching times of six control groups of independent experiments were compared (Fig. 4). On average the mean hatching time of the controls was 13.3 days. The comparison of these independent experiments shows some variation. However, the minimum and maximum value of the arithmetic mean hatching time differs circa 24 h. This maximum observed difference can at least partly be explained by the varying age of the egg masses at the beginning of the experiment. In fact, it is not distinguishable, whether an egg mass has an age of almost one day or whether it is freshly laid. Due to the above-mentioned hatching time in the controls 21 days as the total test duration seem to be adequate. Further, the size of a control embryo was recorded daily. The data showed an exponential curve for embryonic growth (see Fig. 5). It reveals that the embryonic diameter doubles within 3.2 days. The growth rate was 0.21 per day. On day 0, the embryo had a size of 135 μm. After hatching on the 12th day it measured 1736 μm. 3.2. Embryo toxicity test with TBT
Fig. 5. Embryo size (μm) until hatching versus time (d) under control conditions; dashed line: exponential growth equation, r2 ≥ 0.97.
4. Discussion 4.1. Feasibility of the embryo toxicity test Exposing isolated Lymnaea stagnalis eggs individually proves to be a suitable method for generating chronic toxicity endpoints on snail development. Embryo toxicity tests with L. stagnalis have previously been conducted by Boon-Niermeijer et al. (2000) and Canton and Slooff (1977) but in a design that was less suitable for standardisation e.g. because egg masses were used. But nevertheless, working with isolated eggs offers various advantages. First, abnormal (e.g. damaged, multi-fertilised or infertile) eggs can be excluded from the experiment immediately at the start. As mentioned above, eggs can be multi-fertilised (see Fig. 1) and up to 8 embryos in one egg have been reported by Lanzer (1999). Also, if a limited number of egg
The cumulated hatch shows a clear relationship (Fig. 6A). The three highest concentrations, namely 1.0, 3.0 and 10 μg TBT/L, lead to 100% mortality. Even at 0.3 μg TBT/L hatching was significantly reduced. Consequently the NOEC for hatching success was 0.1 μg TBT/L. The calculated LC50 was 0.36 μg TBT/L (95% confidence interval 0.28–0.46 μg TBT/L). Besides this, there was a significant delay in hatching in the treatments of 0.1 and 0.3 μg TBT/L, which is shown in Fig. 6B. Hence, the NOEC for mean hatching time was 0.03 μg TBT/L. 3.3. Toxicity of organic solvents in a limit test Considering the hatching success, no significant differences were observed between the treatments and the control. However, the mean hatching time showed a significant reduction (i.e. faster development) for all glycol containing solvents: TEG, EGME and EGDE (see Fig. 7).
Fig. 4. Mean hatching time with the standard deviation [d] of control treatments of six independent experiments (12 b n b 48); dashed lines are the range of minimum and maximum average values.
Fig. 6. A) Cumulative hatch of Lymnaea stagnalis embryos exposed to TBT (μg TBT/L nominal values) in ovo against time (days); dashed lines are cumulative logistic curves, r² ≥ 0.99, B) Mean hatching time with standard deviation [d] versus nominal TBT-concentration (μg TBT/L); asterisks indicate significant difference from control using Kruskal–Wallis test followed by Dunn's multiple comparison test (* = p b 0.05; *** = pb 0.001) (31 b n b 46).
94
C. Bandow, L. Weltje / Science of the Total Environment 435–436 (2012) 90–95
Fig. 7. Mean hatching time with standard deviation [d] of Lymnaea stagnalis embryos exposed in ovo to various solvents at nominal 100 mg/L L and 100 μL/L, respectively; asterisks indicate significant difference from control using Kruskal–Wallis test followed by Dunn's multiple comparison test (* = p b 0.05; ** = p b 0.01; n = 24; exception: methanol, n = 12).
masses are used, genetic variation is strongly reduced, as proposed by Sawasdee and Köhler (2009) who used a similar method of studying individual eggs of the ramshorn snail Marisa cornuarietis. Finally, an influence of the position of the egg within the egg mass can be avoided. The egg mass consists of gelatine, which protects the eggs. Depending on the location of the egg in the egg mass the protection level offered by the exterior gelatine matrix could vary (Gomot, 1998). Obviously eggs are not isolated under natural exposure conditions, so this test design may simulate the worst case for chemical exposure, possibly leading to an overestimation of toxicity in comparison to a more natural exposure under higher densities (Leung et al., 2004). Otherwise, the protective function of the gelatine will also depend on the chemical. According to physical and chemical characteristics of the substance (for instance logPow or pKS) the penetrating power of the substance into the egg mass could be intensified or inhibited. All in all it cannot be assumed that the protective function of the gelatine is constantly equal. Whether the embryo is protected by the egg capsule itself and how could potentially be investigated in tests where capsules are removed so that the embryo is exposed bare. Similar studies on zebrafish Danio rerio eggs and embryos are already available (Braunbeck et al., 2005; Scholz et al., 2008). The determined growth rate of embryonic development was 0.21 per day. Grosell and Brix (2009) studied the growth rate of 48 h aged L. stagnalis for 38 days. They found a similar growth rate of 0.28 per day. Also for hatching time confirmatory data could be found in the literature. Taylor (1977) reported a hatching time of about 11 days at 25 °C, which is obviously shorter due to the higher temperature, but still fits very well to the data obtained in this work at 20 °C. The control mortality in this work was less than 5%, which is well below the typical validity criteria for a chronic test. Sawasdee and Köhler (2009) regarded a mortality in the controls of less than 15% as valid in their embryo toxicity test with M. cornuarietis.
4.2. Effects of TBT and organic solvents on Lymnaea stagnalis Leung et al. (2007) tested TBT-oxide in a life cycle test with L. stagnalis. Whole egg masses were used and exposed to 0.01, 1.0 and 10 μg TBT/L. In the highest concentration, which equates to the maximum concentration in our experiments, the mortality was 100%. In the case of 1.0 μg TBT/L Leung et al. (2007) assessed a non‐significant mortality of circa 30%, however at a comparatively high snail density. In the present study the mortality was still 100% at this concentration. The maximum concentration of 10 μg TBTO/L was also tested by Morley et al. (2004) in an investigation on adults of another pond snail,
Lymnaea peregra. The adults of this species react as sensitively as the embryos of L. stagnalis, with 100% mortality. In a study by Mathijssen-Spiekman et al. (1989) chronic toxicity of TBT-oxide in L. stagnalis was examined. The NOEC of hatching success was 0.32 μg TBT/L, which is very similar to the NOEC of the present research, i.e. 0.1 μg TBT/L. In the 0.3 μg TBT/L treatment the embryos were significantly delayed in development, so that hatching was not finalised after 21 days. It is probable, that an extended test duration would have led to further hatching success, which would increase the NOEC for this endpoint. Nevertheless, the NOEC of the mean hatching time of the present study (0.03 μg TBT/L) was lower than the one reported by Mathijssen-Spiekman et al. (1989), who assessed 17% of the effects after 9 days exposure of embryos in 0.1 μg TBT/L. Information on the toxicity of TBT to other gastropod molluscs can be found in Leung et al. (2004) and Duft et al. (2005). Exposing egg masses of the pulmonate snail Physa fontinalis, Leung et al. (2004) determined a NOEC and LOEC of 0.01 and 1.0 μg TBT/L, respectively. Duft et al. (2005) developed a sediment toxicity test with the parthenogenetic ovoviviparous snail Potamopyrgus antipodarum. The LOEC for juvenile production after 4 weeks was 24.4 μg TBT/kg dw, which was the lowest concentration. Hence, a NOEC could not be determined. The EC10 was calculated as 2.39 μg TBT/kg dw. Due to the differences in employed test designs, e.g. a spacing factor of 10 (Leung et al., 2004) and sediment exposure (Duft et al., 2005), a comparison of species sensitivities is hampered. Regarding the different solvents, it seems that there could be a stimulating influence of some of these substances (i.e. those containing glycol) on embryo development, albeit at relatively high concentrations. Yang et al. (2008) found such an effect on larvae of the mussel Mytilus galloprovincialis when exposed to various solvents, including ethylene glycol. At present it is not understood if the cause for these effects is of a direct or indirect nature. Comprehensive data regarding solvent effects on aquatic organisms, resulting both in stimulating and inhibiting effects, have been reviewed by Hutchinson et al. (2006). Nonetheless studies on molluscs, especially snails, are rare. In further studies with more replicates and lower concentrations, the influence of these solvents should be confirmed and assessed clearly.
4.3. Embryo testing with other mollusc species There are two acute standard toxicity tests with mollusc embryos, employing marine bivalve species of commercial interest. One test was described by Thain (1991) for the International Council for the Exploration of the Sea (ICES). In this 24-h test the development of embryos of the oyster Crassostrea gigas to the shelled larval stage is examined. The American Society for Testing and Materials (ASTM, 2004) also published a standard test describing a 48-h test on embryos and the resulting larvae of four species of saltwater bivalve molluscs (Pacific oyster, Crassostrea gigas; eastern oyster, Crassostrea virginica; hard clam, Mercenaria mercenaria; and blue mussel, Mytilus edulis). In contrast to L. stagnalis, which produces egg masses, the gametes of bivalves have to be collected and then be fertilised extra corporal. This biological attribute complicates the test, despite its short exposure duration. Embryos of other pulmonate mollusc species have been employed in toxicity tests as well, e.g. Physa acuta (Cheung and Lam, 1998). These authors demonstrated that embryos were slightly more sensitive to cadmium than juvenile snails and suggested also that there was a role for early life stage testing of pulmonate snails in ecotoxicological testing. Obviously, there is a specific interest to develop tests with eggs/ embryos of fish as these do not qualify as vertebrate tests before the fish start eating independently (Braunbeck et al., 2005; OECD 212, 1998; Scholz et al., 2008). When further data for molluscs become available, it would be interesting to systematically study the sensitivity differences between fish and molluscs. A comparison of test results between fish and Marisa cornuarietis embryo toxicity tests showed a comparable
C. Bandow, L. Weltje / Science of the Total Environment 435–436 (2012) 90–95
sensitivity for lithium and copper, while fish were more sensitive to lead and palladium (Sawasdee and Köhler, 2010). 5. Conclusion Considering all data it seems that this method of chemical testing is rather simple for assessing embryo toxicity. For a single 12 well plate a volume of only 36 mL of medium is required at each renewal. Hence, costs for chemicals and laboratory materials are low. Furthermore the multiwell plates need little space and are easily handled. With 21 days the test has a comparatively short duration to achieve explicit and sensitive endpoints as was demonstrated by means of the reference toxicant TBT. In short, this test appears to be adequate to become a part of the aspired guideline for mollusc life cycle toxicity test within the OECD or alternately to serve as a trigger to decide upon the necessity of a (partial) life-cycle test. Acknowledgements The authors wish to thank Jörg Oehlmann (Goethe-University Frankfurt/Main, Germany) and three anonymous reviewers for their helpful advice and comments on a previous version of this paper. This work originates from the diploma thesis of C.B., under supervision of Dieter Greif (University of Applied Sciences Zittau/Görlitz). References ASTM Standard E724 - 98. Standard Guide for Conducting Static Acute Toxicity Tests Starting with Embryos of Four Species of Saltwater Bivalve Molluscs. West Conshohocken, USA: ASTM International; 2004. Available from: www.astm.org. Boon-Niermeijer EK, van den Berg A, Wikman G, Wiegant FAC. Phyto-adaptogens protect against environmental stress-induced death of embryos from the freshwater snail Lymnaea stagnalis. Phytomedicine 2000;7(5):389–99. Braunbeck T, Böttcher M, Hollert H, Kosmehl T, Lammer E, Leist E, et al. Towards an alternative for the acute fish LC50 test in chemical assessment: the fish embryo toxicity test goes multi-species—an update. ALTEX 2005;22(2):87-102. Canton JH, Slooff W. The usefulness of Lymnaea stagnalis L. as a biological indicator in toxicological bio-assays (model substance α-HCH). Water Res 1977;11(1):117–21. Cheung CCC, Lam PKS. Effect of cadmium on the embryos and juveniles of a tropical freshwater snail, Physa acuta (Draparnaud, 1805). Water Sci Technol 1998;38(7): 263–70. Cooney JJ. Microbial transformations of tin and tin compounds. J Ind Microbiol 1988;3(4): 195–204. [as citied in Hazardous Substances Data Bank [2012 May 30]. Available from: http://toxnet.nlm.nih.gov/cgi-bin/sis/search/f?./temp/~8tUXHU:1]. De Lange RPJ, van Minnen J, Boer HH. Expression and translation of the egg-laying neuropeptide hormone genes during post-embryonic development of the pond snail Lymnaea stagnalis. Cell Tissue Res 1994;275(2):369–75. Duft M, Schulte-Oehlmann U, Tillmann M, Weltje L, Oehlmann J. Biological impact of organotin compounds on molluscs in marine and freshwater ecosystems. Coast Mar Sci 2005;29(2):95-110. Elendt B-P, Bias W-R. Trace nutrient deficiency in Daphnia magna cultured in standard medium for toxicity testing. Effects of the optimization of culture conditions on life history parameters of D. magna. Water Res 1990;24(9):1157–67. Eng G, Bathersfield O, May L. Mössbauer studies of the speciation of tributyltin compounds in seawater and sediment samples. Water Air Soil Pollut 1986;27(1–2): 191–7. [as citied in Hazardous Substances Data Bank [2012 May 30]. Available from: http://toxnet.nlm.nih.gov/cgi-bin/sis/search/f?./temp/~8tUXHU:1]. Glöer P, Meier-Brook C, Ostermann O. Süßwassermollusken, ein Bestimmungsschlüssel für die Bundesrepublik Deutschland (Freshwater molluscs, a determination key for the Federal Republic of Germany). 10th ed. Hamburg, Germany: Deutscher Jugendbund für Naturbeobachtung; 1992 [In German]. Gomot A. Toxic effects of cadmium on reproduction, development, and hatching in the freshwater snail Lymnaea stagnalis for water quality monitoring. Ecotoxicol Environ Saf 1998;41(3):288–97. Grosell M, Brix KV. High net calcium uptake explains the hypersensitivity of the freshwater pulmonate snail, Lymnaea stagnalis, to chronic lead exposure. Aquat Toxicol 2009;91(4):302–11. Grosell M, Gerdes RM, Brix KV. Chronic toxicity of lead to three freshwater invertebrates—Brachionus calyciflorus, Chironomus tentans and Lymnaea stagnalis. Environ Toxicol Chem 2006;25(1):97-104.
95
Hutchinson TH, Shillabeer N, Winter MJ, Pickford DB. Acute and chronic effects of carrier solvents in aquatic organisms: a critical review. Aquat Toxicol 2006;76(1): 69–92. Lalah JO, Severin GF, Schramm KW, Lenoir D, Behechti A, Guenther K, et al. Effects of a branched p-nonylphenol isomer (4(3′,6′-dimethyl-3′-heptyl)-phenol) on embryogenesis in Lymnaea stagnalis L. Arch Environ Contam Toxicol 2007;52(1):104–12. Lanzer R. GSF—National Research Center for Environment and Health. Toxicity of nonylphenol during embryo growth in Lymnaea stagnalis L. GSF Report. Munich, Germany; 1999. Leung KMY, Morley NJ, Grist EPM, Morritt D, Crane M. Chronic toxicity of tributyltin on development and reproduction of the hermaphroditic snail Physa fontinalis: influence of population density. Mar Environ Res 2004;58(2–5):157–62. Leung KMY, Grist EPM, Morley NJ, Morritt D, Crane M. Chronic toxicity of tributyltin to development and reproduction of the European freshwater snail Lymnaea stagnalis (L.). Chemosphere 2007;66(7):1358–66. Maguire RJ, Tkacz RJ. Degradation of the tri-n-butyltin species in water and sediment from Toronto Harbor. J Agric Food Chem 1985;33(5):947–53. [as citied in Hazardous Substances Data Bank [2012 May 30]. Available from: http://toxnet.nlm.nih. gov/cgi-bin/sis/search/f?./temp/~8tUXHU:1]. Mathijssen-Spiekman EAM, Canton JH, Roghair CJ. Dutch National Institute for Public Health and the Environment. Onderzoek naar de toxiciteit van TBTO voor een aantal zoetwaterorganismen (Research into the toxicity of TBTO for a number of freshwater organisms). Bilthoven, The Netherlands: RIVM; 1989 [Report No.: 668118 001. In Dutch]. Matthiessen P. An assessment of endocrine disruption in mollusks and the potential for developing internationally standardized mollusk life cycle test guidelines. Integr Environ Assess Manag 2008;4(3):274–84. Morley NJ, Leung KMY, Morritt D, Crane M. Toxicity of anti-fouling biocides to encysted metacercariae of Echinoparyphium recurvatum (Digenea: Echinostomatidae) and their snail hosts. Chemosphere 2004;56(4):353–8. Oehlmann J, Schulte-Oehlmann U. Endocrine disruption in invertebrates. Pure Appl Chem 2003;75(11):2207–18. Organisation for Economic Co-operation and Development. OECD guideline for testing of chemicals. OECD 212 fish, short-term toxicity test on embryo and sac-fry stages. Paris, France; 1998. Organisation for Economic Co-operation and Development. OECD series on testing and assessment, number 23. Guidance document on aquatic toxicity testing of difficult substances and mixtures. Paris, France; 2000. Organisation for Economic Co-operation and Development. OECD (2004a): OECD guidelines for the testing of chemicals. OECD 202 Daphnia sp., Acute immobilisation test. Paris, France; 2004a. Organisation for Economic Co-operation and Development. OECD Guidelines for the testing of chemicals. OECD 218 Sediment-water chironomid toxicity test using spiked sediment. Paris, France; 2004b. Organisation for Economic Co-operation and Development. OECD Guidelines for the testing of chemicals. OECD 219 Sediment-water chironomid toxicity test using spiked water. Paris, France; 2004c. Organisation for Economic Co-operation and Development. OECD Guidelines for the testing of chemicals. OECD 211 Daphnia magna Reproduction test. Paris, France; 2008. Organisation for Economic Co-operation and Development. Detailed Review Paper on Mollusc Life-Cycle Toxicity Testing. Paris, France; 2010. Sawasdee B, Köhler H-R. Embryo toxicity of pesticides and heavy metals to the ramshorn snail, Marisa cornuarietis (Prosobranchia). Chemosphere 2009;75(11):1539–47. Sawasdee B, Köhler H-R. Metal sensitivity of the embryonic development of the ramshorn snail Marisa cornuarietis (Prosobranchia). Ecotoxicology 2010;19(8):1487–95. Scholz S, Fischer S, Gündel U, Küster E, Luckenbach T, Voelker D. The zebrafish embryo model in environmental risk assessment – applications beyond acute toxicity testing. Environ Sci Pollut Res Int 2008;15(5):394–404. Schwab H. Süßwassertiere, ein ökologisches Bestimmungsbuch (Freshwater organisms, an ecological determination guide). 1st ed. Stuttgart, Germany: Ernst Klett Verlag; 2002 [In German]. Taylor HH. The ionic and water relations of embryos of Lymnaea stagnalis, a freshwater pulmonate mollusc. J Exp Biol 1977;69:143–72. Thain JE. International Council for the Exploration of the Sea. Biological effects of contaminants: Oyster (Crassostrea gigas) embryo bioassay, 11. Techniques in marine environmental sciences; 1991.. [Copenhagen, Denmark]. Van Duivenboden YA. Transfer of semen accelerates the onset of egg-laying in female copulants of the hermaphrodite freshwater snail, Lymnaea stagnalis. Int J Invert Reprod 1983;6(4):249–57. Weltje L, Scholz C, Stark C, Ziebart S, van Doornmalen J, Oehlmann J. Endocrine disruption in the freshwater pondsnail Lymnaea stagnalis L. Poster presentation of the 13th annual meeting of SETAC Europe; 2003 Apr 27-May 1; Hamburg, Germany; 2003. Yang J-L, Satuito CG, Bao W-Y, Kitamura H. Induction of Metamorphosis of Pediveliger Larvae of the Mussel Mytilus galloprovincialis Lamarck, 1819 using Neuroactive Compounds, KCl, NH4Cl and Organic Solvents. Biofouling 2008;24(6):461–70.