Sublethal responses in Melanoides tuberculata following exposure to Cylindrospermopsis raciborskii containing cylindrospermopsin

Harmful Algae 6 (2007) 642–650 www.elsevier.com/locate/hal

Sublethal responses in Melanoides tuberculata following exposure to Cylindrospermopsis raciborskii containing cylindrospermopsin S.H.W. Kinnear *, L.J. Duivenvoorden, L.D. Fabbro Freshwater Ecology Group, Centre for Environmental Management, Central Queensland University, Bruce Highway, Rockhampton, Qld 4702, Australia Received 23 January 2006; received in revised form 23 October 2006; accepted 2 January 2007

Abstract Sublethal effects in the aquatic snail Melanoides tuberculata were examined during exposure to whole cell extracts of Cylindrospermopsis raciborskii and live C. raciborskii cultures, containing varying concentrations of algal cells, cellular debris, and the blue-green algal toxin, cylindrospermopsin (CYN). Exposure to whole cell extracts or live algal cultures did not result in significant changes in adult snail behaviour or relative growth rates. However, clear changes in the number of hatchlings released from parent snails were observed. Exposure to whole cell extracts containing 200 mg L 1 extracellular CYN resulted in an increase in the number of hatchlings. In contrast, decreases in hatchling number were recorded from treatments containing 200 mg L 1 CYN during exposures to live C. raciborskii cultures, compared with controls. This suggests that CYN may be more toxic to grazing invertebrates if present in the intracellular form. Since CYN is a protein synthesis inhibitor, it is possible that CYN may be especially toxic to rapidly developing tissues such as snail embryos. This may also explain the lack of effects observed in adult snails. # 2007 Elsevier B.V. All rights reserved. Keywords: Cylindrospermopsin; Cylindrospermopsis raciborskii; Cyanobacteria; Deoxy-cylindrospermopsin; Ovoviviparous; Gastropod

1. Introduction Cylindrospermopsis raciborskii (Seenayya & Subba Raju) is a blue-green alga belonging to the order Nostocales (Koma´rek and Anagnostidis, 1986). Blooms of C. raciborskii occur in almost every continent, being especially prevalent in Australia, Europe, Indonesia and Africa (Padisa´k, 1997). The species is largely tropical to subtropical, but is considered highly adaptive and appears to be invading new habitats even in temperate climates (Briand et al., 2004; Padisa´k, 1997). Furthermore, C. raciborskii is particularly known for its ability

* Corresponding author. Tel.: +61 7 4930 9647; fax: +61 7 4930 9209. E-mail address: [email protected] (S.H.W. Kinnear). 1568-9883/$ – see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.hal.2007.01.004

to produce the potent alkaloid hepatotoxin, cylindrospermopsin (CYN) (Ohtani et al., 1992). Concentrations of CYN from C. raciborskii blooms in Australia have been recorded at up to 589 mg L 1 (Saker and Eaglesham, 1999; Saker, 2000; Saker and Griffiths, 2001). Hence, an understanding of the environmental and health risks associated with toxic C. raciborskii blooms is critical. Cylindrospermopsin exerts a wide range of toxic effects on both animals and plants. The 24 h LC50 of CYN via intraperitoneal injection was reported at 2.1 mg kg 1 in laboratory studies with mice (Ohtani et al., 1992). However, other toxicity values for CYN appear to vary according to ingestion method, type of toxin (purified solutions compared with cell extracts) and test organism sensitivity (Seawright et al., 1999; Shaw et al., 2000; Metcalf et al., 2002).

S.H.W. Kinnear et al. / Harmful Algae 6 (2007) 642–650

Currently, there is a distinct information gap relating to the ecotoxicity of CYN to invertebrate fauna. Acute toxicity tests on aquatic organisms have been limited to a few recent studies. Saker and Eaglesham (1999) examined histological effects of CYN on redclaw crayfish, but reported no abnormalities. In contrast, a number of C. raciborskii strains have shown high mortality, reduced growth, and changes to fecundity in Daphnia magna, regardless of the CYN concentrations in each (Hawkins and Lampert, 1989; Nogueira et al., 2004). Macroinvertebrates are key components of aquatic food webs, therefore studies of the toxic effects of CYN towards these species are vital to understanding the full range of effects that toxic algal blooms may have on ecological systems. In this study, the ecotoxicity of C. raciborskii whole cell extracts and live cultures were tested on Melanoides tuberculata using environmentally relevant exposure concentrations. The key aim of this study was to determine the differences in ecotoxicological effects following exposure to intracellular toxin (live cultures) and extracellular toxin (whole cell extracts) in M. tuberculata. This is important as the relative abundance of different toxin components may change depending on the age of a naturally occurring bloom (White et al., 2005). Thus, the null hypothesis was ‘‘that the responses of M. tuberculata are the same, regardless of mode of exposure to cylindrospermopsin via whole-cell extracts or live cultures of C. raciborskii’’. 2. Materials and methods 2.1. Test procedures M. tuberculata (Mu¨ller, 1774) were collected from a semi-permanent creek in central Queensland. CYN was not detected using High-Performance Liquid Chromatography from water samples taken during specimen collection. Species identification was provided by Dr. Winston Ponder (Australian Museum, Sydney). Three static-renewal ecotoxicity trials were conducted based on the guidelines provided by the American Society for Testing and Materials (ASTM, 2003). The first two trials examined exposure to C. raciborskii whole cell extracts with toxin available in the extracellular form only, using test concentrations of 0 (control), 25, 50, 100, 200 and 400 mg L 1 CYN. Treatments were prepared by double freeze-thawing a C. raciborskii culture of known toxicity and diluting to the desired concentration by adding non-sterile tap water (trial one) or creek water (trial two). The latter

643

was collected at the Moores creek sampling site from which snails had been collected, and was filtered through Whatman GF/F glass microfibre filters. Trial three examined exposure to a live C. raciborskii culture that contained CYN in a mix of both intracellular (cellbound, contained within intact algal cells) and extracellular (dissolved, lysed) components. Treatments were prepared by diluting a large C. raciborskii culture to 0% (control), 10%, 20%, 30%, 40% and 50% of the original concentration using filtered Moores Ck water. Further details regarding test design, water quality monitoring, and toxin determinations can be found in White et al. (2006). Trials commenced by collectively weighing three snails and randomly assigning these to a treatment dish, which was a rectangular glass dish of approximately 350 mL capacity. Snail behaviour was recorded at 24 h, 48 h and every 2 days thereafter, usually within 2–3 h after the (artificial) onset of daylight (12:12 light:dark photoperiod). Behavioural measurements were conducted by removing test chambers from the water bath to a laboratory bench. After 2 min, each snail was observed and assigned a value of five (active: muscular foot extended and/or actively swimming or feeding), three (alive: but operculum closed) or zero (animal dead: no response to stimulation of operculum and/or evidence of tissue decomposition). Individual scores were then summed for each dish: a single test chamber could therefore receive a score ranging from zero (three dead snails) to fifteen (three active snails). The same assessor was used throughout all trials. Since M. tuberculata is an ovoviviparous species that carries developing young in a brood pouch above the head, the number of emerged hatchlings (cumulative value) in each test chamber was also recorded following assessment of snail behaviour. Snails were harvested (n = three flasks) after 7 or 14 days exposure, reweighed (fresh weight, to 0.01 g), and relative growth rates calculated based on biomass change compared with day zero (grams, wet weight). Snails were not fed during the trials since toxin adsorption to food sources may have influenced the bioavailability of extracellular CYN (for example, snails could directly ingest extracellular toxin if this was adsorbed to food). Minimising food sources also reduced faecal production, which can otherwise decrease dissolved oxygen concentrations and hence the biological activity of toxins (ASTM, 2003). Similarly long exposure periods without food have been used successfully in other studies of aquatic snails (Lajtner et al., 1996; Klobucar et al., 1997).

644

S.H.W. Kinnear et al. / Harmful Algae 6 (2007) 642–650

2.2. Statistical analyses Behavioural data were examined using two-way repeated measures ANOVA. Sphericity was checked using Mauchly’s test: when this value was significant, the MANOVA value (Pillai’s Trace statistic) was reported instead of Roy’s largest root. Dunnett’s T3 tests (multiple pairwise comparisons of heterogeneous data) were used to detect significant differences between individual treatment groups (a = 0.01). Numbers of hatchlings were analysed using two-way ANOVA (a = 0.05, data homogeneous) and post hoc Tukey testing, using the cumulative number of hatchlings from each test chamber on harvest days. Relative growth rates (RGR) were examined using two-way ANOVA. A lower significance level (a = 0.01) was applied to reduce type I error rates, since datasets were heterogeneous and could not be suitably transformed (Underwood, 1981). 3. Results 3.1. Water quality and toxin concentrations In all trials, oxygen levels were generally 80% saturation. Total ammonia concentrations remained

within acceptable limits (1.0 ppm), with the exception of the 400 mg L 1 treatments in trials one and two. CYN concentrations met or exceeded nominal concentrations trials one and two. In the live exposure trial, the average total CYN concentrations in the control and test treatments were 1, 91, 167, 223, 294 and 406 mg L 1, respectively. Overall, the test solutions were therefore broadly comparable with those of the extracellular trials (0–400 mg L 1). Extracellular toxin represented between 72% and 81% of total CYN. More detailed descriptions of water quality and the toxin concentrations of the various treatments can be found in White et al. (2006). 3.2. Behaviour The behaviour of M. tuberculata in trial one, where tap water was used to dilute the test solutions, was markedly different compared with trials two and three (Fig. 1), which used filtered Moores Ck water. Treatments containing high concentrations of whole cell extracts, and hence greater CYN concentrations, showed the highest levels of snail activity. In contrast, lower concentrations of the whole cell extracts and the controls were associated with immobile or dead snails.

Fig. 1. Behavioural scores recorded for Melanoides tuberculata following exposure to Cylindrospermopsis raciborskii whole cell extracts in trial one (A) and trial two (B); and live C. raciborskii exposure in trial three (C). Graphs show average values (n = 6, days 1–6; or n = 3, days 8–14).

S.H.W. Kinnear et al. / Harmful Algae 6 (2007) 642–650

645

Table 1 Summary results for two-way repeated measures ANOVA for snail behavioural values Week 1

Week 2

Trial one Mauchly’s test for sphericitya Treatment Exposure time Interaction (time  treatment)

p = 0.010; p = 0.000; p = 0.001; p = 0.000;

Trial two Mauchly’s test for sphericitya Treatment Exposure time Interaction (time  treatment)

NS NS NS NS

p = 0.000; d.f. = 5; M = 0.052 NS p = 0.005; F3,9 = 8.927 NS

Trial three Mauchly’s test for sphericitya Treatment Exposure time Interaction (time  treatment)

NS NS NS NS

NS NS NS NS

d.f. = 5; M = 0.589 F5,30 = 80.114 F3,28 = 7.743 F5,30 = 6.982

p = 0.000; d.f. = 5; M = 0.089 p = 0.000; F5,12 = 89.985 NS p = 0.000; F5,12 = 16.458

a

Values from two-way ANOVA (sphericity assumed statistic); or, where Mauchly’s test was significant, two-way repeated measures MANOVAs (Roy’s Largest Root Statistic). NS = not significant, p > 0.010 for heterogeneous data.

A significant interaction between treatment and time was present in both exposure periods (Table 1); generally, snail activity in treatments containing low toxin concentrations (100 mg L 1 CYN) progressively decreased over time (Fig. 1A). In the second trial, snail behaviour scores were 12, regardless of treatment and exposure time. No snail deaths were recorded in this trial. Significant differences were not detected between treatments during the trial; a significant effect of exposure time was recorded only during the second week (Table 1). Very similar results were recorded from the live C. raciborskii trial (Fig. 1C), except that time was not significant (Table 1). 3.3. Release of hatchlings M. tuberculata hatchlings were released during every trial. All young survived and appeared healthy (= mobile). The pattern of release was generally gradual, with one to six new hatchlings appearing during each 48 interval. In the first trial, exposure to 200 and 400 mg L 1 toxin in the whole cell extract treatments resulted in 30 hatchlings being recorded by day 14, whilst all other treatments recorded 1 (Fig. 2A). These data could not be analysed by ANOVA as some samples had zero standard deviation. In the second trial, treatments containing high CYN concentrations recorded the highest number of hatchlings again, particularly towards the end of the 14-day trial. However, the magnitude of this effect was much less compared with trial one (Fig. 2B). Significant effects of treatment, time or interaction were not

observed during this trial (two-way ANOVA; p > 0.050; F = 0.402, 0.035 and 1.104 for treatment, time and interaction, respectively). Exposure to live C. raciborskii resulted in hatchlings being produced overall (generally, 10 per treatment). Also, the pattern of release changed: the controls and the treatments containing low toxin concentrations recorded the most hatchlings, whilst treatments with high toxin concentrations recorded the least (Fig. 2C). Time was the only factor to exert a significant effect in this trial ( p = 0.005; F 1,24 = 9.690, two-way ANOVA). Pearson Product Moment correlations examined the relationships between snail behaviour scores (treatment averages) and the cumulative number of hatchlings released for each exposure phase. In half of the tested cases, a strongly significant, positive correlation was found between snail activity and the number of emerged hatchlings (Table 2). However, in the remaining half, no significant correlations were observed. 3.4. Relative growth rates During trial one, positive growth was generally limited to whole cell extract treatments containing 100 mg L 1 CYN. In contrast, growth rates were almost always positive during trials two and three (Fig. 3B and C). Two-way ANOVA (a = 0.01, heterogeneous data) did not find significant differences between treatment, time or interaction in any of these trials ( p > 0.010). Deaths of snails in the first trial prevented calculation of some treatment RGRs based on changes in fresh weights (Fig. 3A).

646

S.H.W. Kinnear et al. / Harmful Algae 6 (2007) 642–650 Table 2 Summary of Pearson Product Moment correlations, comparing average activity levels and cumulative number of hatchlings per treatment for each exposure phase Trial

Days 0–7

Days 8–14

Trial one (extracellular) Trial two (extracellular) Trial three (live)

p = 0.000; r = 0.821 NS

p = 0.000; r = 0.918 p = 0.007; r = 0.615 NS

NS

NS = not significant, p  0.050.

Fig. 2. Average number of hatchlings released by M. tuberculata following exposure to C. raciborskii whole cell extracts in trial one (A) and trial two (B); and live C. raciborskii exposure in trial three (C).

4. Discussion 4.1. Release of hatchlings Increased release of hatchlings by M. tuberculata could indicate over-reproduction or abortion, possibly as a sign of stress. For example, exposure to CYN concentrations may prompt a shock response in the

snails and result in early termination of brooding. If so, released embryos may be too underdeveloped to survive (A/Prof. R. Dillon, Department of Biology, College of Charleston, South Carolina, personal communication). However, this kind of toxic shock response would have been better evidenced by the sudden release of all hatchlings early in the exposure phase, rather than gradually. Adult Melanoides typically release hatchlings at a fairly constant rate, determined by the production and maturation of the embryos (A/Prof. Rob Dillon, personal communication). Furthermore, all young in the trials appeared to be healthy, surviving offspring, at least for the trial duration (7–14 days). In oviparous species, the gelatinous capsule surrounding the egg masses may offer some protection to developing embryos, by minimising water (and hence dissolved toxin) uptake (Singh and Agarwal, 1986). In M. tuberculata (an ovoviviparous species), transdermal exposure to the snail young could have been similarly limited by the protective brood pouch. In the third trial, where intracellular toxin was present, the pattern of release was reversed: fewer hatchlings appeared in treatments containing higher CYN concentrations. This suggests that exposure to live C. raciborskii may be more toxic to hatchlings than whole extracts containing extracellular toxin only. For example, whole cell extracts associated with toxin exposures 400 mg L 1 may cause a mild, stimulatory toxicity, whilst live exposures of similar total toxin concentrations (406 mg L 1) may exert sufficient toxicity to decrease reproduction or delay hatching emergence. There has been limited research on the development freshwater snail hatchlings in ovoviviparous species, though some studies have described the effect of pollutants on hatching and survival of eggs from oviparous species. Here, pesticide and heavy metal toxicities and acidic pH values are associated with decreases in egg production, delayed hatching or decreased embryonic development and death of young

S.H.W. Kinnear et al. / Harmful Algae 6 (2007) 642–650

647

Fig. 3. Average relative growth rates (RGR) in M. tuberculata following exposure to C. raciborskii whole cell extracts in trial one (A) and trial two (B); and live C. raciborskii exposure in trial three (C). Graphs show average values (n = 3), bars show standard error.

(Holcombe et al., 1984; Singh and Agarwal, 1986; Gomot, 1998; Tripathi and Singh, 2002). These results suggest that exposure to hazardous substances (such as harmful algal blooms) may be associated with similar toxic effects in ovoviviparous snails, such as reduced embryo production, development, and survival. In particular, delayed hatching may indicate a strategy to prolong development and maximise hatchling survival, especially if the brood pouch can minimise toxicant exposure and uptake. Most recently, EllisTabanor and Hyslop (2005) reported that exposure to the pesticide endosulfan caused M. tuberculata to produce fewer, but larger, hatchlings. It was suggested that endosulfan might inhibit calcium metabolism, thus delaying embryonic shell development and causing the parent snails to retain hatchlings in the brood pouch (Ellis-Tabanor and Hyslop, 2005). The response of M. tuberculata to CYN exposure in regards to hatchling release may therefore represent one of three scenarios. (1) Toxin exposure may be truly beneficial to the species, as reflected by increased development and release of hatchlings. However, the behavioural results from the extracellular trials do not corroborate this. (2) That CYN a is mild stimulant at low toxin concentrations, or when present only in the extracellular form, as evidenced by an increased release of apparently healthy young. On the other hand, higher

exposure concentrations could promote early release of hatchlings (=abortion), causing decreased survivorship. Since CYN primarily exerts toxicity via protein synthesis inhibition, tissues with a fast cell reproduction rate such as embryonic tissues may be strongly affected by CYN exposure. The effect of CYN exposure on calcium metabolism (and hence shell and muscle development) is unknown. Finally, (3) release of hatchlings may represent an actual toxic response; specifically, a trade-off between reducing ‘brooding’ stress on the adult versus increasing the toxin exposure risk to hatchlings once released from the brood pouch. Given the anomalies between the extracellular and live exposure trials, which of these three scenarios is applicable to CYN remains unresolved. A more thorough examination of released hatchlings (e.g., parameters such as size at birth, long-term survival rates) could provide more answers. 4.2. Snail behaviour The lack of significant treatment effects recorded for M. tuberculata suggests that exposure to 400 mg L 1 CYN in the form of whole cell extracts, or as live culture, has no effect on adult snail behaviour. A possible explanation could be that the method of measurement used (snail activity) was not discriminate

648

S.H.W. Kinnear et al. / Harmful Algae 6 (2007) 642–650

enough to detect subtle changes. However, similar techniques have been effective in studies of snail behaviour to examine the effects of aluminium and silicic acid (Campbell et al., 2000) and the algal toxin microcystin (White et al., 2004). The consistently high activity levels across all treatment groups may suggest that snails were hungry, especially if M. tuberculata selects against toxic C. raciborskii cells or cell debris. Poor food availability can potentially provide strong motivation for snail movement (Burris et al., 1990; Dillon, 2000). High activity could also suggest that snails were trying to escape CYN contaminated water, however, this does not explain the high activity in the controls. Furthermore, decreased activity in snails has been noted as a strategy to minimise toxicant uptake and toxicity (Truscott et al., 1995). Ultimately, activity levels therefore represent a trade-off between minimising toxin uptake compared with maximising chances of toxin escape and the ability to forage for food. 4.3. Relative growth rates (RGRs) Significant impacts on RGRs of M. tuberculata were not observed during exposure to either the whole cell extracts or the live C. raciborskii cultures, with total CYN concentrations 400 mg L 1. Biomass measurements may not have been sensitive enough since adult freshwater gastropods, especially prosobranchs, grow relatively slowly (Dillon, 2000). Negative or poor growth rates may also be indicative of lack of food, especially if M. tuberculata is capable of preferentially selecting against toxic C. raciborskii algal debris (the only available food source). Zooplankton species can avoid toxic (and low-nutritional quality) cyanoprokaryote cells (Reinikainen et al., 1994; DeMott, 1999; Mohamed, 2001) and freshwater gastropods may graze selectively, though not specifically against cyanoprokaryote species (Lodge, 1986). However, faecal strings were present in treatments containing higher CYN concentrations (100 mg L 1 in the whole cell extract trials; or 30% culture strength (trial 3)), especially towards the end of trial periods. This appears to indicate that M. tuberculata were grazing on live or dead toxic C. raciborskii cells. 4.4. The importance of toxin fraction availability The toxicity of C. raciborskii cells and CYN has been poorly studied with respect to aquatic macroinvertebrates. This research indicates that the toxic effects of CYN are not limited to vertebrate (particularly

mammalian) species. Crude extracts of C. raciborskii have also been shown to exert toxicity on molluscan neurones and neurotransmitter receptors, though this resulted from the presence of anatoxin, rather than CYN (Kiss et al., 2002). Mortality, decreased body sizes and increased enzymatic activity in Daphnia was also observed after exposure to C. raciborskii (Nogueira et al., 2004). Purified CYN causes death of brine shrimp (Metcalf et al., 2002), whilst redclaw crayfish and rainbowfish survived a C. raciborskii bloom containing up to 589 mg L 1 CYN, apparently with few ill effects (Saker and Eaglesham, 1999). Saker et al. (2004) also recorded high levels of CYN bioaccumulation in the mussel Anodonta, with no apparent side effects. Invertebrates could possibly be less sensitive to CYN due to having less well-developed liver apparatus. In mammals, CYN metabolism occurs within the liver, probably via the use of cytochrome P-450 (Terao et al., 1994; Shaw et al., 2000). However, this breakdown process is known to correlate with the production of molecules more toxic than the parent CYN (Runnegar et al., 1994). In the absence of cytochrome P-450, CYN toxicity may be reduced: inhibition of this enzyme offers some protection from CYN toxicity (Runnegar et al., 1995). On the other hand, toxic effects may occur in adult gastropods, but may not manifest on a gross scale (e.g., behaviour, RGRs). The fact that snail deaths were not recorded during total exposure concentrations 400 mg L 1 (either from intracellular or extracellular toxin) is important, since these are concentrations occur naturally in C. raciborskii blooms. The relative proportion of intracellular and extracellular toxin components influences the possibility for, and likely rate of, toxin ingestion within aquatic organisms (White et al., 2005). The present work has indicated that ecotoxicological effects could be different in grazing snails, depending on whether toxin is available intra- or extracellularly. In extracellular trials, toxicity may have resulted from CYN reaching target tissues: this suggests passive or active transdermal uptake, or via accidental ingestion of dissolved toxins in the test solutions. In contrast, during the live C. raciborskii trial, snail grazing on toxin laden cells may have allowed direct ingestion of far higher toxin doses. Earlier research is consistent with this: bioaccumulation of CYN in M. tuberculata was dramatically increased following exposure to live algal cultures compared with freeze-thawed whole cell extracts (White et al., 2006). Thus, ingestion (and probable accumulation) of greater toxin loads could partly explain the difference in hatchling release by M. tuberculata in the extracellular compared with live culture trials.

S.H.W. Kinnear et al. / Harmful Algae 6 (2007) 642–650

Finally, it should be noted that although CYN is suspected to be the primary agent of toxicity, C. raciborskii whole cell extracts may contain other substances, most notably the toxin analog, deoxyCYN. Early research suggests that deoxy-CYN has a minimal contribution to the toxicity of C. raciborskii, at least in mouse bioassay (Norris et al., 1999), though more recent work aligns the toxicity of deoxy-CYN with that of CYN, at least in vitro (Looper et al., 2005). Thus, it cannot be discounted that deoxy-CYN, other substances like cyanoprokaryote lipopolysaccharides, and/or additive and synergistic effects between these compounds could be at least partly responsible for toxicity in invertebrate fauna. 4.5. Water quality and toxin concentrations Snail deaths recorded from control treatments during trial one suggested that the use of aged tap water used was inappropriate. Subsequent testing (not described here) showed that use of Moores Ck water resulted in the highest behaviour scores for M. tuberculata: trials two and three were therefore prepared using filtered creek water as the diluent. Natural sources (such as local streams) are sometimes recommended in preference to artificial solutions, provided these are adequately filtered and stored prior to use (Riethmuller et al., 2003). 5. Conclusions Exposure to C. raciborskii whole cell extracts and live C. raciborskii cultures did not cause changes to mortality, relative growth rates or behaviour in adult Melanoides snails. However, the release of hatchlings from the brood pouch appeared to be influenced by both the whole cell extract and live culture test solutions. This work indicates that the toxic effects of C. raciborskii blooms containing CYN are likely to impact macroinvertebrate fauna. Furthermore, blooms containing a mix of both intracellular and extracellular toxin fractions may pose a greater risk than those that feature predominantly extracellular toxin, for example, during bloom senescence. Acknowledgements The authors thank TermiMesh Rockhampton for the donation of stainless steel mesh used in cage construction; Geoff Eaglesham for his work with toxin analyses; Prof. Dillon for his comments and Dr. Ponder for snail identification. This work was supported financially by Nebo Shire Council (Lake Elphinstone Alliance), the

649

Centre for Environmental Management, federal Research Training Scheme funds and an Australian Postgraduate Award.[SS] References ASTM, 2003. E 1192-97 Standard Guide for Conducting Acute Toxicity Tests on Aqueous Ambient Samples and Effluents with Fishes, Macroinvertebrates, and Amphibian. American Society for Testing and Materials International, West Conshohocken, Philadelphia, 14 pp. Briand, J.F., Leboulanger, C., Humbert, J.-F., Bernard, C., Dufour, P., 2004. Cylindrospermopsis raciborskii (Cyanobacteria) Invasion at Mid-Latitudes: Selection; Wide Physiological Tolerance, or Global Warming? J. Phycol. 40, 231–238. Burris, J.A., Bamford, M.S., Stewart, A.J., 1990. Behavioural responses of marked snails as indicators of water quality. Environ. Toxicol. Chem. 9, 69–76. Campbell, M.M., White, K.N., Jugdaohsingh, R., Powell, J.J., McCrohan, C.R., 2000. Effect of aluminum and silicic acid on the behaviour of the freshwater snail Lymnaea stagnalis. Can. J. Fish. Aquat. Sci. 57, 1151–1159. DeMott, W.R., 1999. Foraging strategies and growth inhibition in five daphnids feedind on mixtures of a toxic cyanobacterium and a green alga. Freshwater Biol. 42, 263–274. Dillon, R.T., 2000. The Ecology of Freshwater Molluscs. Cambridge University Press, Cambridge, UK. Ellis-Tabanor, M., Hyslop, E., 2005. Effect of sublethal concentrations of endosulfan on growth and fecundity of two species of snails. Bull. Environ. Contam. Toxicol. 74, 1173–1178. Gomot, A., 1998. Toxic effects of cadmium on reproduction, development and hatching in the freshwater snail Lymnaea stagnalis for Water Quality Monitoring. Ecotox. Environ. Saf. 41, 288–297. Hawkins, P.R., Lampert, W., 1989. The effect of Daphnia body size on filtering rate inhibition in the presence of a filamentous cyanobacterium. Limnol. Oceanogr. 34, 1084–1089. Holcombe, G.W., Phipps, G.L., Marier, J.W., 1984. Methods for conducting snail (Aplexa hypnorum) embryo through adult exposures: effects of cadmium and reduced pH levels. Arch. Environ. Contam. Toxicol. 13, 627–634. ´ ., Hiripi, L., Kova´cs, M., Vo¨ro¨s, L., 2002. Kiss, T., Vehovsky, A Membrane effects of toxins isolated from a cyanobacterium, Cylindrospermopsis raciborskii, on identified molluscan neurones. Comp. Biochem. Physiol. Part C 131, 167–176. Klobucar, G.I.V., Lajtner, J., Erben, R., 1997. Lipid peroxidation and histopathological changes in the digestive gland of a freshwater snail Planorbarius corneus L. (Gastropoda, Pulmonata) exposed to chronic and sub-chronic concentrations of PCP. Bull. Environ. Contam. Toxicol. 58, 128–134. Koma´rek, J., Anagnostidis, K., 1986. Modern approach to the classification system of cyanophytes. Arch. Hydrobiol. Suppl. 73.2 Algological Stud. 43, 157–226. Lajtner, J., Erbern, R., Klobucar, G.I.V., 1996. Histopathological effects of phenol on the digestive gland of Amphimelania holandri Fe´r. (Gastropoda, Prosobranchia). Bull. Environ. Contam. Toxicol. 57, 458–464. Lodge, D.M., 1986. Selective grazing on periphyton: a determinant of freshwater gastropod microdistributions. Freshwater Biol. 16, 831–841. Looper, R.E., Runnegar, M.T.C., Williams, R.M., 2005. Synthesis of the putative structure of 7-deoxycylindrospermopsin: C7

650

S.H.W. Kinnear et al. / Harmful Algae 6 (2007) 642–650

oxygenation is not required for the inhibition of protein synthesis. Angew. Chem. Int. Ed. 44, 3879–3881. Metcalf, J.S., Lindsay, J., Beattie, K.A., Birmingham, S., Saker, M.L., To¨ro¨kne´, A.K., Codd, G.A., 2002. Toxicity of cylindrospermopsin to the brine shrimp Artemia salina: comparisons with protein synthesis inhibitors and microcystins. Toxicon 40, 1115–1120. Mohamed, Z.A., 2001. Accumulation of cyanobacterial heptatotoxins by Daphnia in some Egyptian irrigation channels. Ecotox. Environ. Saf. 50, 4–8. Nogueira, I.C.G., Saker, M.L., Pflugmacher, S., Wiegand, C., Vasconcelos, V.M., 2004. Toxicity of the cyanobacterium Cylindrospermopsis raciborskii to Daphnia magna. Environ. Toxicol. 19, 453–459. Norris, K.R., Eaglesham, G., Pierens, G., Shaw, G., Smith, M.J., Chiswell, R.K., Seawright, A.A., Moore, M.R., 1999. Deoxycylindrospermopsin, an analog of cylindrospermopsin from Cylindrospermopsis raciborskii. Environ. Toxicol. 14, 163–165. Ohtani, I., Moore, R.E., Runnegar, M.T.C., 1992. Cylindrospermopsin: a potent hepatotoxin from the blue-green alga Cylindrospermopsis raciborskii. J. Am. Chem. Soc. 114, 7941–7942. Padisa´k, J., 1997. Cylindrospermopsis raciborskii (Woloszynska) Seenayya et Subba Raju, an expanding highly adaptive cyanobacterium: worldwide distribution and review of its ecology. Archiv fu¨r Hydrobiologie. 4 (Suppl. 107), 563–593 (Monographic Studies). Reinikainen, M., Ketola, M., Walls, M., 1994. Effects of the concentrations of toxic Microcystis aeruginosa and an alternative food on the survival of Daphnia pulex. Limnol. Oceanogr. 39, 424–432. Riethmuller, N., Camilleri, C., Franklin, N., Hogan, A., King, A., Koch, A., Markich, S.J., Turley, C., van Dam, R., 2003. Ecotoxicological Testing Protocols for Australian Tropical Freshwater Ecosystems. Environmental Research Institute of the Supervising Scientist, Environment Australia, Darwin, Australia, 140 pp. Runnegar, M.T., Kong, S., Zhong, Y., Lu, S.C., 1995. Inhibition of reduced glutathione synthesis by cyanobacterial alkaloid cylindrospermopsin in cultured rat hepatocytes. Biochem. Pharmacol. 49, 219–225. Runnegar, M.T., Kong, S.M., Zhong, Y.Z., Ge, J.L., Lu, S.C., 1994. The role of glutathione in the toxicity of a novel cyanobacterial alkaloid cylindrospermopsin in cultured rat hepatocytes. Biochem. Biophys. Res. Commun. 201, 235–241. Saker, M., 2000. Cyanobacterial blooms in tropical north Queensland water bodies. Ph.D. Thesis. James Cook University, Townsville, Australia. Saker, M.L., Eaglesham, G.K., 1999. The accumulation of cylindrospermopsin from the cyanobacterium Cylindrospermopsis

raciborskii in tissues of the redclaw crayfish Cherax quadricarinatus. Toxicon 37, 1065–1077. Saker, M.L., Griffiths, D.J., 2001. Occurrence of blooms of the cyanobacterium Cylindrospermopsis raciborskii (Woloszynska) Seenayya and Subba Raju in a north Queensland domestic water supply. Mar. Freshwater Res. 52, 907–915. Saker, M.L., Metcalf, J.S., Codd, G.A., Vasconcelos, V.M., 2004. Accumulation and depuration of the cyanobacterial toxin cylindrospermopsin in the freshwater mussel Anodonta cygnea. Toxicon 43, 185–194. Seawright, A.A., Nolan, C.C., Shaw, G.R., Chiswell, R.K., Norris, R.L., Moore, M.R., Smith, M.J., 1999. The oral toxicity for mice of the tropical cyanobacterium Cylindrospermopsis raciborskii. Environ. Toxicol. 14, 135–142. Shaw, G., Seawright, A.A., Moore, M.R., Lam, P.K.S., 2000. Cylindrospermopsin, a cyanobacterial alkaloid: evaluation of its toxicologic activity. Ther. Drug Monit. 22, 89–92. Singh, D.K., Agarwal, R.A., 1986. Toxicity of pesticides to fecundity, hatchability and survival of young snails of Lymnaea acuminata. Acta Hydrochim. Hydrobiol. 14, 191–194. Terao, K., Ohmori, S., Igarshi, K., Ohtani, I., Watanabe, M.F., Harada, K.I., Ito, E., Watanabe, M., 1994. Electron microscope studies on experimental poisoning in mice induced by cylindrospermopsin isolated from blue-green alga Umezakia natans. Toxicon 32, 833–843. Tripathi, P.K., Singh, A., 2002. Toxic effects of dimethoate and carbaryl pesticides on carbohydrate metabolism of freshwater snail Lymnaea acuminata. Bull. Environ. Contam. Toxicol. 68, 606–611. Truscott, R., McCrohan, C.R., Bailey, S.E.R., White, K.N., 1995. Effect of aluminium and lead on activity in the freshwater pond snail Lymnaea stagnalis. Can. J. Fish. Aquat. Sci. 52, 1623–1629. Underwood, A.J., 1981. Techniques of analysis of variance in experimental marine biology and ecology. Oceanogr. Mar. Biol.: Annu. Rev. 19, 513–605. White, S.H., Duivenvoorden, L.J., Fabbro, L., 2004. Effects of cyanotoxins on aquatic snails and shrimp. In: Proceedings of the AWA (Qld) Regional Conference, Kingfisher Bay, Fraser Island, Australian Water Association, Artarmon, NSW, November 11–14. White, S.H., Duivenvoorden, L.J., Fabbro, L.D., 2005. A decisionmaking framework for ecological impacts associated with the accumulation of cyanotoxins (cylindrospermopsin and microcystin). Lakes Res.: Res. Man. 10, 25–37. White, S.H., Duivenvoorden, L.J., Fabbro, L.D., Eaglesham, G.K., 2006. Influence of intracellular toxin concentration on cylindrospermopsin bioaccumulation in a freshwater gastropod (Melanoides tuberculata). Toxicon 5, 497–509.