- Email: [email protected]
Seasonal lethality of the organophosphate pesticide, azamethiphos to female American lobster (Homarus americanus)
ARTICLE IN PRESS
Ecotoxicology and Environmental Safety 60 (2005) 277–281 www.elsevier.com/locate/ecoenv
Seasonal lethality of the organophosphate pesticide, azamethiphos to female American lobster (Homarus americanus) L.E. Burridgea,, K. Hayaa, S.L Waddyb a
Marine Environmental Sciences Division, Fisheries and Oceans Canada, St. Andrews Biological Station, 531 Brandy Cove Rd., St. Andrews, NB, Canada E5B 2L9 b Aquaculture Division, Fisheries and Oceans Canada, St. Andrews Biological Station, 531 Brandy Cove Rd., St. Andrews, NB, Canada E5B 2L9 Received 16 June 2003; received in revised form 28 May 2004; accepted 8 June 2004 Available online 17 July 2004
Abstract The organophosphate pesticide azamethiphos is the active ingredient in Salmosan, a product formerly registered in Canada for the treatment of cultured Atlantic salmon against infestations of the ectoparasite Lepeophtheirus salmonis. The 48-h LC50 of azamethiphos to female American lobsters was determined bimonthly for 2 years to determine whether the sensitivity of lobsters to azamethiphos varied with time of year, molt stage, or reproductive stage. The LC50’s ranged from 0.61 to 3.24 mg/L. The lobsters were most sensitive to azamethiphos during the spawning and molting seasons which occur in the summer and early fall when seawater temperatures are highest. Testing of compounds on this species for regulatory purposes should take into account that there may be variations in sensitivity during the molt and reproductive cycles. Crown Copyright r 2004 Published by Elsevier Inc. All rights reserved. Keywords: Sea lice pesticides; Azamethiphos; Salmosan; American lobster; Lethality; Spawning; Molting
1. Introduction Since 1994 the salmon aquaculture industry in Atlantic Canada has dealt with a succession of parasite and disease problems. Changes in management practices have resulted in greatly improved husbandry and a reduction in the use of some chemicals. However, fish farmers remain reliant on chemotherapeutants to combat infestations of the most problematic ectoparasite, Lepeophtheirus salmonis, a marine copepod commonly called sea lice. Salmosan was registered for use in Canada as an antisea louse treatment until 2003 (Health Canada, 2003a, b). Salmosan is a wettable powder containing 47.5% (w/w) of azamethiphos (S-6-chloro-2, 3-dihydro2-oxo-1, 3-oxazolo[4,5-b] pyridin-3-ylmethyl) O,O-dimethyl phosphorothioate), an organophosphate pestiCorresponding author. Fax: +506-529-5862.
E-mail address: [email protected] (L.E. Burridge).
cide. Salmon are given a bath treatment with 100 mg/L of azamethiphos for up to 1 h. Health Canada sanctions the use of other anti-sea louse compounds under its emergency drug release policy and access to these treatments appears to have reduced the use of azamethiphos in eastern Canada. Sales of Salmosan have dropped from 89 kg in 1998 to a low of 6 kg in 2001. In 2002 there was an increase in sales to 31 kg (M. Boldon, New Brunswick Department of Environment and Local Government, personal communication). Several papers describing the effects of azamethiphos on indigenous species that might be exposed to the compound after routine anti-sea louse treatments have been published (Burridge et al., 1999, 2000a; Abgrall et al., 2000). These studies focused on the American lobster, a commercially important crustacean fished in areas close to salmon aquaculture sites. A potential weakness with each of these studies is that the experiments were conducted only at one time of year and did not take into account that seawater tempera-
0147-6513/$ - see front matter Crown Copyright r 2004 Published by Elsevier Inc. All rights reserved. doi:10.1016/j.ecoenv.2004.06.004
ARTICLE IN PRESS 278
L.E. Burridge et al. / Ecotoxicology and Environmental Safety 60 (2005) 277–281
ture, the physiological condition of animals, or time of year might affect sensitivity to the compound. Crustaceans are well known to be more sensitive to some chemicals near the time of molt. Rao and Conklin (1986), for example, reported that several organic pollutants were more toxic to molting grass shrimp (Palaemonetes pugio) than to intermolt shrimp. To our knowledge, no studies have considered whether time of year or reproductive stage affects the response of crustaceans to chemicals. The American lobster displays marked seasonal changes in sensitivity to environmental and endocrine manipulation (Waddy and Aiken, 2000) and sublethal responses to some compounds appear to vary with time of year and reproductive condition (Waddy et al., 2003, unpublished). Female American lobsters typically spawn and molt on a biennial cycle. Eggs are extruded in the late spring and summer, molting occurs the following year after the brood has hatched, and the female spawns again in the third summer, 2 years after the previous spawning (Waddy et al., 1995). This paper describes the results of experiments conducted over a 2-year period that determines the lethality of the azamethiphos formulation, Salmosan (47.5% w/w azamethiphos), to lobsters at various times during the biennial molt and reproductive cycle.
2. Methods Adult female lobsters (500 g) were obtained from the commercial fishery at Miminegash, Prince Edward Island (PEI), Canada, in September of 1996 and 1998. From the time of capture until the time of exposure, the lobsters were housed communally with shelter at local seawater temperature (varying seasonally from 0 to 14 1C) and day length. They were fed four or five times per week with a diet consisting of fresh-frozen herring, shrimp, and squid. The 1998 stock were used in exposures done between October 1998 and February 2000 while the lobsters were in late postmolt (stage C3) and intermolt (stage C4) and were either preovigerous or ovigerous (Table 1). Exposures done between April 2000 and October 2000 involved 1996 stock that were nonovigerous and in the molting phase of the 2-year cycle (encompassing molt stages C4, D, A, B, and C1 3). Molt and reproductive stages were assessed prior to the tests using the criteria of Aiken (1973) and Aiken and Waddy (1982). Twenty-four hours prior to testing, lobsters (n=5) were transferred to 200-L glass aquaria filled with 150 L of filtered seawater and placed in a water bath to maintain water temperature throughout the exposures. At the onset of each test the water in each aquarium was treated with a known quantity of Salmosan (47.5% azamethiphos w/w). Although lobsters were exposed to
six concentrations in each test, seven concentrations were used in the study. The lowest concentration was not used in the months with the coldest water temperatures (December, February, and April) and the highest concentration was dropped from the remaining tests. Concentrations tested were approximately 0, 0.4, 0.7, 1.1, 1.8, 3.0, and 5.0 mg/L (nominal, as azamethiphos) and were selected based on previous work with azamethiphos and lobsters (Burridge et al., 1999). These concentrations represent 0–5% of the recommended treatment concentration (100 mg/L). The test water was aerated throughout the experiment and a water sample (1 L) was collected from one test concentration (1.1 mg/ L) at T=0, 3, 6, 12, 24, and 48 h. Water samples were extracted and analyzed for azamethiphos according to Burridge et al. (1999). Mean measured water concentrations were calculated according to Zitko et al. (1977). The 48-h LC50 and confidence interval (CI) were calculated, using measured water concentrations, according to Stephan (1977) by the Spearman–Karber method. This method is appropriate for tests where exposure to a range of concentrations results in very few tests with greater than 0% but less than 100% mortality (so-called partial kills) (Stephan, 1977). In tests where no partial kills were observed and no 95% confidence interval could be calculated, the LC50 was estimated using the binomial method (Stephan, 1977) as the geometric mean of the lowest concentration with greater than 50% mortality and the highest concentration with less than 50% mortality. The confidence interval (93.6%) about this mean is the two concentrations used to calculate the LC50. With only five lobsters being exposed to any concentration, it was decided that the death of one lobster among the five in the control group (0 mg/L) would invalidate the test. Statistical significance of differences between calculated LC50 values was assessed using the method of Sprague and Fogels (1977).
3. Results The measured concentration of azamethiphos in water at the onset of each experiment ranged from 84% to 110% of the nominal concentrations. The degradation curves used to determine the mean concentration over 48 h were similar to that described by Burridge et al. (1999). No lobsters in the control treatment (0 mg/L) died during this study. Fig. 1 shows the 48-h LC50’s and confidence intervals calculated for azamethiphos and adult female lobsters. The calculated LC50 ranges from a low of 0.61 mg/L in late August 2000 to a high of 3.24 mg/L in April 2000. At no time during this study was an LC50 estimate found to be statistically different from the estimated LC50 for the test immediately before
ARTICLE IN PRESS L.E. Burridge et al. / Ecotoxicology and Environmental Safety 60 (2005) 277–281
279
Table 1 Water temperatures, reproductive condition, and molt and setogenic stages of the American lobsters used in each 48-h lethality test Date
1C
Molt and setogenic stages
Reproductive condition and predicted time of spawning
October 21, 1998 December 21, 1998 February 15, 1999 April 27, 1999 June 21, 1999 August 23, 1999 October 26, 1999 December 14, 1999 February 22, 2000 April 25, 2000 June 20, 2000
11.3 6.2 3.0 5.8 11.0 14.0 11.2 8.9 2.0 6.0 10.0t
Preovigerous (summer 1999) Preovigerous (summer 1999) Preovigerous (summer 1999) Preovigerous (summer 1999) 93% preovigerous (summer 1999) 7% ovigerous 13% preovigerous (Summer 1999) 87% ovigerous Ovigerous Ovigerous Ovigerous Summer 2001 Summer 2001
July 17, 2000
13.0
August 1, 2000
14.5
August 9, 2000
13.9
August 24, 2000 October 23, 2000
14.6 11.8
Postmolt (C3) Intermolt (C4) Intermolt (C4) Intermolt (C4) Intermolt (C4) Intermolt (C4) Intermolt (C4) Intermolt (C4) Intermolt (C4) Intermolt (C4) Molt stage D0 Setogenic stages 2.0–2.5 Early to mid premolt (D1) Setogenic stages 3.0–4.0 Late premolt (D2 3) Setogenic stages 5.0–5.5 (5 of 30 molted during test) Late premolt (D2 3) Setogenic stages 5.0–5.5 (4 of 30 molted during test) Postmolt (B and C1 2 ) Postmolt (C3)
Summer 2001 Summer 2001 Summer 2001 Preovigerous (summer 2001) Preovigerous (summer 2001)
Molt and reproductive condition were determined by the methods of Aiken (1973) and Aiken and Waddy (1982). Preovigerous groups are in the reproductive phase of the biennial molt and reproductive cycle and will spawn and incubate eggs before molting.
Seasonal lethality (azamethiphos and female lobsters) 4.5 4
48 h LC50 (µg/L)
3.5 3
*
* 2.5 2
*
1.5 1 0.5 0 Oct-98 Dec- Feb98 99
Apr99
Jun- Aug99 99
Oct- Dec- Feb99 99 00
Apr00
Jun00
Jul- Aug Aug Aug Oct-00 00 (A) 00 (B) 00 (C) 00
Date
Fig. 1. Calculated 48-h LC50 for azamethiphos and adult female lobsters. Bars represent the 95% confidence interval about the mean except where marked with an * which indicates 93.6% CI.
or after (P>0.05). When the results of individual tests were compared to all others, statistically significant (Po0.05) differences are observed. There is a general trend for the lobsters to be the least sensitive to azamethiphos in February and April, to be the most sensitive during June, August, and October, and to show intermediate sensitivity in between. The lowest LC50s were observed in June and August 1999 (during and following the spawning season) and in August and October 2000 (during the molting season). The lowest LC50 (0.61 mg/L) was observed in a group of lobsters that had recently molted and were in molt stages B and C1 2 at the time that they were tested.
Table 1 shows the water temperatures at the beginning of each lethality test and the molt and reproductive stages of the lobsters tested. Water temperatures were essentially the same in each year with the only exception being in December where the water temperature was 6.2 1C in 1998 and 8.9 1C in 1999 (Table 1). There was no significant difference between the LC50’s for these months. In August 2000, some lobsters molted during the tests (see Table 1). Those that molted during the August 1st test were from the control (two), 3.0-mg/L (one), 2.0mg/L (one), and 0.8-mg/L (one) groups. The lobsters that molted in the control and 0.8-mg/L groups
ARTICLE IN PRESS 280
L.E. Burridge et al. / Ecotoxicology and Environmental Safety 60 (2005) 277–281
survived, but the two lobsters that molted in the 2.0 and 3.0-mg/L groups died and were the first in the groups to die at those concentrations. Those that molted during the August 9th test were in the two lowest azamethiphos concentrations. The two lobsters that molted in the 1-mg/L group both died during the test, while the two that molted in the 0.5-mg/L group survived. In tests where less than 100% of the females were carrying eggs (June and August 1999) there was no obvious difference in the sensitivity of ovigerous and nonovigerous lobsters to the treatments.
4. Discussion Several publications have compared the sensitivity of the larval stages of lobsters to contaminants (Wells and Sprague, 1976; Young-Lai et al., 1991; Burridge and Haya, 1997; Burridge et al., 1999, 2000b). No work has been done, however, to compare the sensitivity of female lobsters to chemical contaminants at various times during the 2-year molt and reproductive cycle. Rao and Conklin (1986) reported that molting grass shrimp were more sensitive to a number of organic contaminants and metals than intermolt shrimp and Lee and Buikema (1979) reported increased susceptibility of Daphnia pulex to chromium at, or shortly after, molting. A similar result was reported by Price and Uglow (1979) from tests in which Crangon crangon were exposed to copper and zinc. The 48-h LC50 values reported here are in general agreement with those reported by Burridge et al. (1999). In three lethality tests there were no partial kills (these are identified in Fig. 1 by an asterisk). In these tests the LC50 was calculated using the binomial method with an estimation of the 93.6% CI. This more conservative estimate of the LC50 may affect assessments of statistical significance that are based on 95% CI. Fortunately, these tests did not produce LC50 estimates near either extreme. The pattern of LC50 values is interesting in that there appear to be two distinct times during the biennial molt and reproductive cycle when the lobsters are more sensitive to azamethiphos. These are near the time of spawning (June and August 1999) and during the molting season (August–October 2000). These months have the highest seawater temperatures measured during the study, so the possibility that the differences are related to water temperature cannot be dismissed. However, it seems clear from the August 2000 data that temperature alone is not responsible for the differences in sensitivity. Although, water temperatures remained constant at 14 1C, the LC50 was significantly lower in lobsters that had recently molted (stages B and C1 2) than in those in late premolt (setogenic stages 5.0 and 5.5, Aiken, 1973).
Azamethiphos has a high water solubility (1.1 g/L) and a low octanol–water partition coefficient (log Kow=1.05) (Tomlin, 1997). The octanol–water partition coefficient provides an indication of a chemical’s likely partitioning between water and sediment, its availability from food, and the likelihood of bioaccumulation of the compound by aquatic biota. A value greater than 5.0 is considered indicative of a compound that is likely to persist in the environment or bioaccumulate in tissue (Beek et al., 2000). Azamethiphos is unlikely to accumulate in tissue or in sediment and is more likely to be present in water and therefore available to the lobsters. These data show that time of year of exposure to azamethiphos may have a significant effect on the response of female American lobsters to the compound. The estimated LC50 can vary by as much as 5 depending on molt and reproductive stage of the lobster or seawater temperature. The timing of anti-sea louse treatments is often coincident with the period when lobsters are at their highest concentration near aquaculture operations (Burridge et al., 1999, 2000a). While our data suggest that a single anti-louse treatment will pose no significant lethal threat to spawning or molting lobsters, this can be confirmed only by field studies. In addition, our data do not address questions of repeated exposures or sublethal effects that might also vary with time of year of exposure. Novartis, the producer of azamethiphos, recently applied to Health Canada to have use of their product discontinued. Aquaculturists may continue to use azamethiphos until April 1, 2005 (Cathy Morris, Health Canada, personal communication). Despite the reduction in use of Salmosan in eastern Canada, the variability in the sensitivity of adult female lobsters during the year is an important finding and contributes to the realization that responses of American lobsters to chemicals are complex and difficult to predict as they can vary with time of year and physiological condition of the animals at the time of exposure (this study; Waddy et al., 2003; L.E. Burridge and S.L. Waddy, unpublished data). As more compounds become available to treat salmon against infestations of sea lice and as more chemicals are used in other aquaculture operations it will be necessary not only to report the lethality of these compounds to the American lobster but also to determine the variation in lethality that can occur during the molt and reproductive cycles.
Acknowledgments We thank the following for valuable contributions and participation in this work: Dr. D. Wildish, Dr. F. Page, Ms. Sarah Mercer, Mr. W. Young-Lai, Ms. N.
ARTICLE IN PRESS L.E. Burridge et al. / Ecotoxicology and Environmental Safety 60 (2005) 277–281
Hamilton-Gibson, and Mr. Ken MacKeigan. This project was partially funded by Fisheries and Oceans Canada, Environmental Sciences Strategic Research Fund. References Abgrall, P., Rangeley, W.R., Burridge, L.E., Lawton, P., 2000. Aquaculture 181, 1–10. Aiken, D.E., 1973. J. Fish. Res. Board Can. 30, 1337–1344. Aiken, D.E., Waddy, S.L., 1982. J. Crustacean Biol. 2, 315–327. Beek, B., Bohling, S., Bruckmann, U., Franke, C., Johncke, U., Studinger, G., 2000. The assessment of bioaccumulation. In: Beek, B. (Ed.), The Handbook of Environmental Chemistry, vol. 2 (Part J), Bioaccumulation, New Aspects and Developments. Springer, Berlin, pp. 239–276. Burridge, L.E., Haya, K., 1997. Ecotoxicol. Environ. Saf. 38, 150–154. Burridge, L.E., Haya, K., Zitko, V., Waddy, S., 1999. Ecotoxicol. Environ. Saf. 43, 165–169. Burridge, L.E., Haya, K., Waddy, S.L., Wade, J., 2000a. Aquaculture 182 (1–2), 27–35. Burridge, L.E., Haya, K., Page, F., Waddy, S.L., Zitko, V., Wade, J., 2000b. Aquaculture 182, 37–47. Health Canada, 2003a. http://www.hc-sc.gc.ca/vetdrugs-medsvet/ aquaculture_english.html (accessed June 6, 2003). Health Canada, 2003b. http://www.eddenet.pmra-arla.gc.ca/4.0/ 4.1.asp (accessed June 6, 2003). Lee, D.R., Buikema Jr., A.L., 1979. J. Fish. Res. Board Can. 36, 1129–1133. Price, R.K.J., Uglow, R.F., 1979. Mar. Environ. Res. 2, 287–299. Rao, K.R., Conklin, P.J., 1986. Molt-related susceptibility and regenerative limb growth as sensitive indicators of aquatic
281
pollutant toxicity to crustaceans. In: Thompson, M.-F., Sarojini, R., Nagabhushanam, E. (Eds.), Biology of Benthic Marine Organisms. Techniques and Methods as Applied to the Indian Ocean. Indian Ed. Ser. 12, pp. 523–534. Sprague, J.B., Fogels, A., 1977. Watch the Y in bioassay. In: Parker, W.R., Pessah, E., Wells, P.G., Westlake, G.F. (Ed.), Environmental Protection Service Technical Report No. EPS-5-AR-77-1, Halifax, pp. 107–118. Stephan, C.E., 1977. Methods for calculating an LC50. In: Mayer, F.L., Hamelink, J.L. (Eds.), Aquatic Toxicology and Hazard Evaluation, ASTM STP 634. American Society for Testing and Materials, Philadelphia, pp. 65–84. Tomlin, C.D.S. (Ed.), 1997. The Pesticide Manual—A World Compendium. British Crop Protection Council, Surrey. Waddy, S.L., Aiken, D.E., 2000. Endocrinology and the culture of homarid lobsters. In: Fingerman, M., Nagabhushanam, R. (Eds.), Recent Advances in Marine Biotechnology. Aquaculture (Part A, Seaweeds and Invertebrates), Vol. 4. Science Publishers Inc., Enfield, pp. 195–247. Waddy, S.L., Aiken, D.E., de Kleijn, D.P.V., 1995. Control of growth and reproduction. In: Factor, J.R. (Ed.), Biology of the Lobster, Homarus americanus. Academic Press Inc., San Diego, pp. 217–266. Waddy, S.L., Hamilton, M.N., Burridge, L.E., Mercer, S.M., Aiken, D.E., Haya, K., 2003. Aquacult. Assoc. Can. Spec. Publ. 6, 75–77. Wells, P.G., Sprague, J.B., 1976. J. Fish. Res. Board Can. 33, 1604–1614. Young-Lai, W.W., Charmantier-Daures, M., Charmantier, G., 1991. Mar. Biol. 110, 293–300. Zitko, V., Carson, W.G., Metcalfe, C.D., 1977. Bull. Environ. Contam. Toxicol. 18, 35–41.
Ecotoxicology and Environmental Safety 60 (2005) 277–281 www.elsevier.com/locate/ecoenv
Seasonal lethality of the organophosphate pesticide, azamethiphos to female American lobster (Homarus americanus) L.E. Burridgea,, K. Hayaa, S.L Waddyb a
Marine Environmental Sciences Division, Fisheries and Oceans Canada, St. Andrews Biological Station, 531 Brandy Cove Rd., St. Andrews, NB, Canada E5B 2L9 b Aquaculture Division, Fisheries and Oceans Canada, St. Andrews Biological Station, 531 Brandy Cove Rd., St. Andrews, NB, Canada E5B 2L9 Received 16 June 2003; received in revised form 28 May 2004; accepted 8 June 2004 Available online 17 July 2004
Abstract The organophosphate pesticide azamethiphos is the active ingredient in Salmosan, a product formerly registered in Canada for the treatment of cultured Atlantic salmon against infestations of the ectoparasite Lepeophtheirus salmonis. The 48-h LC50 of azamethiphos to female American lobsters was determined bimonthly for 2 years to determine whether the sensitivity of lobsters to azamethiphos varied with time of year, molt stage, or reproductive stage. The LC50’s ranged from 0.61 to 3.24 mg/L. The lobsters were most sensitive to azamethiphos during the spawning and molting seasons which occur in the summer and early fall when seawater temperatures are highest. Testing of compounds on this species for regulatory purposes should take into account that there may be variations in sensitivity during the molt and reproductive cycles. Crown Copyright r 2004 Published by Elsevier Inc. All rights reserved. Keywords: Sea lice pesticides; Azamethiphos; Salmosan; American lobster; Lethality; Spawning; Molting
1. Introduction Since 1994 the salmon aquaculture industry in Atlantic Canada has dealt with a succession of parasite and disease problems. Changes in management practices have resulted in greatly improved husbandry and a reduction in the use of some chemicals. However, fish farmers remain reliant on chemotherapeutants to combat infestations of the most problematic ectoparasite, Lepeophtheirus salmonis, a marine copepod commonly called sea lice. Salmosan was registered for use in Canada as an antisea louse treatment until 2003 (Health Canada, 2003a, b). Salmosan is a wettable powder containing 47.5% (w/w) of azamethiphos (S-6-chloro-2, 3-dihydro2-oxo-1, 3-oxazolo[4,5-b] pyridin-3-ylmethyl) O,O-dimethyl phosphorothioate), an organophosphate pestiCorresponding author. Fax: +506-529-5862.
E-mail address: [email protected] (L.E. Burridge).
cide. Salmon are given a bath treatment with 100 mg/L of azamethiphos for up to 1 h. Health Canada sanctions the use of other anti-sea louse compounds under its emergency drug release policy and access to these treatments appears to have reduced the use of azamethiphos in eastern Canada. Sales of Salmosan have dropped from 89 kg in 1998 to a low of 6 kg in 2001. In 2002 there was an increase in sales to 31 kg (M. Boldon, New Brunswick Department of Environment and Local Government, personal communication). Several papers describing the effects of azamethiphos on indigenous species that might be exposed to the compound after routine anti-sea louse treatments have been published (Burridge et al., 1999, 2000a; Abgrall et al., 2000). These studies focused on the American lobster, a commercially important crustacean fished in areas close to salmon aquaculture sites. A potential weakness with each of these studies is that the experiments were conducted only at one time of year and did not take into account that seawater tempera-
0147-6513/$ - see front matter Crown Copyright r 2004 Published by Elsevier Inc. All rights reserved. doi:10.1016/j.ecoenv.2004.06.004
ARTICLE IN PRESS 278
L.E. Burridge et al. / Ecotoxicology and Environmental Safety 60 (2005) 277–281
ture, the physiological condition of animals, or time of year might affect sensitivity to the compound. Crustaceans are well known to be more sensitive to some chemicals near the time of molt. Rao and Conklin (1986), for example, reported that several organic pollutants were more toxic to molting grass shrimp (Palaemonetes pugio) than to intermolt shrimp. To our knowledge, no studies have considered whether time of year or reproductive stage affects the response of crustaceans to chemicals. The American lobster displays marked seasonal changes in sensitivity to environmental and endocrine manipulation (Waddy and Aiken, 2000) and sublethal responses to some compounds appear to vary with time of year and reproductive condition (Waddy et al., 2003, unpublished). Female American lobsters typically spawn and molt on a biennial cycle. Eggs are extruded in the late spring and summer, molting occurs the following year after the brood has hatched, and the female spawns again in the third summer, 2 years after the previous spawning (Waddy et al., 1995). This paper describes the results of experiments conducted over a 2-year period that determines the lethality of the azamethiphos formulation, Salmosan (47.5% w/w azamethiphos), to lobsters at various times during the biennial molt and reproductive cycle.
2. Methods Adult female lobsters (500 g) were obtained from the commercial fishery at Miminegash, Prince Edward Island (PEI), Canada, in September of 1996 and 1998. From the time of capture until the time of exposure, the lobsters were housed communally with shelter at local seawater temperature (varying seasonally from 0 to 14 1C) and day length. They were fed four or five times per week with a diet consisting of fresh-frozen herring, shrimp, and squid. The 1998 stock were used in exposures done between October 1998 and February 2000 while the lobsters were in late postmolt (stage C3) and intermolt (stage C4) and were either preovigerous or ovigerous (Table 1). Exposures done between April 2000 and October 2000 involved 1996 stock that were nonovigerous and in the molting phase of the 2-year cycle (encompassing molt stages C4, D, A, B, and C1 3). Molt and reproductive stages were assessed prior to the tests using the criteria of Aiken (1973) and Aiken and Waddy (1982). Twenty-four hours prior to testing, lobsters (n=5) were transferred to 200-L glass aquaria filled with 150 L of filtered seawater and placed in a water bath to maintain water temperature throughout the exposures. At the onset of each test the water in each aquarium was treated with a known quantity of Salmosan (47.5% azamethiphos w/w). Although lobsters were exposed to
six concentrations in each test, seven concentrations were used in the study. The lowest concentration was not used in the months with the coldest water temperatures (December, February, and April) and the highest concentration was dropped from the remaining tests. Concentrations tested were approximately 0, 0.4, 0.7, 1.1, 1.8, 3.0, and 5.0 mg/L (nominal, as azamethiphos) and were selected based on previous work with azamethiphos and lobsters (Burridge et al., 1999). These concentrations represent 0–5% of the recommended treatment concentration (100 mg/L). The test water was aerated throughout the experiment and a water sample (1 L) was collected from one test concentration (1.1 mg/ L) at T=0, 3, 6, 12, 24, and 48 h. Water samples were extracted and analyzed for azamethiphos according to Burridge et al. (1999). Mean measured water concentrations were calculated according to Zitko et al. (1977). The 48-h LC50 and confidence interval (CI) were calculated, using measured water concentrations, according to Stephan (1977) by the Spearman–Karber method. This method is appropriate for tests where exposure to a range of concentrations results in very few tests with greater than 0% but less than 100% mortality (so-called partial kills) (Stephan, 1977). In tests where no partial kills were observed and no 95% confidence interval could be calculated, the LC50 was estimated using the binomial method (Stephan, 1977) as the geometric mean of the lowest concentration with greater than 50% mortality and the highest concentration with less than 50% mortality. The confidence interval (93.6%) about this mean is the two concentrations used to calculate the LC50. With only five lobsters being exposed to any concentration, it was decided that the death of one lobster among the five in the control group (0 mg/L) would invalidate the test. Statistical significance of differences between calculated LC50 values was assessed using the method of Sprague and Fogels (1977).
3. Results The measured concentration of azamethiphos in water at the onset of each experiment ranged from 84% to 110% of the nominal concentrations. The degradation curves used to determine the mean concentration over 48 h were similar to that described by Burridge et al. (1999). No lobsters in the control treatment (0 mg/L) died during this study. Fig. 1 shows the 48-h LC50’s and confidence intervals calculated for azamethiphos and adult female lobsters. The calculated LC50 ranges from a low of 0.61 mg/L in late August 2000 to a high of 3.24 mg/L in April 2000. At no time during this study was an LC50 estimate found to be statistically different from the estimated LC50 for the test immediately before
ARTICLE IN PRESS L.E. Burridge et al. / Ecotoxicology and Environmental Safety 60 (2005) 277–281
279
Table 1 Water temperatures, reproductive condition, and molt and setogenic stages of the American lobsters used in each 48-h lethality test Date
1C
Molt and setogenic stages
Reproductive condition and predicted time of spawning
October 21, 1998 December 21, 1998 February 15, 1999 April 27, 1999 June 21, 1999 August 23, 1999 October 26, 1999 December 14, 1999 February 22, 2000 April 25, 2000 June 20, 2000
11.3 6.2 3.0 5.8 11.0 14.0 11.2 8.9 2.0 6.0 10.0t
Preovigerous (summer 1999) Preovigerous (summer 1999) Preovigerous (summer 1999) Preovigerous (summer 1999) 93% preovigerous (summer 1999) 7% ovigerous 13% preovigerous (Summer 1999) 87% ovigerous Ovigerous Ovigerous Ovigerous Summer 2001 Summer 2001
July 17, 2000
13.0
August 1, 2000
14.5
August 9, 2000
13.9
August 24, 2000 October 23, 2000
14.6 11.8
Postmolt (C3) Intermolt (C4) Intermolt (C4) Intermolt (C4) Intermolt (C4) Intermolt (C4) Intermolt (C4) Intermolt (C4) Intermolt (C4) Intermolt (C4) Molt stage D0 Setogenic stages 2.0–2.5 Early to mid premolt (D1) Setogenic stages 3.0–4.0 Late premolt (D2 3) Setogenic stages 5.0–5.5 (5 of 30 molted during test) Late premolt (D2 3) Setogenic stages 5.0–5.5 (4 of 30 molted during test) Postmolt (B and C1 2 ) Postmolt (C3)
Summer 2001 Summer 2001 Summer 2001 Preovigerous (summer 2001) Preovigerous (summer 2001)
Molt and reproductive condition were determined by the methods of Aiken (1973) and Aiken and Waddy (1982). Preovigerous groups are in the reproductive phase of the biennial molt and reproductive cycle and will spawn and incubate eggs before molting.
Seasonal lethality (azamethiphos and female lobsters) 4.5 4
48 h LC50 (µg/L)
3.5 3
*
* 2.5 2
*
1.5 1 0.5 0 Oct-98 Dec- Feb98 99
Apr99
Jun- Aug99 99
Oct- Dec- Feb99 99 00
Apr00
Jun00
Jul- Aug Aug Aug Oct-00 00 (A) 00 (B) 00 (C) 00
Date
Fig. 1. Calculated 48-h LC50 for azamethiphos and adult female lobsters. Bars represent the 95% confidence interval about the mean except where marked with an * which indicates 93.6% CI.
or after (P>0.05). When the results of individual tests were compared to all others, statistically significant (Po0.05) differences are observed. There is a general trend for the lobsters to be the least sensitive to azamethiphos in February and April, to be the most sensitive during June, August, and October, and to show intermediate sensitivity in between. The lowest LC50s were observed in June and August 1999 (during and following the spawning season) and in August and October 2000 (during the molting season). The lowest LC50 (0.61 mg/L) was observed in a group of lobsters that had recently molted and were in molt stages B and C1 2 at the time that they were tested.
Table 1 shows the water temperatures at the beginning of each lethality test and the molt and reproductive stages of the lobsters tested. Water temperatures were essentially the same in each year with the only exception being in December where the water temperature was 6.2 1C in 1998 and 8.9 1C in 1999 (Table 1). There was no significant difference between the LC50’s for these months. In August 2000, some lobsters molted during the tests (see Table 1). Those that molted during the August 1st test were from the control (two), 3.0-mg/L (one), 2.0mg/L (one), and 0.8-mg/L (one) groups. The lobsters that molted in the control and 0.8-mg/L groups
ARTICLE IN PRESS 280
L.E. Burridge et al. / Ecotoxicology and Environmental Safety 60 (2005) 277–281
survived, but the two lobsters that molted in the 2.0 and 3.0-mg/L groups died and were the first in the groups to die at those concentrations. Those that molted during the August 9th test were in the two lowest azamethiphos concentrations. The two lobsters that molted in the 1-mg/L group both died during the test, while the two that molted in the 0.5-mg/L group survived. In tests where less than 100% of the females were carrying eggs (June and August 1999) there was no obvious difference in the sensitivity of ovigerous and nonovigerous lobsters to the treatments.
4. Discussion Several publications have compared the sensitivity of the larval stages of lobsters to contaminants (Wells and Sprague, 1976; Young-Lai et al., 1991; Burridge and Haya, 1997; Burridge et al., 1999, 2000b). No work has been done, however, to compare the sensitivity of female lobsters to chemical contaminants at various times during the 2-year molt and reproductive cycle. Rao and Conklin (1986) reported that molting grass shrimp were more sensitive to a number of organic contaminants and metals than intermolt shrimp and Lee and Buikema (1979) reported increased susceptibility of Daphnia pulex to chromium at, or shortly after, molting. A similar result was reported by Price and Uglow (1979) from tests in which Crangon crangon were exposed to copper and zinc. The 48-h LC50 values reported here are in general agreement with those reported by Burridge et al. (1999). In three lethality tests there were no partial kills (these are identified in Fig. 1 by an asterisk). In these tests the LC50 was calculated using the binomial method with an estimation of the 93.6% CI. This more conservative estimate of the LC50 may affect assessments of statistical significance that are based on 95% CI. Fortunately, these tests did not produce LC50 estimates near either extreme. The pattern of LC50 values is interesting in that there appear to be two distinct times during the biennial molt and reproductive cycle when the lobsters are more sensitive to azamethiphos. These are near the time of spawning (June and August 1999) and during the molting season (August–October 2000). These months have the highest seawater temperatures measured during the study, so the possibility that the differences are related to water temperature cannot be dismissed. However, it seems clear from the August 2000 data that temperature alone is not responsible for the differences in sensitivity. Although, water temperatures remained constant at 14 1C, the LC50 was significantly lower in lobsters that had recently molted (stages B and C1 2) than in those in late premolt (setogenic stages 5.0 and 5.5, Aiken, 1973).
Azamethiphos has a high water solubility (1.1 g/L) and a low octanol–water partition coefficient (log Kow=1.05) (Tomlin, 1997). The octanol–water partition coefficient provides an indication of a chemical’s likely partitioning between water and sediment, its availability from food, and the likelihood of bioaccumulation of the compound by aquatic biota. A value greater than 5.0 is considered indicative of a compound that is likely to persist in the environment or bioaccumulate in tissue (Beek et al., 2000). Azamethiphos is unlikely to accumulate in tissue or in sediment and is more likely to be present in water and therefore available to the lobsters. These data show that time of year of exposure to azamethiphos may have a significant effect on the response of female American lobsters to the compound. The estimated LC50 can vary by as much as 5 depending on molt and reproductive stage of the lobster or seawater temperature. The timing of anti-sea louse treatments is often coincident with the period when lobsters are at their highest concentration near aquaculture operations (Burridge et al., 1999, 2000a). While our data suggest that a single anti-louse treatment will pose no significant lethal threat to spawning or molting lobsters, this can be confirmed only by field studies. In addition, our data do not address questions of repeated exposures or sublethal effects that might also vary with time of year of exposure. Novartis, the producer of azamethiphos, recently applied to Health Canada to have use of their product discontinued. Aquaculturists may continue to use azamethiphos until April 1, 2005 (Cathy Morris, Health Canada, personal communication). Despite the reduction in use of Salmosan in eastern Canada, the variability in the sensitivity of adult female lobsters during the year is an important finding and contributes to the realization that responses of American lobsters to chemicals are complex and difficult to predict as they can vary with time of year and physiological condition of the animals at the time of exposure (this study; Waddy et al., 2003; L.E. Burridge and S.L. Waddy, unpublished data). As more compounds become available to treat salmon against infestations of sea lice and as more chemicals are used in other aquaculture operations it will be necessary not only to report the lethality of these compounds to the American lobster but also to determine the variation in lethality that can occur during the molt and reproductive cycles.
Acknowledgments We thank the following for valuable contributions and participation in this work: Dr. D. Wildish, Dr. F. Page, Ms. Sarah Mercer, Mr. W. Young-Lai, Ms. N.
ARTICLE IN PRESS L.E. Burridge et al. / Ecotoxicology and Environmental Safety 60 (2005) 277–281
Hamilton-Gibson, and Mr. Ken MacKeigan. This project was partially funded by Fisheries and Oceans Canada, Environmental Sciences Strategic Research Fund. References Abgrall, P., Rangeley, W.R., Burridge, L.E., Lawton, P., 2000. Aquaculture 181, 1–10. Aiken, D.E., 1973. J. Fish. Res. Board Can. 30, 1337–1344. Aiken, D.E., Waddy, S.L., 1982. J. Crustacean Biol. 2, 315–327. Beek, B., Bohling, S., Bruckmann, U., Franke, C., Johncke, U., Studinger, G., 2000. The assessment of bioaccumulation. In: Beek, B. (Ed.), The Handbook of Environmental Chemistry, vol. 2 (Part J), Bioaccumulation, New Aspects and Developments. Springer, Berlin, pp. 239–276. Burridge, L.E., Haya, K., 1997. Ecotoxicol. Environ. Saf. 38, 150–154. Burridge, L.E., Haya, K., Zitko, V., Waddy, S., 1999. Ecotoxicol. Environ. Saf. 43, 165–169. Burridge, L.E., Haya, K., Waddy, S.L., Wade, J., 2000a. Aquaculture 182 (1–2), 27–35. Burridge, L.E., Haya, K., Page, F., Waddy, S.L., Zitko, V., Wade, J., 2000b. Aquaculture 182, 37–47. Health Canada, 2003a. http://www.hc-sc.gc.ca/vetdrugs-medsvet/ aquaculture_english.html (accessed June 6, 2003). Health Canada, 2003b. http://www.eddenet.pmra-arla.gc.ca/4.0/ 4.1.asp (accessed June 6, 2003). Lee, D.R., Buikema Jr., A.L., 1979. J. Fish. Res. Board Can. 36, 1129–1133. Price, R.K.J., Uglow, R.F., 1979. Mar. Environ. Res. 2, 287–299. Rao, K.R., Conklin, P.J., 1986. Molt-related susceptibility and regenerative limb growth as sensitive indicators of aquatic
281
pollutant toxicity to crustaceans. In: Thompson, M.-F., Sarojini, R., Nagabhushanam, E. (Eds.), Biology of Benthic Marine Organisms. Techniques and Methods as Applied to the Indian Ocean. Indian Ed. Ser. 12, pp. 523–534. Sprague, J.B., Fogels, A., 1977. Watch the Y in bioassay. In: Parker, W.R., Pessah, E., Wells, P.G., Westlake, G.F. (Ed.), Environmental Protection Service Technical Report No. EPS-5-AR-77-1, Halifax, pp. 107–118. Stephan, C.E., 1977. Methods for calculating an LC50. In: Mayer, F.L., Hamelink, J.L. (Eds.), Aquatic Toxicology and Hazard Evaluation, ASTM STP 634. American Society for Testing and Materials, Philadelphia, pp. 65–84. Tomlin, C.D.S. (Ed.), 1997. The Pesticide Manual—A World Compendium. British Crop Protection Council, Surrey. Waddy, S.L., Aiken, D.E., 2000. Endocrinology and the culture of homarid lobsters. In: Fingerman, M., Nagabhushanam, R. (Eds.), Recent Advances in Marine Biotechnology. Aquaculture (Part A, Seaweeds and Invertebrates), Vol. 4. Science Publishers Inc., Enfield, pp. 195–247. Waddy, S.L., Aiken, D.E., de Kleijn, D.P.V., 1995. Control of growth and reproduction. In: Factor, J.R. (Ed.), Biology of the Lobster, Homarus americanus. Academic Press Inc., San Diego, pp. 217–266. Waddy, S.L., Hamilton, M.N., Burridge, L.E., Mercer, S.M., Aiken, D.E., Haya, K., 2003. Aquacult. Assoc. Can. Spec. Publ. 6, 75–77. Wells, P.G., Sprague, J.B., 1976. J. Fish. Res. Board Can. 33, 1604–1614. Young-Lai, W.W., Charmantier-Daures, M., Charmantier, G., 1991. Mar. Biol. 110, 293–300. Zitko, V., Carson, W.G., Metcalfe, C.D., 1977. Bull. Environ. Contam. Toxicol. 18, 35–41.