Chemosphere 74 (2008) 70–77 Contents lists available at ScienceDirect Chemosphere j o u r n a l h o m e p a g e : w w w . e l s e v i e r. c o m / l o c a t e / ch e m o s p h e r e Effects of spinosad and Bacillus thuringiensis israelensis on a natural population of Daphnia pulex in field microcosms C. Duchet a,b, M. Larroque c, Th. Caquet b, E. Franquet d, C. Lagneau a, L. Lagadic b,* Entente Interdépartementale de Démoustication du Littoral Méditerranéen, 165 Avenue Paul-Rimbaud, F-34184 Montpellier, France INRA, UMR985 Écologie et Santé des Écosystèmes, Équipe Écotoxicologie et Qualité des Milieux Aquatiques, 65 rue de Saint Brieuc, F-35042 Rennes, France c Laboratoire de Chimie Analytique, UMR Qualisud, Faculté de Pharmacie – Université Montpellier I, 15 Avenue Charles-Flahault, BP14491, F-34093 Montpellier Cedex 5, France d Université Paul Cézanne, Institut Méditerranéen d’Écologie et de Paléoécologie, Faculté des Sciences et Techniques Saint Jérôme, C31, F-13397 Marseille, France a b a r t i c l e i n f o Article history: Received 4 April 2008 Received in revised form 12 August 2008 Accepted 4 September 2008 Available online 1 November 2008 Keywords: Biopesticide In situ microcosm Dissipation kinetics Mosquito control Ecological risk assessment a b s t r a c t Spinosad, a candidate biological larvicide for mosquito control, was evaluated for its effects on a field population of Daphnia pulex, using Bacillus thuringiensis serovar israelensis (Bti) as a reference larvicide. Microcosms (125 L enclosures) were placed in a shallow temporary oligohaline marsh where D. pulex was present. Three concentrations of spinosad (8, 17 and 33 lg L¡1) and two concentrations of Bti (0.16 and 0.50 lL L¡1) were applied (5 replicates per concentration, including the controls). Effects of larvicides on D. pulex were evaluated after 2, 4, 7, 14 and 21 d of exposure, through measurements of abundance and individual size. Dissipation of spinosad from the water phase was rapid. Four days after treatment, residue concentration represented 11.8%, 3.9% and 12.7% of the initial exposure level for the nominal concentrations of 8, 17 and 33 lg L¡1, respectively. Spinosyns A and D dissipated at similar rates. Analysis of abundance and size structure of the D. pulex population showed an impact of spinosad. Both survival and size structure were affected. However, at the lowest concentration (8 lg L¡1), population recovered after the first week. In microcosms treated with Bti, the abundance of D. pulex was not affected but the size structure of the population changed after 21 d. As compared to laboratory tests, the use of in situ microcosms improved the environmental risk assessment of larvicides, taking into account the influence of environmental factors (e.g., temperature, light, salinity) and intrinsic capacity of recovery of D. pulex under field conditions. © 2008 Elsevier Ltd. All rights reserved. 1. Introduction Chemicals used for mosquito control may be applied as larvicides or adulticides. Larvicides are introduced into aquatic ecosystems where mosquito larvae develop (marshes, ponds, sanitation devices) whereas adulticides are sprayed into the atmosphere to kill flying mosquitoes. The use of larvicides optimizes the efficiency of chemical control because mosquito larvae cannot escape from the treated water. However, treatment of aquatic ecosystems raises the question of possible effects of larvicides on non-target aquatic species. To be used in the European Union, larvicides must fulfil the criteria of the “Biocides Directive” 98/8/EC, ensuring an adequate level of protection for humans and the environment. As a consequence, many active substances used to control mosquito larvae have been withdrawn from the market. For example, the organophosphorus insecticide temephos was banned in 2007 in Europe. The number of available larvicides is therefore decreasing, and there is an urgent need to identify environment-friendly compounds. * Corresponding author. Tel.: +33 223 485 237; fax: +33 223 485 440. E-mail address: Laurent.Lagadic@rennes.inra.fr (L. Lagadic). 0045-6535/$ - see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2008.09.024 In such a context, a number of natural products have been developed as insecticides, some with specificity to certain insect taxa. For example, pyrethrum and neem are well established commercially, and pesticides based on plant essential oils have recently entered the marketplace (Isman, 2006). Another bioinsecticide, the bacterial larvicide Bacillus thuringiensis serovar israelensis (Bti), is well-known for its selectivity for Nematocera dipterans. Moreover, the highest sensitivity is observed within a restricted number of families, namely Culicidae (mosquitoes), Simuliidae (black flies) and Chironomidae (non-biting midges) (Boisvert and Boisvert, 2000). This larvicide is therefore widely used for mosquito control all over the world (Boisvert and Lacoursière, 2004). Bti is a positive Gram (Bacillaceae) and endospore-forming aerobic bacterium. During the spore-forming stage of its life-cycle, Bti produces a protein crystal. When ingested by mosquito larvae, the crystal is dissolved, due to the alkaline environment of insect gut, and converted into toxic proteins (Ali, 1980; Garcia et al., 1981; Merritt et al., 1989; He and Ong, 2000). The toxins bind to specific receptors of the epithelial cell wall, causing membrane perforations of the gut, leaking of gut internal fluids, and eventually death (Van Frankenhuyzen, 1993). A new biological insecticide that seems promising for mosquito control is spinosad (DowElanco, Indianapolis, IN, USA). C. Duchet et al. / Chemosphere 74 (2008) 70–77 Spinosad is a mixture of spinosyns A and D known as fermentation products of a soil bacterium (Saccharopolyspora spinosa, Actinomycetes) (Crouse et al., 2001). Spinosad acts as a contact and stomach poison (DowElanco, 1996; Salgado, 1998). It persistently stimulates the insect central nervous system by interacting with nicotinic acetylcholine receptors through a mechanism distinct from those of other nicotinic agonists (Watson, 2001). Up to now, it has been used against lepidopteran, dipteran, and thysanopteran pests, and it is used for integrated pest management (IPM) programmes in vegetables and ornamentals (Pineda et al., 2006). Spinosad is also a potential candidate for mosquito control (Cetin et al., 2005). Although it is considered as a selective insecticide for insect pest species (Miles and Dutton, 2000), some studies indicate that it may be toxic to beneficial species (Nasreen et al., 2000; Tillman and Mulrooney, 2000; Consoli et al., 2001), and previous studies demonstrated effects of spinosad on cohort dynamics of the zooplanktonic crustacean Daphnia pulex (Crustacea, Cladocera) in laboratory conditions (Stark and Vargas, 2003). Ecological risk assessment is usually based on toxicity studies conducted on single species in laboratory. Although these tests may allow mortality, growth, reproduction and behaviour of individuals to be estimated, they can rarely be used to assess environmental hazards at higher levels of biological organization. As natural ecosystems are more complex and variable than laboratory standardised systems (Burnett and Liss, 1990; Rand and Clark, 2000), laboratory toxicity bioassays should be complemented with higher tier assessment in field conditions. In situ microcosms or enclosures take into account physicochemical and biological natural variability which is not possible with single-species laboratory toxicity tests designed to achieve homogenous test conditions (van den Brink et al., 2005). The ecological risk of insecticides in the field following a treatment depends on their environmental fate and on the ability of non-target species to recover their pre-treatment population size (Medina et al., 2004; Caquet et al., 2007). There is a large amount of data available on the effects of Bti on non-target species (Boisvert and Boisvert, 2000; Boisvert and Lacoursière, 2004), but few studies on effects in the field, especially in wetlands, have been published (Purcell, 1981; Hershey et al., 1998). In the case of spinosad, the published studies were carried out in greenhouses (Miles, 2003; Holt et al., 2006), and none were performed in natural aquatic ecosystems. The present study was therefore undertaken to assess the impact of spinosad on a field population of Daphnia pulex, with Bti as the reference compound. In situ microcosms were used to enclose parts of this natural population, in order to control exposure to the larvicides. Changes in spinosyns A and D concentrations in water were monitored, and D. pulex population-level effects were concomitantly assessed by analysing abundance and sizestructure for increasing levels of exposure. Results are discussed with regard to the interest of using in situ microcosms in the ecological risk assessment of mosquito larvicides. 2. Materials and methods 2.1. Microcosms The microcosms, 125 L cube-shaped plexiglas enclosures (50 £ 50 £ 50 cm), were placed in a shallow temporary oligohaline marsh located in Le Tour du Parc (Morbihan, Brittany, France; 47°319250N–02°389280O). They were pushed into the sediment surface (5–10 cm deep) to avoid leaking of contaminated water from the enclosures where the insecticides were applied. Thirty units were used in this study. Microcosms were allowed to stabilize for 24 h before larvicide application. 71 2.2. Experimental design Ten microcosms were treated with Bti (Vectobac® 12AS; Valent Biosciences, Libertyville, IL, USA), fifteen microcosms were treated with spinosad (Spinosad 120SC; DowElanco, Indianapolis, IN, USA), and five microcosms remained as untreated controls. Treatments were randomly assigned to the microcosms using a random number table (R for Windows Version 2.7.0). Vectobac® 12AS was applied at 0.8 and 2.5 L ha¡1 (nominal concentration for 30 cm water depth: 0.16 and 0.50 lL L¡1, respectively), each concentration being applied to 5 enclosures (replicates). These concentrations correspond to the minimum recommended and the maximum registered rates for terrestrial and aerial treatments, respectively (ACTA, 2007). Spinosad was applied as a suspension concentrate formulation containing 120 g active substance per litre at 25, 50 and 100 g ha¡1 (nominal concentration for 30 cm water depth: 8, 17 and 33 lg L¡1, respectively). The treatment rates were chosen in order to encompass the rate of 50 g ha¡1 which would be the mean presumed recommended rate for field application. Five replicates were used for each spinosad concentration. The treatments were performed on May 30, 2006. Each insecticide was diluted in tap water before spraying at the water surface using a portable spraying apparatus. To prevent cross-contamination, the treated microcosms were covered with a PVC plate with a hole in its centre to allow the spray to enter. Additionally, adjacent microcosms were covered with PVC plates while spraying. Monitoring started just before the treatments (day 0), and was carried on 2, 4, 7, 14 and 21 d after insecticide spraying. 2.3. Water quality parameters At each sampling date, the water temperature, dissolved oxygen, salinity, and pH were measured in every enclosure at ca. 5 cm below the water surface, using portable probe apparatus (Wissenschaftlich-Technische-Werkstätten – WTW, Champagne au Mont d’Or, France). Water level was measured in every microcosm using a graduated aluminium gauge. Measurements were always performed between 10:00 and 12:00 AM to ensure data consistency in the event of diurnal effects. Suspended Matter (SM) concentration was determined in 250 mL water samples filtered through pre-weighed oven-dried Whatman GF/C fiberglass filters (2 h at 500 °C; 1.2-lm mesh size; Whatman International, Maidstone, UK) and weighed according to the AFNOR (1996) method after 48 h at 105 °C. Chlorophyll a levels were determined in 250 mL water samples filtered through Whatman GF/C fiberglass filters. Pigments were extracted overnight using 5 mL of an acetone/distilled water (90/10, v/v) mixture. Chlorophyll a was quantified spectrophotometrically (UVK-LAB Technologies, Trappes, France) according to Lorenzen (1967). 2.4. Spinosad residue analysis To determine spinosad concentrations in treated microcosms, water samples were collected 15 min after treatment and at each subsequent sampling date in 3 randomly chosen treated microcosms per concentration. Samples were taken at mid-depth, using 250 mL glass amber bottles. They were stored at -20 °C until analysis. From each sample, 100 mL were acidified with 6 mL of HCl (pH 2). Spinosad was extracted three times with 50 mL dichloromethane. The pooled extracts were evaporated to dryness at 30 °C under a nitrogen flow, and residues were resuspended in 1 mL acetonitrile. Samples of 50 lL were then injected into a HPLC device (Thermoquest P4000) equipped with a UV detector set at 243 nm, and an Eclipse XDB C8 column (150 £ 4.6 mm, 5 lm, Agilent Technologies, Santa Clara, CA, USA). A mobile phase consisting of acetonitrile and ammonium formate buffer at 150 mg L¡1 (90:10, v/v) C. Duchet et al. / Chemosphere 74 (2008) 70–77 72 was delivered at 0.8 mL min¡1 flow rate. Calibration curves were established using stock solutions of spinosyns A and D (150 and 165 mg L¡1, respectively) by successive dilutions in acetonitrile in order to obtain concentrations ranging from 75 to 1500 ng mL¡1 and from 82.5 to 1650 ng mL¡1 for spinosyns A and D, respectively. The response was linear in the range of concentrations tested (data not shown). Under these conditions, the limit of quantification was 0.2 lg L¡1. 2.5. Sampling procedures and endpoints in daphnids Daphnia pulex samples were taken using home-made PVC tube samplers (70 cm length, 6 cm inner diameter) equipped with a 2 £ 4 mm mesh screen-covered one-way valve at the bottom. Water column samples were collected from twenty regularly spaced locations within each enclosure in order to reduce the effects of plankton patchiness (Stephenson et al., 1984; SETAC, 1991). The resulting composite sample (mean ± S.E. volume = 120 ± 26 mL, depending on the water level in the microcosm) was filtered through 30-lm mesh nylon net. The retained organisms (daphnids and some other pelagic invertebrates) were transferred to a 500 mL plastic vial and preserved using neutral aqueous formaldehyde/sucrose (4%, v/v; 40 g L¡1) containing 250 lg L¡1 Bengal pink dye. All the D. pulex found in the samples were identified using the taxonomic key proposed by Amoros (1984). They were counted using a stereomicroscope (Stemi SV 6, Zeiss, Thornwood, NY, USA) and their body length was measured from the eye to base of the tail spine (Boronat and Miracle, 1997). Abundances of D. pulex were expressed as the number of individuals per litre based on the volume of the composite samples collected in the enclosures. D. pulex size structure was obtained by sorting body length data into 5 size-classes according to Gurney et al. (1990) and McCauley et al. (1990): class 1, (<0.6 mm), for neonates; class 2, (0.6–1 mm), for daphnids between 1 and 2 d old; class 3, (1–1.4 mm) for daphnids between 2 and 4 d old; class 4, (1.4–1.7 mm), for daphnids between 4 and 6 d old; class 5, (>1.7 mm), for daphnids older than 6 d, corresponding to the mean age at first reproduction. Invertebrates collected while D. pulex were being sampled were also identified and counted, using the taxonomic key proposed by Amoros (1984) for Cladocera and the taxonomic key proposed by Tachet et al. (2000) for macroinvertebrates. 2.6. Data analysis One-way Repeated Measures-ANOVA (RM-ANOVA) followed by Duncan’s post hoc test were used to compare the data obtained from physicochemical, chlorophyll a and SM measurements, and for the abundance of D. pulex and other invertebrates. Abundance data were log-transformed before analysis. Statistical comparisons were performed over the whole study period to determine overall trends among treatments. When RM-ANOVA indicated a significant effect of treatments on the different parameters, data were further analyzed for each sampling date using one-way ANOVA followed by Duncan’s post hoc test to detect when differences between treated and control microcosms occurred. For body length, Kolmogorov-Smirnov two-sample test was performed to compare central tendency of Daphnia size as well as the general shapes of distributions. This test was performed when the abundance of daphnids per treatment was higher than 10 individuals. All RM- and one-way ANOVA and Kolmogorov-Smirnov two-sample test were performed using Statistica® for Windows Version 6.0 (Statsoft, Tulsa, OK, USA). Differences were considered significant at a = 0.05 for all tests. 3. Results 3.1. Physicochemical and biological parameters The mean values (±SE) of the parameters monitored over the whole study period in the different types of microcosms are given in Table 1. Water depth slightly decreased (from 38.2 cm at d 0 to 24.5 cm at d 21) over the course of the experiment, whereas temperature fluctuated between 13 and 19 °C, salinity fluctuated between 0.3 and 0.8 g L¡1, and pH ranged from 6.8 to 7.5. Greater variations were observed for dissolved oxygen (1.5–11.30 mg L¡1), Table 1 Mean values (±SE; n = 30) and results of the Duncan’s post hoc test for the physicochemical and biological parameters measured in the enclosures during the whole exposure period Parameter Water depth (cm) Temperature (°C) Salinity (g L¡1) Dissolved O2 (mg L¡1) pH Chlorophyll a (lg L¡1) SM (mg L¡1) Chaoborus sp. (nb indiv L¡1) Cladocera (nb indiv L¡1) Culex sp. (nb indiv L¡1) Control microcosms 32 (±1.0) 16.6 (±0.9) 0.6 (±0.1) 4.51 (±0.92) 7.1 (±0.1) 64 (±4) 37 (±4) 17 (±6.7) 50 (±21.7) 1 (±6) Microcosms treated with: Bti Spinosad 0.16 (lL L¡1) 0.50 (lL L¡1) 8 (lg L¡1) 17 (lg L¡1) 33 (lg L¡1) 34 (±1.0) 16.6 (±0.9) 0.6 (±0.1) 4.68 (±0.93) 7.1 (±0.1) 54 (±10) 28 (±6) 13 (±3.6) 69 (±39.0) 0 34 (±1.0) 16.6 (±0.8) 0.6 (±0.1) 4.87 (±1.13) 7.1 (±0.1) 62 (±5) 34 (±8) 29 (±7.3) 49 (±20.5) 3 (±1.2) ** 33 (±1.0) 16.5 (±0.9) 0.6 (±0.1) 5.71 (±1.10)* 7.2 (±0.1) 73 (±7) 33 (±6) 8 (± 4.0) 10 (± 6.5) 1 (± 1.0) 34 (±1.0) 16.5 (±0.9) 0.6 (±0.1) 6.40 (±1.30)** 7.2 (±0.1)* 85 (±12) 37 (±9) 3 (± 1.3) 6 (± 3.2) 0 33 (±0.9) 16.6 (±0.9) 0.6 (±0.1) 5.87 (±1.31)* 7.2 (±0.1)* 74 (±5) 35 (±6) 3 (±1.5) 3 (±1.4) 1 (±0.5) Significant difference as compared to control (Duncan’s post hoc test): *, p < 0.05; **, p < 0.01. C. Duchet et al. / Chemosphere 74 (2008) 70–77 100 Spinosyn concentration (µg L-1) Chlorophyll a concentration (µg L-1) Spinosad, 8 µg L-1 Spinosad, 17 µg L-1 Spinosad, 33 µg L-1 Control Bti, 0.16 µg L-1 Bti, 0.50 µg L-1 160 140 120 100 80 60 40 73 Full symbols: Spinosyn A, for spinosad nominal concentrations of -1 ( ) 8, ( ) 17 and ( ) 33 µg L Open symbols: Spinosyn D, for spinosad nominal concentrations of -1 ( ) 8, ( ) 17 and ( ) 33 µg L 10 1 0.1 20 0 2 4 6 8 Time (days) 0 3 6 9 12 15 18 21 Time (days) Fig. 1. Changes in mean values (±SE; n = 5) of chlorophyll a concentration in the water of the control microcosms, the microcosms treated with Bti at 0.16 and 0.50 lL L¡1, and the microcosms treated with spinosad at 8, 17 and 33 lg L¡1. SM concentration (10–80 mg L¡1), and chlorophyll a concentration (16–130 lg L¡1). Concentrations of chlorophyll a increased in all the microcosms during the first 14 d and decreased from 14 to 21 d after treatments (Fig. 1). Larvae of the phantom midge Chaoborus sp. (Diptera: Nematocera), larvae of the mosquito Culex sp. (Diptera: Nematocera), and Chydorus sp., Pleuroxus sp., Ceriodaphnia sp., Simocephalus sp., and Scapholeberis sp. (Crustacea: Cladocera) grouped as Cladocera in Table 1, were collected. Considering the 30 microcosms, the general tendency is an increase of the abundance of these three taxa over the course of the experiment (from 1 at d 0 to 42 indiv L¡1 at d 21 for Chaoborus sp. larvae, from 7 at d 0 to 153 indiv L¡1 at d 21 for Cladocera, and from 0 at d 0 to 1 indiv L¡1 at d 21 for Culex sp. larvae). However Chaoborus sp. larvae and cladocera were less abundant in spinosad-treated microcosms, as compared to Bti-treated and control enclosures (Table 1). RM-ANOVA did not indicate significant differences in the environmental parameters between treated and control microcosms from the day before the applications (d 0) to the end of the study (d 21), except for pH (p = 0.011), dissolved oxygen (p = 0.004), and Culex sp. abundance (p = 0.02). Water pH values were significantly different between the controls and the enclosures treated with 17 and 33 lg L¡1 spinosad (Duncan’s post hoc test, p = 0.01 and p = 0.013, respectively). Dissolved oxygen concentration was significantly different between the controls and the enclosures treated with 8, 17 and 33 lg L¡1 spinosad (Duncan’s post hoc test, p = 0.034, p = 0.002 and p = 0.019, respectively). Density of Culex sp. larvae was significantly different between the controls and the enclosures treated with 0.16 lL L¡1 Bti (Duncan’s post-hoc test, p = 0.008). When Duncan’s post hoc test showed significant differences between the controls and spinosad-sprayed microcosms, mean pH values, mean dissolved oxygen values and mean Culex sp. larvae density values were always higher in the treated microcosms than in the controls. Fig. 2. Changes in spinosyn A (full symbols) and spinosyn D (open symbols) mean concentrations (±SE; n = 3) measured in microcosm water for spinosad nominal concentrations of 8, 17 and 33 lg L¡1. Day 0 corresponds to sampling immediately (within the first 15 min) after spinosad application. A logarithmic scale has been used for spinosyn concentration (Y-axis). The horizontal dotted line represents the limit of quantification. 90.1 ± 2.1% of the total spinosad level, whatever the initial water concentration. Two days after treatment, spinosad concentrations dramatically decreased to 45.8%, 24.1% and 12.9% of the initial exposure levels, for the nominal concentrations of 8, 17 and 33 lg L¡1, respectively. Spinosyn A concentration still represented on average 88.5 ± 2.7% of the insecticide initial concentration. Within the first two days of exposure, the spinosyn A/spinosyn D ratio did not vary (12.2 ± 0.6, 6.3 ± 0.5 and 9.0 ± 1.3, for the nominal concentrations of 8, 17 and 33 lg L¡1, respectively). From four days on, residue concentrations were at the limit of quantification for the 8 and 17 lg L¡1 treatments, representing 11.8% and 3.9% of the initial exposure level, respectively. For the 33 lg L¡1 nominal concentration, the concentration of spinosad measured after two days was 12.7 ± 0.3% of the initial exposure level. At d 7, residue levels were 1000 Abundance (number of individuals per litre 0 800 600 400 * * 200 *** 0 0 3.2. Spinosad exposure concentrations Analysis of spinosad residues in water showed that the exposure concentrations were in the range of 45–118% of the nominal concentrations at the time of application (Fig. 2). Within the 15 min following the treatment, spinosyn A concentration represented Spinosad, 8 µg L-1 Spinosad, 17 µg L-1 Spinosad, 33 µg L-1 Control Bti, 0.16 µg L-1 Bti, 0.50 µg L-1 *** 3 * *** 6 9 12 *** 15 18 21 Time (days) Fig. 3. Changes in mean values (±SE; n = 5) of Daphnia pulex abundance (expressed as the number of individuals per litre) in the control microcosms, the microcosms treated with Bti at 0.16 and 0.50 lL L¡1, and the microcosms treated with spinosad at 8, 17 and 33 lg L¡1. Significant difference from control (Duncan’s post hoc test): *: p < 0.05; **: p < 0.01; ***: p < 0.001. C. Duchet et al. / Chemosphere 74 (2008) 70–77 74 Table 2 Daphnia pulex mean body size (±SE; n = 5) (mm) at each sampling date after larvicide application, and results of Kolmogorov–Smirnov two-sample test Treatment Sampling date Control Bti, 0.16 lL L ¡1 Bti, 0.50 lL L¡1 Spinosad, 8 lg L¡1 Spinosad, 17 lg L¡1 Spinosad, 33 lg L¡1 Day 0 Day 2 Day 4 Day 7 Day 14 Day 21 1.050 (±0.019) 1.022 (±0.014) 1.018 (±0.015) 1.040 (±0.016) 1.042 (±0.019) 1.048 (±0.016) 1.173 (±0.019) 1.178 (±0.020) 1.155 (±0.022) 1.092 (±0.034)** 1.042 (±0.086) 0.856 (±0.083)** 1.268 (±0.022) 1.309 (±0.029) 1.326 (±0.027) 1.388 (±0.034)* 1.297 (±0.063) 1.176 (±0.086) nt 1.006 (±0.031) 1.015 (±0.031) 0.992 (±0.037) 1.115 (±0.046)*** 0.930 (±0.044)* 0.906 (±0.225) nt 1.064 (±0.037) 1.013 (±0.024) 0.973 (±0.033) 1.309 (±0.028)*** 1.038 (±0.134) 0.739 (±0.064) nt 1.037 (±0.032) 1.430 (±0.042)*** 1.303 (±0.032)*** 1.267 (±0.039)*** 1.334 (±0.057)*** 0.994 nt Significant difference as compared to control (Kolmogorov-Smirnov two-sample test):*: p < 0.05; **: p < 0.01; ***: p < 0.001. nt: not tested (number of daphnids 6 10). 8.4%, 6.5% and 4.4% of the initial exposure level for the nominal concentrations of 8, 17 and 33 lg L¡1, respectively. Spinosyn A represented on average 79.1 ± 7.3%, and the spinosyn A/spinosyn D ratio was 4.5 ± 1.5 whatever the insecticide initial concentration. 3.3. Daphnia pulex The mean abundance of D. pulex varied between 730 and 280 individuals per litre in Bti-treated and control microcosms from the beginning to the end of the whole study period (Fig. 3). In the microcosms treated with 8 lg L¡1 spinosad, D. pulex population density ranged between 780 and 180 individuals per litre whereas there was a sharp density decrease in the enclosures treated with 17 and 33 lg L¡1 spinosad (respectively from 501 to 15 individuals per litre and from 479 to 3 individuals per litre). RM-ANOVA showed significant differences between the treatments (p < 0.0001). Duncan’s post hoc test did not show any significant difference between Bti-treated and untreated microcosms. In contrast, spinosad at 17 and 33 lg L¡1 caused a significant reduction in the density of D. pulex as compared to the controls from d 2 after treatment until the end of the study (Duncan’s post hoc test, p < 0.0001 for both treatments). Density of D. pulex also decreased in the microcosms treated with 8 lg L¡1 spinosad (Fig. 3), but differences were only significant at days 2 and 4 after treatment as compared to the controls (Duncan’s post hoc test, p = 0.046 and p = 0.035, respectively), indicating that population recovery occurred in those enclosures. The effect of treatment on mean size of D. pulex is shown in Table 2. In the controls, the first cohort grew between d 0 and d 4 (mean size ranging from 1.050 to 1.268 mm). At days 7, 14 and 21, a new cohort was identified in the samples (size-class 2; Fig. 4). In Bti-treated microcosms, similar results were observed, except at d 2 in the microcosms treated with Bti at 0.50 lL L¡1 where size-class 3 was more abundant, and at d 21 for both concentrations where the younger daphnids did not represent the major class (Fig. 4). In microcosms treated with 8 lg L¡1 spinosad, Kolmogorov-Smirnov two-sample test showed significant differences in the size structure of D. pulex population as compared to the controls from d 2 to d 21 (Fig. 4). In the microcosms treated with 17 lg L¡1, Kolmogorov-Smirnov two-sample test showed significant differences in the size structure of D. pulex population at d 7 and d 21 as compared to the control. In the microcosms treated with 33 lg L¡1 spinosad, Kolmogorov-Smirnov two-sample test was carried out only at d 0 and d 2 because there were fewer than 10 daphnids per litre for each of the other sampling dates. Furthermore, the new cohort that appeared in the control enclosures at d 7, 14 and 21 was not observed in any of the spinosad-treated microcosms. 4. Discussion For the first time, the present study demonstrates effects of spinosad on a natural population of D. pulex. Both survival and size structure were affected by the treatments, but the lowest concentration (8 lg L¡1) allowed population recovery after the first week of exposure. In contrast, Bti had no effect on D. pulex population survival, confirming a number of previous studies (reviewed in Boisvert and Boisvert, 2000 and Boisvert and Lacoursière, 2004). Effects of spinosad were associated with a short time window of exposure (less than the first two days), as both spinosyns A and D were nearly undetectable in the water after four to seven days. Overall, physicochemical water conditions (e.g., pH, salinity, SM) were not affected by the treatments. Water temperatures differed between d 0 and d 21, and increased inside the microcosms as air temperature increased during the study. Water level decreased and salinity concomitantly increased during the study due to evaporation. Concentrations of dissolved oxygen in the microcosms treated with spinosad were significantly higher than in the controls from d 2 to d 21 (Table 1), which may reflect depletion of the population of D. pulex and possibly of other invertebrates. Other water quality parameters (e.g., chlorophyll a concentration) were insensitive to the effects of the insecticide (Fig. 1). However, chlorophyll a level is a rough indicator of microalgal biomass (Dauta and Feuillade, 1995), and further investigations are needed to assess the possible consequences of changes in arthropod densities on phytoplankton. Spinosad concentrations measured in microcosm water were close to the expected concentrations shortly after treatment (Fig. 2), the slight discrepancy observed for the lowest concentration being likely due to the sampling procedure and/or a rapid degradation of spinosyns as both the A and D forms phototransform rapidly in aqueous solutions. Dissipation kinetics of spinosad in our microcosms are in close agreement with the results obtained by Cleveland et al. (2002a) in outdoor microcosms of comparable size, and confirm the half-life of 1–2 d for the sum of spinosyns A and D (Saunders and Bret, 1997; Cleveland et al., 2002a). Within the first two days, the ratio between spinosyns A and D concentrations remained constant, indicating that both compounds degraded at similar rates, as shown by Cleveland et al. (2002a,b) who determined half-lives of 1.6–1.8 d in outdoor microcosms. In the present study, a mean half-life value of 0.76 ± 0.02 d was estimated using exponential regression for the three nominal spinosad concentrations tested. Rapid removal of spinosad from the water column results from both photolysis and sediment partitioning (Saunders and Bret, 1997; Cleveland et al., 2002a). Both processes were favoured in our experimental conditions since the microcosms C. Duchet et al. / Chemosphere 74 (2008) 70–77 240 240 A. Control 160 120 80 40 160 120 ** 80 * *** *** *** 40 0 0 0 2 4 240 7 14 21 0 2 240 B. Bti - 0.16 µL L-1 4 7 14 21 E. Spinosad - 17 µg L-1 200 Abundance 200 Abundance D. Spinosad - 8 µg L-1 200 Abundance Abundance 200 160 120 80 *** 40 160 120 * 80 *** 40 0 0 2 4 240 14 21 0 2 240 C. Bti - 0.50 µL L-1 ** 200 7 4 7 14 21 F. Spinosad - 33 µg L-1 200 160 120 80 *** 40 Abundance 0 Abundance 75 160 120 80 40 0 ** nt nt nt nt 2 4 7 14 21 0 0 2 4 7 14 21 Time (days) 0 Time (days) class 1 (< 0.6 mm; neonates); class 2 (0.6 - 1 mm; 1 to 2 days old); class 3 (1 - 1.4 mm; 2 to 4 days old); class 4 (1.4 - 1.7 mm; 4 to 6 days old); class 5 (> 1.7 mm; older than 6 days) Fig. 4. Changes in size-structure of the Daphnia pulex population exposed to 0.16 and 0.50 lL L¡1 Bti (B and C, respectively) and 8, 17 and 33 lg L¡1 spinosad (D, E and F, respectively) as compared to control (A). Significant difference from control (Kolmogorov-Smirnov test): *: p < 0.05; **: p < 0.01; ***: p < 0.001. nt: not tested (number of daphnids 6 10). were exposed to June sunlight and contained a 5–10 cm layer of organic matter rich sediments. Effects of spinosad on D. pulex were compared to those of Bti, applied at 0.16 and 0.50 lL L¡1. As expected, Bti had no effect on the survival of D. pulex, as shown by the absence of change in abundance (Fig. 3). According to previous studies, Bti is highly specific to Nematocera (Diptera) like Culicidae, Simuliidae and Chironomidae (Boisvert and Boisvert, 2000). Other aquatic invertebrates (molluscs, crustaceans, and insects) and vertebrates (fish and amphibians) are not sensitive to Bti. Moreover, Ali (1981) and Miura et al. (1981) showed that Ephemeroptera, Amphipoda, Cladocera and Copepoda were not affected by Bti. However, our study shows that there were fewer younger daphnids and more older daphnids at d 21 in the enclosures treated with Bti (0.16 and 0.50 lL L¡1), as compared to the control (Fig. 4). As this change was only observed at d 21 after Bti application, direct effects are unlikely. Size selection by predation and/or competition in the D. pulex population might be an explanation. Chaoborus sp. larvae, which were abundant in the enclosures treated with Bti (Table 1), may have impacted some specific size-classes in the D. pulex population. Indeed, phantom midge larvae are major predators in zooplankton (Lüning-Krizan, 1997), and medium-sized preys represent a great proportion of the diet because of their vulnerability (defined as the product of the encounter rate between predator and prey and the capture success of the predator; Pastorok, 1981). In addition, there were more Cladocera (Chydorus sp., Pleuroxus sp., Ceriodaphnia sp., Simocephalus sp., and Scapholeberis sp.) in enclosures treated with 0.16 lL L¡1 Bti, and more Culex sp. larvae in enclosures treated with 0.50 lL L¡1 Bti, as compared to the controls (Table 1). Cladocera and mosquito larvae are filter-feeders and may have competed with D. pulex for food, resulting in a decrease in the number of young daphnids which are more vulnerable to food deprivation. 76 C. Duchet et al. / Chemosphere 74 (2008) 70–77 In contrast to Bti, spinosad had a significant effect on the natural population of D. pulex. At 17 and 33 lg L¡1, spinosad affected the whole population and did not impact a particular size-class (Figs. 3 and 4). This indicates that all size-classes of D. pulex were equally sensitive to the highest concentrations applied to the microcosms. In the enclosures treated with 8 lg L¡1 spinosad, D. pulex density measured within the first four days of exposure was significantly lower than controls (Fig. 3). High mortality rates due to the exposure to 17 and 33 lg L¡1 nominal concentrations were also observed, which are inconsistent with the laboratory data published by Stark (2005), who estimated the acute LC50 at 129 lg L¡1. Recovery of the D. pulex population in microcosms treated with spinosad at 17 and 33 lg L¡1 did not occur, even after pesticide dissipation (Fig. 3). This may be due to the short duration of the observation period and/or to an eutrophication process (i.e., increase of algal biomass) resulting in deoxygenation (Lampert and Sommer, 1997). Moreover, in our microcosms, concentrations of dissolved oxygen decreased drastically (mean value at d 2 in all the enclosures: 5.47 mg L¡1 versus 2.58 mg L¡1 in spinosad-treated microcosms; 1.52 mg L¡1 in Bti-treated microcosms; 1.82 mg L¡1 in the controls at d 21) resulting in degradation of water quality, which impeded recovery of the D. pulex population affected by the two highest spinosad concentrations. Chronic NOECs of spinosad were estimated at 6.7 and 8 lg L¡1 in static and semi-static laboratory tests, respectively (National Registration Authority for Agricultural and Veterinary Chemicals, 1998; WHO, 2007). In our conditions, 8 lg L¡1 spinosad produced significant mortality in D. pulex over the first four days of exposure. Seven days after the treatment, the D. pulex population recovered to densities similar to those measured in the control and Bti-treated microcosms. At d 4, the most abundant size-class in the microcosms treated with 8 lg L¡1 spinosad was class 4 (between 1.4 and 1.7 mm) which was able to reproduce two days later, leading to an increase in the abundance of neonates (class 1, <0.6 mm) and juveniles (class 2, 0.6–1 mm) at d 7 (Fig. 4). In contrast, recovery was not observed in laboratory experiments where the exposure of D. pulex to 10 lg L¡1 spinosad led to a decline in survival, birth rate (b), net reproductive rate (R0), and intrinsic rate of increase (rm), population extinction (negative rm values) occurring after eight days of exposure to spinosad (Stark and Vargas, 2003). This discrepancy may be due to the fact that, in the laboratory study, D. pulex were maintained in static renewal tests and moved to newly made pesticide solutions every other day. Additionally, spinosad may have been more persistent in the laboratory conditions as little photodegradation would have occurred. 5. Conclusion Unlike Bti, spinosad had a strong lethal effect on D. pulex when applied at 17 and 33 lg L¡1, resulting in population extinction. This result does not confirm the LOEC that was estimated at 52 lg L¡1 in static laboratory tests (National Registration Authority for Agricultural and Veterinary Chemicals, 1998). When applied at 8 lg L¡1 in the enclosures, spinosad slowed down the growth of the population but recovery was observed after seven days. This is not in agreement with laboratory studies which predicted population extinction within eight days of exposure to 10 lg L¡1 spinosad (Stark and Vargas, 2003). In situ microcosms therefore appear to be a good compromise between laboratory tests and open field studies, allowing a realistic assessment of larvicide impact, including recovery capacity, on natural invertebrate populations. Acknowledgements The authors are grateful to the French Ministry for Ecology, Sustainable Development and Spatial Planning for financial support through the National Programme for Ecotoxicology (PNETOX). The help of M. Roucaute and P. Le Goff for field sampling was greatly appreciated. The authors are most grateful to Dow AgroSciences for the generous gift of spinosad 120SC and for helpful information on analytical procedures. They also gratefully acknowledge Mr. Jacques Defois for giving them access to the study site. References ACTA, 2007. Index Phytosanitaire. Association de Coordination Technique Agricole, Paris. AFNOR, 1996. Qualité de l’eau – Dosage des matières en suspension – Méthode par filtration sur filtre en fibres de verre. In: Association Française de Normalisation (Eds.), Qualité de l’eau – 6e édition Tome 2 – Analyses organoleptiques – Mesures physico-chimiques – Paramètres globaux – Composés organiques. AFNOR, Paris, France, pp. 101–110. Ali, A., 1980. Nuisant Chironomids and their control: a review. Entomol. Soc. Am. Bull. 26, 3–16. Ali, A., 1981. Bacillus thuringiensis serovar israelensis (ABG-6108) against chironomids and some nontarget aquatic invertebrates. J. Invertebr. Pathol. 38, 264– 272. Amoros, C., 1984. Crustacés Cladocères. Bull. mensuel de la Soc. Linn. de Lyon 53, 72–145. Boisvert, J., Lacoursière, J.O., 2004. Le Bacillus thuringiensis et le contrôle des insectes piqueurs au Québec. ENV/2004/0278. Ministère de l’Environnement Québécois, Québec Canada. Boisvert, M., Boisvert, J., 2000. Effects of Bacillus thuringiensis var. israelensis on target and non target organisms: a review of laboratory and field experiments. Biocontrol Sci. Technol. 10, 517–561. Boronat, M.D., Miracle, M.R., 1997. Size distribution of Daphnia longispina in the vertical profile. Hydrobiologia 360, 187–196. Burnett, K.M., Liss, W.J., 1990. Multi-steady-state toxicant fate and effect in laboratory aquatic ecosystems. Environ. Toxicol. Chem. 9, 637–647. Caquet, Th., Hanson, M.L., Roucaute, M., Graham, W.G., Lagadic, L., 2007. Influence of isolation on the recovery of pond mesocosms from the application of an insecticide. II. Benthic macroinvertebrate responses. Environ. Toxicol. Chem. 26, 1280–1290. Cetin, H., Yanikoglu, A., Cilek, J.E., 2005. Evaluation of the naturally-derived insecticide spinosad against Culex pipiens L. (Diptera: Culicidae) larvae in septic tank water in Antalya, Turkey. J. Vector Ecol. 30, 151–154. Cleveland, C.B., Bormett, G.A., Saunders, D.G., Powers, F.L., McGibbon, A.S., Reeves, G.L., Rutherford, L., Balcer, J.L., 2002a. Environmental fate of spinosad. 1. Dissipation and degradation in aqueous systems. J. Agric. Food Chem. 50, 3244– 3256. Cleveland, C.B., Mayes, M.A., Cryer, S.A., 2002b. An ecological risk assessment for spinosad use on cotton. Pest Manage. Sci. 58, 70–84. Consoli, F.L., Botelho, P.S.M., Parra, P.J.R., 2001. Selectivity of insecticides to the egg parasitoid Trichogramma galloi Zucchi, 1988 (Hym., Trichogrammatidae). J. Appl. Entomol. 125, 37–43. Crouse, G.D., Sparks, T.C., Schoonover, J., Gifford, J., Dripps, J., Bruce, T., Larson, L., Garlich, J., Hatton, C., Hill, R.L., Worden, T.V., Martynow, J.G., 2001. Recent advances in the chemistry of spinosyns. Pest Manage. Sci. 57, 177–185. Dauta, A., Feuillade, J., 1995. Croissance et dynamique des populations algales. In: Pourriot, R., Meybeck, M. (Eds.), Limnologie générale. Masson, Paris France, pp. 328–350. DowElanco, 1996. Spinosad technical guide. Form No 200-03-001 (4/96). Indianapolis, USA. Garcia, R., Des Rochers, B., Tozer, W., 1981. Studies on the toxicity of Bacillus thuringiensis var. israelensis against organisms found in association with mosquito larvae. In: California Mosquito and Vector Control Association, Inc (Eds.), Proceeding of the Forty-eighth Annual Conference of the California Mosquito and Vector Control Association, Inc, pp. 33–36. Gurney, W.S.C., McCauley, E., Nisbet, R.M., Murdoch, W.W., 1990. The physiological ecology of Daphnia: a dynamic model of growth and reproduction. Ecology 71, 716–732. He, L.S., Ong, K.H., 2000. Effects of Bacillus thuringiensis H-14 on bloodworms (Diptera: Chironomidae). Singapore. J. Pri. Ind. 28, 7–12. Hershey, A.E., Lima, A.R., Niemi, G.J., Regal, R.R., 1998. Effects of Bacillus thuringiensis israelensis (Bti) and methoprene on nontarget macroinvertebrates in Minnesota wetlands. Ecol. Appl. 8, 41–60. Holt, K.M., Opit, G.P., Nechols, J.R., Margolies, D.C., 2006. Testing for non-target effects of spinosad on two spotted spider mites and their predator Phytoseiulus persimilis under greenhouse conditions. Exp. Appl. Acarol. 38, 141–149. Isman, M.B., 2006. Botanical insecticides deterrents, and repellents, in modern agriculture and an increasingly regulated world. Annu. Rev. Entomol. 51, 45–66. Lampert, W., Sommer, U., 1997. Limnology: The Ecology of Lakes and Streams. Oxford University Press, Oxford, UK. Lorenzen, C.J., 1967. Determination of chlorophyll and pheopigments: Spectrophotometric equations. Limnol. Oceanogr. 12, 343–346. Lüning-Krizan, J., 1997. Selective feeding of third- and fourth-instar larvae of Chaoborus flavicans in the field. Arch. Hydrobiol. 140, 347–365. McCauley, E., Murdoch, W.W., Nisbet, R.M., Gurney, W.S.C., 1990. The physiological ecology of Daphnia: development of a model of growth and reproduction. Ecology 71, 703–715. C. Duchet et al. / Chemosphere 74 (2008) 70–77 Medina, M., Barata, C., Telfer, T., Baird, D.J., 2004. Effects of cypermethrin on marine plankton communities: a simulated field study using mesocosms. Ecotoxicol. Environ. Saf. 58, 236–245. Merritt, R.W., Walker, E.D., Wilzbach, M.A., Cummins, K.W., Morgan, W.T., 1989. A broad evaluation of Bti for black fly (Diptera: Simuliidae) control in a Michigan river: efficacy, carry and non target effects on invertebrates and fish. J. Am. Mosquito Control Assoc. 5, 397–415. Miles, M., 2003. The effects of spinosad, a naturally derived insect control agent to the honeybee. Bull. Insectol. 56, 119–124. Miles, M., Dutton, R., 2000. Spinosad: a naturally derived insect control agent with potential use in glasshouse integrated pest management systems. Meded. Fac. Landbouwkundige en Toegepaste Biol. Wet. Univ. Gent 65, 393–400. Miura, T., Takahashi, R.M., Mulligan III, F.S., 1981. Impact of the use of candidate bacterial mosquito larvicides on some selected aquatic organisms. In: California Mosquito and Vector Control Association, Inc (Eds.), Proceedings of the Annual Conference of the Californian Mosquito Control Association 49, pp. 45–48. Nasreen, A., Ashfaq, M., Mustafa, G., 2000. Intrinsic toxicity of some insecticides to egg parasitoid Trichogramma chilonis (Hym. Trichogrammatidae). Bull. Inst. Trop. Agric. Kyushu Univ. 23, 41–44. National Registration Authority for Agricultural and Veterinary Chemicals, 1998. Spinosad in the products Laser Naturalyte Insect Control–Tracer Naturalyte Insect Control. NRA, Camberra, Australia. Pastorok, R.A., 1981. Prey vulnerability and size selection by Chaoborus larvae. Ecology 62, 1311–1324. Pineda, S., Smagghe, G., Schneider, M.I., Del Estal, P., Viñuela, E., Martìnez, A.M., Budia, F., 2006. Toxicity and pharmacokinetics of spinosad and methoxyfenozide to Spodoptera littoralis (Lepidoptera: Noctuidae). Environ. Entomol. 35, 856–864. Purcell, B.H., 1981. Effects of Bacillus thuringiensis var. israelensis on Aedes taeniorhynchus and some non target organisms in the salt marsh. Mosq. News 41, 476–484. Rand, G.M., Clark, J.R., 2000. Hazard/risk assessment of pyridaben: I. Aquatic toxicity and environmental chemistry. Ecotoxicology 9, 157–168. 77 Salgado, V.L., 1998. Studies on the mode of action of spinosad: insect symptoms and physiological correlates. Pestic. Biochem. Physiol. 60, 91–102. Saunders, D.G., Bret, B.L., 1997. Fate of spinosad in the environment. Down to Earth 52, 14–20. SETAC, 1991. Guidance Document on Testing Procedures for Pesticides in Freshwater Static Mesocosms. In: SETAC (Eds.), A Meeting of Experts on Guideline for Static Field Mesocosms Tests, Huntingdon, UK. Stark, J.D., 2005. How closely do acute lethal concentration estimates predict effects of toxicants on populations? Int. Environ. Assess. Manage. 1, 109–113. Stark, J.D., Vargas, R.I., 2003. Demographic changes in Daphnia pulex (Leydig) after exposure to the insecticides spinosad and diazinon. Ecotoxicol. Environ. Saf. 56, 334–338. Stephenson, G.L., Hamilton, P., Kaushik, N.K., Robinson, J.B., Solomon, K.R., 1984. Spatial distribution of plankton in enclosures of three sizes. Can. J. Fish. Aquat. Sci. 41, 1048–1054. Tachet, H., Richoux, P., Bournaud, M., Usseglio-Polatera, P., 2000. Invertébrés d’eau douce - Systématique, biologique, écologie. CNRS Editions, Paris France. Tillman, P.G., Mulrooney, J.E., 2000. Effects of selected insecticides on the natural enemies Coleomegilla maculata and Hippodamia convergens (Coleoptera: Coccinellidae), Geocoris punctipes (Hemiptera: Lygaeidae), and Bracon mellitor, Cardiochiles nigriceps, and Cotesia marginiventris (Hymenoptera: Braconidae) in cotton. J. Econ. Entomol. 93, 1638–1643. van den Brink, P.J., Tarazona, J.V., Solomon, K.R., Knacker, T., van den Brink, N.W., Brock, T.C.M., Hoogland, J.P.H., 2005. The use of terrestrial and aquatic microcosms and mesocosms for the ecological risk assessment of veterinary medicinal products. Environ. Toxicol. Chem. 24, 820–829. van Frankenhuyzen, K., 1993. The challenge of Bacillus thuringiensis. In: Entwistle, P.F.J., Cory, S., Bailey, M.J., Higgs, S. (Eds.), Bacillus thuringiensis, an Environmental Biopesticide: Theory and Practice. John Wiley & Sons, pp. 1–35. Watson, G.B., 2001. Actions of insecticidal spinosyns on gamma-aminobutyric acid receptors from small-diameters cockroach neurones. Pestic. Biochem. Physiol. 71, 20–28. WHO, 2007. Spinosad. World Health Organization, Geneva, Switzerland.