Interaction of chemical cues from fish tissues and organophosphorous pesticides on Ceriodaphnia dubia survival

Environmental Pollution 141 (2006) 90e97 www.elsevier.com/locate/envpol

Interaction of chemical cues from fish tissues and organophosphorous pesticides on Ceriodaphnia dubia survival Jonathan D. Maul a,b,*, Jerry L. Farris a, Michael J. Lydy b a

Ecotoxicology Research Facility, Environmental Sciences Program, PO Box 847, Arkansas State University, State University, AR 72467, USA b Fisheries and Illinois Aquaculture Center and Department of Zoology, Southern Illinois University, Carbondale, IL 62901, USA Received 17 November 2004; accepted 5 August 2005

Potentiation of organophosphorous pesticide toxicity to Ceriodaphnia dubia by fathead minnow (Pimephales promelas) chemical cues was observed. Abstract Cladocera are frequently used as test organisms for assessing chemical and effluent toxicity and have been shown to respond to stimuli and cues from potential predators. In this study, the interactive effects of visual and chemical cues of fish and two organophosphorous pesticides on survival of Ceriodaphnia dubia were examined. A significant chemical cue (homogenized Pimephales promelas) and malathion interaction was observed on C. dubia survival (P Z 0.006). Chemical cue and 2.82 mg/L malathion resulted in a 76.0% reduction in survival compared to malathion alone (P ! 0.01). Furthermore, potentiation of malathion toxicity varied based on the source of chemical cues (i.e., epithelial or whole body). It is unclear in this study whether these chemical cues elicited a predation-related stress in C. dubia. Future research should examine the mechanism of this interaction and determine what role, if any, stress responses by C. dubia might play in the interaction. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Ceriodaphnia dubia; Dicrotophos; Malathion; Potentiation; Kairomone

1. Introduction Impairment to aquatic systems by agricultural chemicals is often assessed using single species and single toxicant laboratory studies (Cairns, 1983; Cooney, 2003). These studies are typically conducted with little or no environmental variability or population-mediating factors such as predation, competition, and resource limitation. The utility of these studies for protection of aquatic resources can be questioned when considering that pesticide exposures occur simultaneously with other environmental variables and may potentially result in complex interactive effects on non-target organisms (Cairns, 1986). * Corresponding author. Fisheries and Illinois Aquaculture Center and Department of Zoology, Southern Illinois University, Carbondale, IL 62901, USA. Tel.: C1 618 453 6080; fax: C1 618 453 6095. E-mail addresses: [email protected], [email protected] (J.D. Maul). 0269-7491/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2005.08.013

Chemical stimuli or cues from potential predators are one important environmental variable that can influence organism response to pesticides (Hanazato and Dodson, 1992; Relyea, 2004; Relyea and Mills, 2001). In amphibians, carbamate and organophosphorous (OP) pesticide toxicity was greater when test organisms were exposed simultaneously to the pesticide and a chemical predator cue (Relyea, 2004; Relyea and Mills, 2001), and a significant interaction between the presence of a predator and a carbamate was reported for a sublethal response (mass) (Boone and Semlitsch, 2001). Similar to amphibians, cladoceran population growth and survival is strongly dependent upon ecological processes such as predation (Lynch, 1979) with perturbations often causing predictable patterns (Odum, 1985). Although variation in abiotic environmental factors has been shown to modify cladoceran survival and reproduction responses to contaminant exposure (Folt et al., 1999), interaction effects of contaminants and predator

J.D. Maul et al. / Environmental Pollution 141 (2006) 90e97

cues on cladocerans have received limited attention (Hanazato, 1999; Hanazato and Dodson, 1992; Hanazato and Dodson, 1995). In terms of predator chemical cues, the source (intestinal tract) and type of molecule for a chemical cue from the invertebrate predator Chaoborus americanus has been identified (Parejko and Dodson, 1990). For fish-based chemical cues, cladoceran responses vary among potential fish predators (Hanazato et al., 2001; Weber, 2003), and it has been suggested these cues are composed of a complex mixture of signaling chemicals (Weber, 2003). Examining interactions of chemical cues and pesticides is particularly important for cladocerans because species such as Ceriodaphnia dubia, Daphnia pulex, and D. magna are frequently used to indicate waterbody impairment, are used in whole effluent toxicity (WET) testing, and often function in toxicity testing as a surrogate testing organism for protection of a broad spectrum of aquatic organisms. Thus, further identification of these interactions is critical for providing a morerealistic estimation of risk to aquatic systems (Hanazato, 2001). The objective of this study was to examine the interaction effects of visual and chemical cues from a fish and two organophosphorous (OP) insecticides (dicrotophos and malathion) on C. dubia with the prediction that contaminant effects are enhanced in the presence of the visual and chemical cues. Five experiments were conducted to address this objective, the first two of which focused on interactions between OPs and two distinct stimuli (visual and chemical) from a potential fish predator. The following two experiments focused specifically on interactions between OPs and chemical stimuli with increased replication and reduced design complexity. The final experiment explored organism response to OPs and two different types of chemical stimuli from a fish to gain a better understanding of possible sources of fish chemical cues. 2. Materials and methods 2.1. Chemicals, testing conditions, and doseeresponse experiments Dicrotophos (dimethyl 2-dimethylcarbamoyl-1-methylvinyl phosphate) and malathion (O,O-dimethyl S-(1,2-dicarbethoxyethyl) phosphorodithioate) were selected for this study because they are two of the most frequently and abundantly applied OP compounds for insect control on cotton production areas (USDA, 2003). In Arkansas alone, 1.36 ! 104 kg active ingredient (a.i.) of dicrotophos and 6.46 ! 105 kg a.i. of malathion were applied on cotton fields in 2001 (USDA, 2003), and both compounds have been detected in aquatic habitats (i.e., streams and rivers) of the same region (Thurman et al., 1998). Dicrotophos and malathion used in the study were obtained from ChemService (Westchester, PA, USA) with certified purities of 98.0 and 99.2%, respectively. Stock solutions were prepared by pipetting neat compound into known volumes of pesticide grade acetone. Nominal stock concentrations were 150.78 mg/L of dicrotophos and 25.83 mg/L of malathion. Ceriodaphnia dubia were originally obtained from the US Environmental Protection Agency (Duluth, MN, USA) and reared in cultures at the Ecotoxicology Research Facility at Arkansas State University (State University, AR, USA) following standardized culturing procedures (USEPA, 2002). Cultures were maintained at 25 G 1  C in synthetically prepared, moderately hard water and fed daily a mixed algal species solution containing Selenastrum capricornutum (R80%) and Ankistrodesmus sp. resulting in a concentration

91

of 2.0e2.3 ! 105 cells/mL and 0.1 mL of a yeast/cereal/trout chow (YCT)/ 20 mL. Neonates used for experiments were within 24 G 2 h of age and !24 G 2 h old. Experimental chambers consisted of 33-mL borosilicate glass vials and contained 20 mL of test water and five C. dubia neonates. Static acute toxicity experiments were conducted following standard procedures (USEPA, 2002) with a feeding modification to simulate realistic environmental conditions. Experimental chambers were administered a single feeding of algae and YCT at the start of each test in the amounts described above. All tests were run in environmental chambers with temperature maintained at 25 G 1  C, a 16:8-h light:dark photoperiod, and moderately hard dilution water was used for negative and solvent controls and pesticide exposures. Solvent controls contained pesticide grade acetone at a concentration equivalent to the highest contaminant concentration in a test and never exceeded 300 ml/L. Acute (48-h) doseeresponse experiments were conducted for dicrotophos and malathion (USEPA, 2002). The highest test concentration for each dosee response test was prepared by pipetting a known volume of pesticide stock solution into dilution water. The highest concentration was serially diluted resulting in nominal acute test concentrations of 2.83, 5.65, 11.31, 22.62, and 45.24 mg/L for dicrotophos, and 2.68, 4.46, 7.44, 12.40, and 20.66 mg/L for malathion.

2.2. Malathion and dicrotophos stability experiment Stability of both OPs was examined prior to the definitive predatorcontaminant experiments to understand changes in contaminant exposure concentrations over time under the experimental conditions used in all experiments. Malathion and dicrotophos concentrations were measured after 24 and 48 h of test conditions (including cladoceran food additions) from three samples for each time point and pesticide. Each sample was composed of a composite of the contents of ten 33-mL clear borosilicate experimental chambers, each of which contained 20 mL of test water.

2.3. Visual cues and contaminant interactions The effects of a visual cue of fish, pesticide exposure, and their interaction on 48-h C. dubia survival were evaluated using a split-plot factorial design in two blocks. In the first experiment, the whole plot factor was visual cues at two levels: presence or absence of fathead minnow (Pimephales promelas). Sub-plot factors consisted of dicrotophos at three concentrations (0.0, 18.02, and 19.17 mg/L), and malathion at three concentrations (0.0, 2.69, and 2.82 mg/L). Contaminant concentrations below the LC50 s (i.e., concentration that resulted in 50% mortality) observed in the doseeresponse experiments were selected for all visual cue-contaminant experiments. Exposure concentrations below the LC50 s were targeted because they resulted in measurable pesticide effects that were low enough to assess potentiation or greater than additive effects when the pesticide was presented in combination with an additional factor. P. promelas is a common predator of cladocerans, and in some populations, cladocerans are the predominant prey item consumed (Held and Peterka, 1974). In laboratory cultures, 6e8 month old P. promelas were reared in 75.7 L tanks following a standardized protocol (USEPA, 2002). P. promelas assigned to these experiments regularly foraged on cladocerans by additions of C. dubia, Daphnia pulex, and D. magna into P. promelas mass culture tanks at least twice a week for 4 months prior to the experiments. Two 57.2 ! 40.6 ! 15.2 cm plastic containers were partitioned into quadrants with opaque plexiglass partitions. Containers were filled with dechlorinated tap water from the same source used for daily culturing of P. promelas. Either presence or absence of P. promelas was randomly assigned to each quadrant. Those quadrants assigned presence of fish received six P. promelas. C. dubia were added to each experimental chamber and suspended into quadrants. As such, this method ensured that test organisms were not consumed by the fish, were exposed to the designated contaminant concentration, and foraging behavior of P. promelas was not inhibited by the contaminant. Contaminant treatment combinations were randomly assigned to 15 chambers within each quadrant resulting in six replicates for each treatment combination. Survival at 48 h was selected as an acute response in these experiments

92

J.D. Maul et al. / Environmental Pollution 141 (2006) 90e97

to facilitate comparisons to previously published OP toxicity data for daphnids, and is a commonly used endpoint.

scaled surfaces and fins and the mucus was collected and brought to 50 mL with deionized water. Six replicates were used for every treatment combination.

2.4. Chemical cues and contaminant interactions

2.7. Water quality and protein analysis

A second experiment was conducted to evaluate the effects of chemical cues from fish, pesticides, and their interactions on C. dubia survival using a completely randomized factorial design. The response variable and contaminant concentrations were identical to those described for experiment 1; however, the P. promelas cue was a chemical stimulus. Treatment combinations were randomly assigned to chambers and there were six replicates for every treatment combination. Methods in previous studies for introducing a chemical predator cue include containing caged predators within the test system (Relyea and Mills, 2001), using predator conditioned water (Hanazato et al., 2001), and preparing predator extracts (Hanazato and Ooi, 1992; Hebert and Grewe, 1985; Parejko and Dodson, 1990). In this study, the chemical cue was the presence or absence of a homogenized P. promelas. This technique for preparing a chemical cue was selected to minimize the confounding effects that fish-conditioned water would present to our study. The effects of interest in this study involved organic contaminants, and test water preconditioned with fish could substantially influence targeted pesticide concentrations and water quality. Thus, a solution of homogenized minnow enabled experimental chambers to be spiked with only a small aliquot of potential chemical cues, similar to the method used for invertebrate (Chaoborus flavicans) extracts (Hanazato and Ooi, 1992). In addition, the origin and type of chemical cues for P. promelas are unknown, but if they originate from the digestive organs of the fish, as in the invertebrate Chaoborus americanus (Parejko and Dodson, 1990), this method should facilitate the chemical cues to be homogenously in solution and available. Chambers assigned the presence of fish chemical cues were spiked with a 10 mL aliquot of a 1:8 w/v solution of homogenized adult male (3.95 g) and female (2.07 g) P. promelas in deionized water. Treatment combinations were randomly assigned to borosilicate experimental chambers with six replicates for every treatment combination.

Temperature, dissolved oxygen, pH, and conductivity of test waters were measured prior to exposure and at the termination of the acute doseeresponse and fish cue and contaminant experiments. Temperature of environmental chambers was monitored daily. Additional experimental chambers were included in experiments 3 and 4 for a comparison of protein content in chambers with and without a 10-mL spike of homogenized P. promelas solution. These chambers contained 20 mL of dilution water and the cladoceran food additions. Protein concentration was determined using a modified version of the Lowry method designed for 96-well microplates (BioRad, Richmond, CA, USA). The protein assay was performed using a 96-well microplate spectrophotometer (Tecan Austria GmbH, Gro¨dig/Salzburg, Austria) taking five single absorbance readings every 10 s at l Z 750 nm at 25  C. Bovine gamma globulin was used as a protein standard.

2.5. Chemical cues and contaminant interactions: additional experiments Experiments 3 and 4 focused more specifically on interactions between P. promelas chemical cues and each OP individually using randomized factorial designs with chemical cue at two levels (presence/absence) in both experiments and malathion at four concentrations (0.0, 2.69, 2.82, and 2.96 mg/L) in experiment 3, and dicrotophos at three concentrations (0.0, 15.00, and 17.20 mg/L) in experiment 4. Chambers randomly assigned chemical cues were spiked with a 10 mL aliquot of a 1:8 w/v homogenate solution of an adult male (4.26 g) and female (2.21 g) P. promelas in deionized water (i.e., a completely new preparation as that used in experiment 2). For both experiments, there were 10 replicates for every treatment combination including negative and solvent controls. For consistency with the pesticide stability experiment and experiments 1 and 2, the measured response was 48-h C. dubia survival. Although 48-h survival was the reported response variable, these and subsequent experiments were continued beyond 48 h and survival was monitored for insight into trends over time.

2.6. Comparison of two P. promelas chemical cues Experiment 5 explored the interactions between two types of P. promelas chemical cues and malathion on 48-h C. dubia survival using a completely randomized factorial design. Chemical cue was at three levels (absence, presence type I, presence type II) and malathion at two concentrations (0.0 and 2.96 mg/L). Chambers assigned the presence of type 1 chemical cue were spiked with a 10-mL aliquot of the P. promelas homogenate solution prepared for experiments 3 and 4. Type II chemical cue consisted of a 10-mL aliquot of a solution prepared from a collection of P. promelas epithelial mucus secretions. Specifically, a male and female P. promelas of similar age and size as those used in the homogenized solution were scraped with a dissecting knife on all

2.8. Chemical analyses Chemical concentrations were measured in triplicate for the stock solution of each pesticide used in all experiments and the initial, 24 h, and 48 h OP concentrations of the stability experiment. A known amount of tributylphosphate (TBP) was added to each sample immediately prior to extraction for OP recovery estimation. Solid-phase extraction (SPE) methods (Belden et al., 2000) were conducted using preconditioned AccuBOND II C18 cartridges (Agilent Technologies, Palo Alto, CA, USA). Aqueous samples were pulled through vacuum straws attached to C18 SPE cartridges with a vacuum manifold (Fisher Scientific, Pittsburgh, PA, USA) with pressure maintained around 15 psi. Compounds were eluted from C18 cartridges with three washes totaling 7 mL of reagent grade hexane:acetone:methylene chloride (1:1:1 v/v/v). Solvents were evaporated to 1 mL under a stream of N2 at 50  C using a Reacti-Vap Evaporator (Pierce, Rockford, IL, USA). Samples were transferred to gas chromatograph (GC) vials and analyzed using a Hewlett-Packard 6890 GC (Palo Alto, CA, USA) with a nitrogen-phosphorus detector.

2.9. Statistical analyses For the acute doseeresponse tests, 48-h LC50 estimates were calculated with maximum likelihood probit analysis (USEPA, 2002). Percent survival data were arcsine square root transformed prior to analyses. All dosee response analyses were performed with ToxCalc v.5.0.20 (TidePool Scientific Software, McKinleyville, CA, USA). The visual cue-contaminant experiment (i.e., experiment 1) was analyzed using a mixed-model analysis of variance (ANOVA) with visual cue as the random factor and dicrotophos and malathion as fixed factors (PROC MIXED, SAS Institute Inc., 1989). Hypotheses of interest were first-order interactions between visual cue and dicrotophos or malathion. The chemical cue-contaminant experiment (i.e., experiment 2) was analyzed using a three-factor analysis of variance (ANOVA) testing for the first-order interaction (PROC GLM, SAS Institute Inc., 1989). The remaining three experiments were analyzed using two-factor ANOVA testing for the first-order interaction effect of predator chemical stimulus and pesticide on 48-h C. dubia survival (PROC GLM, SAS Institute Inc., 1989). Prior to analyses, percent survival data from the acute tests were arcsine square root transformed. All analyses were performed using type III sum of squares and a Z 0.05.

3. Results 3.1. Doseeresponse experiments The doseeresponse tests met the recommended minimum control survival (O 90%) and recommended dissolved oxygen (O4.0 mg/L) and temperature (25 G 1  C) guidelines

J.D. Maul et al. / Environmental Pollution 141 (2006) 90e97

(USEPA, 2002). The 48-h LC50 values from the dosee response tests were 3.35 mg/L (95% confidence interval [95% CI]: 2.68e3.93) for malathion and 19.17 mg/L (95% CI: 14.18e20.25) for dicrotophos. 3.2. Malathion and dicrotophos stability experiment Measured pesticide concentrations for stock solutions were within 84.3 and 95.0% of nominal concentrations for dicrotophos and malathion, respectively. For our experimental test conditions (i.e., static acute tests in moderately hard water with an initial cladoceran feeding), 80.7% of the initial dicrotophos concentration and 53.4% of the initial malathion concentration was measured after 48 h of testing conditions (Table 1). For dicrotophos and malathion aqueous extractions, TBP surrogate recoveries were 94.3 and 94.4%, respectively. In the aqueous testing matrix used in these experiments, malathion concentrations remaining after 48 h was similar to the reported half-life of less than 1 week for malathion in river water (Kamrin, 1997). The expected half-life for dicrotophos resulting from aerobic aquatic metabolism is unknown (USEPA, 2003); however, the aqueous half-life (i.e., hydrolysis) of dicrotophos was 72 d at pH 7.0 under sterile laboratory conditions (Lee et al., 1989) and the reported dicrotophos half-life in an aerobic soil metabolism study was 2.7 d in Hanford sandy loam soil (USEPA, 2003). Because the measured stock concentrations were within 16% of nominal concentrations and results of the malathion component of the stability experiment was similar to the reported half-life under typical environmental conditions, nominal concentrations of dicrotophos and malathion are reported throughout the manuscript. Nominal and measured concentrations used in these experiments were less than the peak expected environmental concentrations for dicrotophos (21.3 mg/L) and malathion (7.9 mg/L) as calculated by the EPA using the PRZM-EXAMS model for the use rates and frequencies of these compounds typically observed for cotton fields in Arkansas (USEPA, 2003, 2004). 3.3. Visual and chemical stimuli and contaminant interactions

93

In the second experiment that examined P. promelas chemical cues, no main effect of the cue was detected at the 48-h endpoint; however, the interaction effect of chemical cues and malathion on percent survival of C. dubia was significant (F2,90 Z 5.3, P Z 0.006). Potentiation of malathion toxicity to C. dubia was observed in the presence of the fish chemical cue and the percent reduction appeared similar for both concentrations tested (Fig. 1). Similarly, decreasing trends in C. dubia survival over several different dicrotophos concentrations differed between experimental chambers receiving fish chemical stimulus and those not receiving the stimulus (F2,90 Z 3.6, P Z 0.032). However, chambers assigned chemical cues and dicrotophos had greater C. dubia survival than those assigned only dicrotophos; a pattern opposite than observed for malathion. In the chemical cue-malathion experiment (experiment 3), there was a significant interaction between predator chemical cue and malathion (F3,72 Z 26.5, P ! 0.0001). At malathion concentrations of 2.69, 2.82, and 2.96 mg/L, 48-h C. dubia survival for the treatment combination containing malathion and predator stressor was significantly reduced from that of malathion alone (P ! 0.01; Fig. 2). For the chemical cue-dicrotophos experiment (experiment 4), the interaction effect was not significant (P O 0.05). At both 15.0 and 17.2 mg/L of dicrotophos, 48-h C. dubia survival for dicrotophos and chemical cue did not differ from those exposed to dicrotophos alone (Fig. 3). In the final experiment that compared two different types of fish chemical cues (experiment 5), 48-h survival did not differ between chambers containing 2.90 mg/L malathion with or without the P. promelas scale/mucus cue (Fig. 4A). However, 48-h C. dubia survival when exposed to 2.90 mg/L malathion and homogenized P. promelas was reduced 72.0% from survival of organisms exposed to 2.90 mg/L malathion alone (F1,8 Z 36.9, P Z 0.0003; Fig. 4B). In experiments 3 and 4, protein concentration was greater (P ! 0.05) in chambers that received a 10-mL aliquot of the P. promelas homogenate solution suggesting that the resulting concentration added protein to experimental chambers, some of which may be compounds important for signaling to cladocerans. 4. Discussion

For experiment 1 that examined visual P. promelas cues, there was no main effect of the cue on 48-h percent C. dubia survival nor interaction effect detected for visual cue and malathion or visual cue and dicrotophos on 48-h percent C. dubia survival.

4.1. LC50 determinations The 48-h LC50 (3.35 mg/L) observed for malathion in this study was greater than that reported for another cladoceran

Table 1 Stability of aqueous dicrotophos and malathion concentrations over time under experimental conditions used for all experiments Pesticide

Time (h)

Mean (mg/L)

SD

CV (%)

n

Decrease (%)

Dicrotophos

0 24 48

9.00 8.49 7.27

e 0.16 0.64

e 1.80 8.83

1 3 3

e 5.7 19.3

Malathion

0 24 48

0.87 0.64 0.47

e 0.03 0.01

e 4.83 3.00

1 3 3

e 26.4 46.6

J.D. Maul et al. / Environmental Pollution 141 (2006) 90e97

94

No Fish Cue Fish Cue

Proportion Surviving

1.0 0.8 0.6 0.4

The interaction effect between fish cues and contaminants also varied by the type of stimulus; the interaction was not observed when the fish cues were visual, although P. promelas were observed pursuing and attempting to consume C. dubia contained within experimental chambers. Alternatively, Daphnia readily respond to chemical cues and such stimuli from predators have been shown empirically to induce anti-predator behavioral responses in eight species of Daphnia (Dodson, 1988).

0.2 0.0 0.0 µg/L

2.69 µg/L

100%

2.82 µg/L

Malathion Survival

80%

Fig. 1. The interaction effect of malathion and P. promelas chemical cue (i.e., addition of homogenized P. promelas) on 48-h Ceriodaphnia dubia survival (P ! 0.05). Each treatment-level combination mean is derived from six replicate experimental chambers. Untransformed proportion affected data are presented. Error bars are standard deviation.

60% 40% 0.0 µg/L 0.0 µg/L + Fish Cue 2.69 µg/L 2.69 µg/L + Fish Cue

20% 0% 0

12

24

36

48

60

72

84

96

Time (h)

100% 80%

Survival

species Daphnia pulex (1.8 mg/L; 95% CI: 1.4e2.4 (USEPA, 1986)). Similarly, the 48-h LC50 for dicrotophos (19.17 mg/L) was greater than that reported for Daphnia magna (12.7 mg/L; 87.7% a.i.; (USEPA, 2003)). This variability may be due in part to the addition of an algal community and additional food resources for C. dubia. These additions introduced organic matter to experimental chambers, which had potential binding capacity for organic contaminants. Presence of an algal community may better simulate environmental conditions in receiving systems during pesticide runoff events as opposed to laboratory toxicity tests utilizing only synthetic water. Because variability in the concentration and type of food items can affect cladoceran survival and reproduction (Enserink et al., 1992) and modify toxicity of chemicals (Folt et al., 1999), it was important that resource limitation did not confound our effort to understand potential modifying factors of contaminant toxicity.

60%

0.0 µg/L 0.0 µg/L + Fish Cue 2.82 µg/L 2.82 µg/L + Fish Cue

40% 20% 0% 0

12

24

36

48

60

72

84

96

Time (h)

100%

4.2. Interactions between contaminant and fish-based chemical cues Survival

80%

Results of our study suggest that chemicals cues from fish affected contaminant toxicity to cladocerans. The ANOVA interactions between contaminant and chemical cues indicate a biological response that differs from a simple additive response (Billick and Case, 1994; Folt et al., 1999). Although the main effect of the chemical stimulus was not significant, when presented to C. dubia in combination with malathion, percent survival was reduced suggesting potentiation of this contaminant. For cladocerans, several studies have explored the interaction of abiotic factors and contaminants (Folt et al., 1999) or modification of contaminant toxicity by environmental variables (Bond and Bradley, 1997). However, only a few studies have examined interactions between chemical cues of a biotic source and exposure to contaminants (Hanazato and Dodson, 1992, 1995).

60% 0.0 µg/L 0.0 µg/L + Fish Cue 2.96 µg/L 2.96 µg/L + Fish Cue

40% 20% 0% 0

12

24

36

48

60

72

84

96

Time (h) Fig. 2. Potentiation of malathion toxicity by P. promelas chemical cues (i.e., addition of homogenized P. promelas). The interaction effect of malathion and chemical cues was significant (P ! 0.05) and acute 48-h Ceriodaphnia dubia survival at all three malathion concentrations tested was significantly reduced in chambers receiving malathion and chemical cues than those receiving malathion alone (P ! 0.05). Error bars are standard deviation.

J.D. Maul et al. / Environmental Pollution 141 (2006) 90e97

100%

100%

80%

Survival

80%

Survival

95

60%

60% 40%

40% 0.0 µg/L 0.0 µg/L + Fish Cue 15.0 µg/L 15.0 µg/L + Fish Cue

20%

0.0 µg/L 0.0 µg/L + epithelial 2.96 µg/L 2.96 µg/L + epithelial

20% 0%

0%

0 0

12

24

36

48

60

12

24

72

36

48

60

72

Time (h)

Time (h) 100% 100% 80%

Survival

Survival

80% 60% 40%

40%

0.0 µg/L 0.0 µg/L + Fish Cue 17.2 µg/L 17.2 µg/L + Fish Cue

20%

12

24

0.0 µg/L 0.0 µg/L + homogenized 2.96 µg/L 2.96 µg/L + homogenized

20%

0% 0

60%

36

48

60

72

Time (h) Fig. 3. Acute 48-h Ceriodaphnia dubia survival in the presence of single and binary combinations of dicrotophos and P. promelas chemical cues. The interaction effect of dicrotophos and chemical cues on 48-h survival was not significant for both dicrotophos concentrations tested (P O 0.05). Error bars are standard deviation.

Furthermore, Cladoceran responses can vary based on the species the chemical extracts are isolated from (Hanazato et al., 2001; Weber, 2003; Weber and Declerck, 1997; Weber and Van Noordwijk, 2002; Weber and Vesela, 2002). Of the two potential types of chemical stimuli examined in the current experiments, the homogenized P. promelas had greater potentiation of malathion toxicity than the epithelial-based (i.e., scale/mucus) solution, demonstrating the variation in strength of each type of chemical cue and the importance of examining multiple sources of chemical cues (Parejko and Dodson, 1990). Weber (2003) suggested that for chemical cues from fish, cladocerans may be responding to a suite of chemicals rather than a single exuded compound. The weaker response observed for epithelial-based chemical cues versus whole fish-based cues for the same species of fish may provide an additional perspective into the fish chemical cocktail proposed by Weber (2003). For example, understanding the precise source of chemical cues from a fish may help to explain variation in cladoceran responses among chemical cues from different fish species (Weber, 2003). It is unknown whether fish chemicals in the current study stimulated anti-predatory responses in C. dubia. However, chemical cues from both vertebrate and invertebrate predators have been shown to induce a number of defense mechanisms and variability in life history traits in cladocerans (Hanazato et al., 2001; Hebert and Grewe, 1985; Kreuger and Dodson,

0% 0

12

24

36

48

60

72

Time (h) Fig. 4. Acute 48-h Ceriodaphnia dubia survival in the presence of single and binary combinations of malathion and P. promelas chemical cues consisting of either epithelial mucus or addition of homogenized whole fish. Significant potentiation of malathion toxicity occurred with the addition of homogenized whole fish (P ! 0.05), but not for addition of epithelial mucus (P O 0.05). Error bars are standard deviation.

1981; Weber, 2003; Weber and Declerck, 1997). These include an immediate escape response that involves vertical (downward) movement in a water column (Dodson, 1988) and predator-induced cyclic morphological change (i.e., cyclomorphosis) in offspring that reduces predator handling ability and increases escape ability (Dodson, 1989; Havel and Dodson, 1984). Both responses offer a defense against predation, but may entail an energetic and reproductive cost (e.g., production of fewer offspring; Dodson, 1984). If the treatment of P. promelas chemical cues in the current study did indeed induce such responses, one explanation for the interaction with malathion may be that the associated energetic costs may have reduced C. dubia capacity to cope with a simultaneously incurred contaminant stressor as suggested by Hanazato (1999). Similar potentiation of malathion and carbaryl toxicity by chemical cues of predatory salamanders has been reported recently for gray tree frog tadpole (Hyla versicolor) survival (Relyea, 2004; Relyea and Mills, 2001). The chemical cue and contaminant interaction was observed for malathion, but not dicrotophos. Variation among these pesticides may be related to malathion being a bioreactive phosphorodithioate OP compound. Phase I metabolism of malathion by cytochrome P450 monooxygenases (Di Giulio et al., 2003) produce metabolites that are more

96

J.D. Maul et al. / Environmental Pollution 141 (2006) 90e97

efficient acetylcholinesterase inhibiters than the parent form (i.e., increased toxicity). Unlike malathion, dicrotophos is only metabolized to less toxic metabolites. Hypotheses for the difference in patterns observed between OPs include (1) that an increase in malathion toxicity may be due to alteration in C. dubia metabolism in response to fish-based chemical cues or (2) the contents of the homogenized solution may contain constituents that are capable of metabolizing malathion to its more toxic metabolite. However, these hypotheses need to be tested experimentally before drawing strong conclusions as to the mechanism of these interactions. In the event that fish chemical cues in this study did elicit a predator response, our design differed from Smith and Weis (1997) and Schulz and Dabrowski (2001) in that effects of predator and contaminants on cladoceran survival were separated. As such, the ultimate source of mortality occurred solely by toxicity of contaminant while chemical cues acted as a modifier of pesticide toxicity and not a contributor to the response. Although results of the current study are consistent with other similarly designed studies (Hanazato and Dodson, 1992; Hanazato and Dodson, 1995), additional effects of predators on C. dubia, such as mortality, cannot be evaluated from these studies when only examining chemical cues. 5. Conclusions Results of this study suggest that fish-based chemical cues alone do not have a significant impact on C. dubia survival, but induce substantial potentiation of malathion toxicity when organisms are exposed to both in combination. These results provide further evidence of the additional information that can be obtained from toxicity tests that include multiple variables for protecting the aquatic environment (Cairns, 1983, 1986). Furthermore, C. dubia are frequently used as an indicator organism for aqueous toxicity or waterbody impairment for freshwater organisms. Use of C. dubia for aquatic monitoring or other toxicity tests that do not consider environmental variables such as chemical cues of other species may underestimate risk to aquatic organisms.

Acknowledgments This research was funded by a Sigma Xi Grant-in-Aid of Research (GIAR) and the Judd Hill Foundation of Arkansas State University. We especially thank J. Bouldin, R. Buchanan, C. Cooper, A. Grippo, R. Maul, N. Enger, H. McIntire, G. Ogendi, and W. Stephens for constructive criticisms on earlier drafts of the manuscript, L. Harding for assistance with the bioassays, and K. Cole and H. Ju for chemical analyses. References Belden, J.B., Hofelt, C.S., Lydy, M.J., 2000. Analysis of multiple pesticides in urban storm water using solid-phase extraction. Archives of Environmental Contamination and Toxicology 38, 7e10.

Billick, I., Case, T.J., 1994. Higher order interactions in ecological communities: what are they and how can they be detected? Ecology 75, 1529e1543. Bond, J., Bradley, B.P., 1997. Resistance to malathion in heat-shocked Daphnia magna. Environmental Toxicology and Chemistry 16, 705e712. Boone, M.D., Semlitsch, R.D., 2001. Interactions of an insecticide with larval density and predation in experimental amphibian communities. Conservation Biology 15, 228e238. Cairns Jr., J., 1983. Are single species toxicity tests alone adequate for estimating environmental hazard? Hydrobiologia 100, 47e57. Cairns Jr., J., 1986. The myth of the most sensitive species. BioScience 36, 670e672. Cooney, J.D., 2003. Freshwater tests. In: Rand, G. (Ed.), Fundamentals of Aquatic Toxicology. Taylor & Francis, Washington, DC, pp. 71e102. Di Giulio, R.T., Benson, W.H., Sanders, B.M., Van Veld, P.A., 2003. Biochemical mechanisms: metabolism, adaptation, and toxicity. In: Rand, G. (Ed.), Fundamentals of Aquatic Toxicology. Taylor & Francis, Washington, DC, pp. 523e561. Dodson, S.I., 1984. Predation of Heterscope septentrionalis on two species of Daphnia: morphological defenses and their cost. Ecology 65, 1249e1257. Dodson, S.I., 1988. The ecological role of chemical stimuli for the zooplankton: predator-avoidance behavior in Daphnia. Limnology and Oceanogeography 33, 1431e1439. Dodson, S.I., 1989. Predator-induced reaction norms. BioScience 39, 447e452. Enserink, L., de la Haye, M., Maas, H., 1992. Reproductive strategy of Daphnia magna: implications for chronic toxicity. Aquatic Toxicology 25, 111e124. Folt, C.L., Chen, C.Y., Moore, M.V., Burnaford, J., 1999. Synergism and antagonism among multiple stressors. Limnology and Oceanography 44, 864e877. Hanazato, T., 1999. Anthropogenic chemicals (insecticides) disturb natural organic chemical communication in the plankton community. Environmental Pollution 105, 137e142. Hanazato, T., 2001. Pesticide effects on freshwater zooplankton: an ecological perspective. Environmental Pollution 112, 1e10. Hanazato, T., Dodson, S.I., 1992. Complex effects of a kairomone of Chaoborus and an insecticide on Daphnia pulex. Journal of Plankton Research 14, 1743e1755. Hanazato, T., Dodson, S.I., 1995. Synergistic effects of low oxygen concentration, predator kairomone, and a pesticide on the cladoceran Daphnia pulex. Limnology and Oceanography 40, 700e709. Hanazato, T., Fueki, K., Yoshimoto, M., 2001. Fish-induced life-history shifts in the cladocerans Daphnia and Simocephalus: are they positive or negative responses. Journal of Plankton Research 23, 945e951. Hanazato, T., Ooi, T., 1992. Morphological responses of Daphnia ambigua to different concentrations of a chemical extract from Chaoborus flavicans. Freshwater Biology 27, 379e385. Havel, J.E., Dodson, S.I., 1984. Chaoborus predation on typical and spined morphs of Daphnia pulex: behavioral observations. Limnology and Oceanography 29, 487e494. Hebert, P.D.N., Grewe, P.M., 1985. Chaoborus-induced shifts in the morphology of Daphnia ambigua. Limnology and Oceanography 30, 1291e1297. Held, J.W., Peterka, J.J., 1974. Age, growth, and food habits of the fathead minnow, Pimephales promelas, in North Dakota saline lakes. Transactions of the American Fisheries Society 103, 743e756. Kamrin, M.A., 1997. Pesticide Profiles. Lewis Publishers, New York. Kreuger, D.A., Dodson, S.I., 1981. Embryological induction and predation ecology in Daphnia pulex. Limnology and Oceanography 26, 219e223. Lee, P.W., Stearns, S.M., Hernandez, H., Powell, W.R., Naidu, M.V., 1989. Fate of dicrotophos in the soil environment. Journal of Agricultural and Food Chemistry 37, 1169e1174. Lynch, M., 1979. Predation, competition, and zooplankton community structure: an experimental study. Limnology and Oceanography 24, 253e272. Odum, E.P., 1985. Trends expected in stressed ecosystems. BioScience 35, 419e422. Parejko, K., Dodson, S.I., 1990. Progress towards characterization of a predator/prey kairomone: Daphnia pulex and Chaoborus americanus. Hydrobiologia 198, 51e59. Relyea, R.A., 2004. Synergistic impacts of malathion and predatory stress on six species of North American tadpoles. Environmental Toxicology and Chemistry 23, 1080e1084.

J.D. Maul et al. / Environmental Pollution 141 (2006) 90e97 Relyea, R.A., Mills, N., 2001. Predator-induced stress makes the pesticide carbaryl more deadly to gray treefrog tadpoles (Hyla versicolor). Proceedings of the National Academy of Sciences of the USA 98, 2491e2496. SAS Institute Inc, 1989. In: SAS/STAT User’s Guide. Version 6, fourth ed., vol. 2. SAS Institute Inc., Cary, NC. Schulz, R., Dabrowski, J.M., 2001. Combined effects of predatory fish and sublethal pesticide contamination on the behavior and mortality of mayfly nymphs. Environmental Toxicology and Chemistry 20, 2537e2543. Smith, G., Weis, J.S., 1997. Predator/prey interactions in Fundulus heteroclitus: Effects of living in a polluted environment. Journal of Experimental Marine Biology and Ecology 209, 75e87. Thurman, E.M., Zimmerman, L.R., Scribner, E.A., Coupe Jr., R.H., 1998. Occurrence of cotton pesticides in surface water of the Mississippi Embayment. US Geological Survey Fact Sheet 022-98. !http://ks.water. usgs.gov/Kansas/pubs/fact-sheets/fs.022-98.htmlO. USDA, 2003. Agricultural Chemical Usage 2001 Field Crops Summary. US Department of Agriculture, National Agricultural Statistics Service, !http://usda.mannlib.cornell.edu/reports/nassr/other/pcu-bb/agcs0502.pdfO. USEPA, 1986. Quality Criteria for Water. EPA 440/5-86-001. US Environmental Protection Agency, Washington, DC.

97

USEPA, 2002. Methods for Measuring the Acute Toxicity of Effluents and Receiving Waters to Freshwater and Marine Organisms, fifth ed. EPA821-R-02e012. US Environmental Protection Agency, Washington, DC. USEPA 2003. EFED Red Chapter for Dicrotophos. US Environmental Protection Agency, Washington, DC. !http://www.epa.gov/pesticides/ op/dicrotophos/efed-dicro.pdfO. USEPA 2004. Malathion Reregistration Eligibility Document Environmental Fate and Effects Chapter. US Environmental Protection Agency, Washington, DC. !http://www.epa.gov/pesticides/op/malathion/efed_part1.pdfO. Weber, A., 2003. More than one ‘fish kairomone’? Perch and stickleback kairomones affect Daphnia life history traits differently. Hydrobiologia 498, 143e150. Weber, A., Declerck, S., 1997. Phenotypic plasticity of Daphnia life history traits in response to predator kairomones: genetic variability and evolutionary potential. Hydrobiologia 360, 89e99. Weber, A., Van Noordwijk, A., 2002. Swimming behaviour of Daphnia clones: differentiation through predator infochemicals. Journal of Plankton Research 24, 1335e1348. Weber, A., Vesela, S., 2002. Optimizing survival under predation: chemical cues modify curvature in Daphnia galeata. Aquatic Ecology 36, 519e527.