Sequential exposure to low levels of pesticides and temperature stress increase toxicological sensitivity of crustaceans

Sequential exposure to low levels of pesticides and temperature stress increase toxicological sensitivity of crustaceans

Science of the Total Environment 610–611 (2018) 563–569 Contents lists available at ScienceDirect Science of the Total Environment journal homepage:...

652KB Sizes 2 Downloads 68 Views

Science of the Total Environment 610–611 (2018) 563–569

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Sequential exposure to low levels of pesticides and temperature stress increase toxicological sensitivity of crustaceans Renato Russo, Jeremias Martin Becker, Matthias Liess ⁎ UFZ, Helmholtz-Centre for Environmental Research, Department of System-Ecotoxicology, Permoserstraße 15, 04318 Leipzig, Germany RWTH Aachen University, Institute for Environmental Research (Biology V), Aachen, Germany

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Agricultural streams are subjected to sequential pesticide contaminations. • Environmental stressors interact with pesticides in the field. • Crustaceans from agricultural streams showed increased toxicological sensitivity. • A synergistic interaction between pesticide and temperature stress was revealed. • A realistic risk assessment needs to account for this pesticide-stress interaction.

a r t i c l e

i n f o

Article history: Received 20 April 2017 Received in revised form 8 August 2017 Accepted 8 August 2017 Available online xxxx Editor: D. Barcelo Keywords: Multiple stressors Synergistic interaction Risk assessment Bioindicator SPEAR Environmental stressors Agricultural streams

a b s t r a c t Frequent pesticide-related impacts on ecosystems at concentrations considered environmentally safe indicate that the current risk assessment framework for registration of pesticides is not protective enough. Causes may include difficulties in assessing the effects of sequential pesticide pulses and their interaction with environmental stressors. By contrast to such realistic scenarios, risk assessment for registration of pesticides is typically based on tests of a single exposure period under benign laboratory conditions. Here, we investigated the toxicological sensitivity of Gammarus pulex, an ecologically relevant crustacean, from uncontaminated control streams and pesticide-contaminated agricultural streams by exposing them to pesticide contamination in the laboratory. Individuals from contaminated streams were 2.7-fold more sensitive to pesticide exposure than individuals from the reference streams. We revealed that this increase in sensitivity was the result of a synergistic interaction of sequential pesticide exposure and temperature stress. Such multiple stressor scenarios are typical for agricultural streams. We conclude that the interactive effects of sequential toxicant exposure and additional environmental stressors need to be considered in a realistic risk assessment framework. © 2017 Elsevier B.V. All rights reserved.

1. Introduction

Abbreviations: CS, contaminated sites; US, uncontaminated sites. ⁎ Corresponding author at: UFZ, Helmholtz-Centre for Environmental Research, Department of System-Ecotoxicology, Permoserstraße 15, 04318 Leipzig, Germany. E-mail address: [email protected] (M. Liess).

http://dx.doi.org/10.1016/j.scitotenv.2017.08.073 0048-9697/© 2017 Elsevier B.V. All rights reserved.

Risk assessment for the registration of pesticides has been established to protect non-target communities. To address uncertainties related to the projection of toxicity assessments from benign laboratory conditions towards the field conditions and to predict the regulatory acceptable concentration (RAC) (Office of pesticide programs U.S.

564

R. Russo et al. / Science of the Total Environment 610–611 (2018) 563–569

environmental protection agency, 1999), an assessment factor of 100 below the acute LC50 (concentration that is lethal to 50% of the test organisms) has been established. However, the impacts of pesticides on the structure (Liess and Von Der Ohe, 2005; Schäfer et al., 2012) and the biodiversity (Beketov et al., 2013) of agricultural streams have been observed frequently. Non-compliance with regulations during agricultural practices may contribute to this problem; however, frequent occurrence of environmental impacts of pesticides indicates that the current framework of risk assessment omits relevant processes that determine ecologically effective concentrations. The European Food Safety Authority (EFSA) lists several sources of uncertainty in the projection from test systems to the field (European Food Safety Authority (EFSA), Parma, Italy, 2013), including sequential exposure to a mixture of pesticides within one generation (Ashauer et al., 2007), combined effects of pesticides and environmental stressors (Liess et al., 2016) and culminating effects induced by sequential contamination over several generations (Liess et al., 2013). Single short-term pulses of pesticides that are typical in agricultural streams (Liess et al., 1999; Handy, 1994) are known to cause delayed adverse effects on aquatic invertebrates (Abel, 1980) and fish (Floyd et al., 2008). For example, Liess (2002) showed that caddisflies exposed to fenvalerate for 1 h at 1/1000 of the acute LC50 suffered increased mortality 8 months after this brief exposure. Similar delayed effects of shortterm exposure were identified for various invertebrate species exposed to the insecticides esfenvalerate (Beketov and Liess, 2005), thiacloprid (Liess et al., 2013; Beketov and Liess, 2008), imidacloprid (Nyman et al., 2013; Agatz et al., 2014) and endosulfan (Barry and Logan, 1998). Apparently, short-term exposure to a toxicant may result in long-term weakening of individuals that can cause the observed delayed effects. Furthermore, recent investigations showed that sequential exposure patterns may progressively increase the sensitivity of soil (Jordaan et al., 2012; Reinecke and Reinecke, 2005) and aquatic (Ashauer et al., 2017; Ashauer et al., 2015) organisms within one generation. Toxicokinetic and toxicodynamic (TK/TD) models, such as the GUTS (General Unified Threshold model for Survival) approach (Ashauer et al., 2016) have been suggested to predict the effects of sequential pulse exposure; it accounts for variable toxicant exposure over time, allowing the prediction of survival after different exposure patterns and time-scales. However, they are not validated in the field context where environmental stressors are present. For a realistic assessment of pesticide effects in the field, this additional stress needs to be included. Recently, a meta-analysis has shown that combined stressors increased toxicological sensitivity of organisms by more than one order of magnitude (Liess et al., 2016). Accordingly, in the present study, we investigated whether the sensitivity of Gammarus pulex in agricultural streams was affected by sequential exposure to pesticides in combination with additional environmental stressors. 2. Materials and methods 2.1. Study design Gammarus pulex (Linnaeus, 1758) was selected as the test organism because of its ecological relevance in many stream ecosystems. It plays a pivotal role in the degradation of allochthonous leaf litter (Dangles et al., 2004), a crucial ecosystem function that can be disturbed by pesticide contamination (Brosed et al., 2016). G. pulex individuals were sampled from four uncontaminated stream sections and from four sites contaminated with agricultural pesticides, and they were subsequently exposed to the pyrethroid insecticide esfenvalerate in the laboratory. Sampling of test organisms was scheduled according to the expected regime of insecticide exposure in the field, which was estimated based on the 2015 governmental recommendations for pesticide application in Saxony, Saxony-Anhalt and Lower Saxony (Germany) provided by the Saxony-Anhalt State Institute for Agriculture, Forestry and Horticulture

(LLFG), Bernburg (Fig. S1). At each site, crustaceans were sampled during the following three time periods: (i) Autumn (October 2015): 3–4 months after maximum pesticide application (Liess et al., 1999; Huber et al., 2000), corresponding to “no/low pesticide exposure” in the field (Fig. S1); (ii) Spring (March–April 2015): at the beginning of pesticide application (Liess et al., 1999; Huber et al., 2000), corresponding to “low pesticide exposure” in the field (Fig. S1); and (iii) Summer (June 2015): during maximum pesticide application (Liess et al., 1999; Huber et al., 2000), corresponding to “highest pesticide exposure” in the field for the three sampling periods (Fig. S1).

Notably, G. pulex produces two or three overlapping generations per year (Welton, 1979a). Reproduction is particularly low in winter. Therefore, individuals sampled in early spring generally belonged to the same generation as those that found in the previous autumn. By contrast, reproduction strongly increases in early summer causing high generation turnover from spring to autumn. Accordingly, in our study, we sampled different generations of G. pulex according to the timing of field campaigns. 2.2. Selection of the study sites Selected streams were located in Saxony and Saxony-Anhalt, Germany (Fig. S2) and they were characterized by the following parameters: average width of 2 m, average depth of 0.3 m, available hard structure (i.e., stones and wood) in the sampling area, buffer strips on both banks and an uncontaminated refuge area within a range of 5– 15 km up- or down-stream from the sampling point. Additionally, we selected sites with no wastewater treatment plants present within at least 3.0 km upstream to exclude exposure to contaminants other than pesticides. Water conductivity, temperature and pH were measured at each sampling date. Due to technical issues (i.e., low water levels), we could not use the complete data set of 24 observations (8 observations for each of the 3 field campaigns) for all analyses (Table S4). In the following sections, the data set used for each analysis is specified. 2.3. Assessment of pesticide exposure and categorization of sampling sites It is highly demanding to accurately assess the overall exposure of aquatic invertebrates to pesticides in the field. Discharge of pesticides generally occurs in several peaks (Liess et al., 1999; Handy, 1994) driven by rainfall events with run-off. Such sequential exposure to various pesticides exerts complex mixture effects. Additionally, bioavailability and effects of toxicants depend on various parameters including suspension load (Schulz and Liess, 2001), temperature (Harwood et al., 2009) and behavior of individuals (Rasmussen et al., 2013). Because of these considerations, we quantified the magnitude of toxic pressure with a biological measure, the indicator SPEARpesticides (Liess and Von Der Ohe, 2005). This indicator system analyzes the community composition of macroinvertebrates at a given site to estimate toxic pressure on invertebrates. It provides the advantage of assessing the overall biological effects of pesticides, including the bioavailability and the combined effects of sequential exposure and mixtures of pesticides (Liess and Von Der Ohe, 2005; Schäfer et al., 2012; Münze et al., 2017). Macroinvertebrate communities were sampled in early summer during maximum pesticide exposure (Liess et al., 1999), the time period for which the SPEAR index has been validated and best indicates pesticide exposure (Liess and Von Der Ohe, 2005; Schäfer et al., 2012; Orlinskiy et al., 2015a; Münze et al., 2015). At each stream, ten subsamples were collected across different habitats to obtain a representative sample of the macroinvertebrate community (according to the protocol for SPEAR sampling, http://www.systemecology.eu/spearcalc/index.en.

R. Russo et al. / Science of the Total Environment 610–611 (2018) 563–569

html). The macroinvertebrates were collected using a 25-cm2 kick-net with a mesh size of 500 μm. The content of the net was transferred to a plastic tray filled with water to allow for the identification of species and their abundance. If a species could not be identified in the field, some specimens were preserved in ethanol for determination in the laboratory. Determination was performed using a stereo-microscope (Zeiss Discovery V20, Oberkochen, Germany). SPEAR indices were calculated using the program SPEAR Calculator (version 0.9.0, Leipzig, Germany, http://www.systemecology.eu/ spearcalc/index.en.html). Study sites were categorized as contaminated (CS) or uncontaminated (US) according to the results of the calculations (Table S1). All streams with a SPEAR value of ≤33 were considered contaminated (Beketov et al., 2009). According to the SPEAR calculations, contaminated streams were characterized by a low level of pesticide exposure, resulting in an altered community structure with only sublethal effects for crustaceans. This result was confirmed by using t-tests to compare the abundance of G. pulex populations in CS and US during each season; the tests showed no significant differences in population densities for all seasons (results not shown). 2.4. Land-use calculation To identify streams with agricultural catchment, we estimated landuse coverage of selected streams using a section of 1.5 km upstream from the sampling point and 100 m wide from each bank. We applied ArcGIS 10.4 (ESRI, Redlands, CA, USA), using shape files from the ATKIS database provided by the German Federal Agency for Cartography and Geodesy, Leipzig, Germany. CS were predominantly adjacent to land used for agriculture (72 ± 11.8%), while US were not (0%, Table S1). 2.5. Acute toxicity tests G. pulex were collected using a kick-net with a 500-μm mesh. Individuals within a size range of 6–10 mm were transferred to 5-L plastic beakers filled with an adequate volume of stream water, provided with aeration and transported to the laboratory within cooling boxes. In the laboratory, the animals were placed in aerated plastic trays containing stream water and provided with field-collected leaves ad libitum for 24 h. This procedure was performed to (i) exclude an altered response to exposure due to different environmental conditions in the field and (ii) allow individuals to acclimate to the test temperature (18 ± 1 °C). Although G. pulex collected at different seasons may show different temperature optima (Krog, 1954), we decided to continually apply the same temperature to ensure the same conditions of exposure during laboratory acute tests because chemical properties of pesticides vary at different temperatures (Cairns et al., 1975). Consequently, individuals collected in autumn and spring might have been subjected to some additional stress from adaptation to the test temperature (18 °C) due to the difference in stream temperature at different seasons. Nonetheless, this setting represented the best compromise among the competing requirements. The substance selected for sensitivity testing in the laboratory was esfenvalerate, (s)-cyano(3-phenoxyphenyl) methyl-(s)-4-chloroalpha-(1-methylethyl) benzeneacetate. This pyrethroid insecticide has been reported as highly toxic to G. pulex (Hill, 1985) but was detected in streams of the study area at concentrations that, compared to other insecticides, exert lower toxicity on macroinvertebrates (Münze et al., 2017). Esfenvalerate was obtained from Sigma-Aldrich (Merck KGaA, Darmstadt, Hessen, Germany) in powder form. Stock solutions were prepared by dissolving a known weight of esfenvalerate in DMSO. Bowman et al. (Bowman et al., 1981) investigated the tolerance of three different crustacean species to DMSO; they have observed a LOEC (Lowestobserved-effect concentration) of 2%. Therefore, to prevent additional effects of the solvent on G. pulex, the dilution volume was adjusted to obtain final DMSO concentration in the contamination medium b 0.3%

565

- one order of magnitude lower than the LOEC - for all tested conditions. The concentration of esfenvalerate in the test media was subsequently determined based on the European standard procedure for water quality (ISO 10695:2000, 2000). Analyses were performed by Wessling GmbH, Landsberg OT, Oppin, Germany, using a TSQ™ 8000 Evo Triple Quadrupole GC-MS/MS (Thermo Fisher Scientific, Hennigsdorf, Germany; detection limit 5.7 ng/L). A TG-5HT guard column (ID 0.53 mm, film thickness 0.15 μm; Thermo Fisher Scientific, Hennigsdorf, Germany) was applied, and dichloromethane was used as the eluent. TraceFinder™ software (Thermo Fisher Scientific, Hennigsdorf, Germany) was used to process the data. We analyzed the contaminated medium for the three different esfenvalerate test concentrations and the control medium. On average, measured concentrations deviated from the nominal concentrations by ±2.8%. All results reported in subsequent sections refer to nominal concentrations. Acute toxicity tests were performed using an approach adapted from the OECD guidelines for the testing of chemicals (OECD, 2002). Nominal test concentrations used were the control, 0.32 μg/L, 0.63 μg/L and 1.28 μg/L. Test organisms were exposed to the contaminant in glass vessels containing Elendt M7 medium (OECD, 1998) for 1 h to simulate relevant exposure times in the field. Eight individuals were grouped in one glass vessel with 1.5 L of the test medium. One vessel was used for the 0.32 μg/L and control levels, while two vessels per site and season were used for the 0.63 μg/L and 1.28 μg/L contamination levels to reduce variability of setups in which high mortality was expected. Throughout the experiment, temperature was maintained at 18 ± 1 °C, and test solutions were oxygenated at approximately 100%. After 1 h of exposure, test animals were transferred to glass vessels containing fresh M7 medium. Mortality was recorded at 1 h, 24 h and 48 h after exposure. G. pulex individuals were considered dead if they were unable to swim and there was no visible movement (e.g., leg movement) within 15 s after transferring them to the fresh medium. A data set of 21 observations was applied (Table S4). The results shown here refer to mortality after 48 h, as differences among experimental groups were highest at this test duration. In the following sections, LC50 refers to the median lethal concentration after exposure for 1 h and 48 h of observation. 2.6. Data analyses Statistical analyses were conducted with the software RStudio for Mac (version 0.98.501; R Core Team 2014, Vienna, Austria). A dataset of 18 observations, 6 for each of the three field campaigns, was used (Table S4). From the acute toxicity tests with esfenvalerate, LC50 and the 95% confidence intervals were estimated using the package drc (Ritz and Streibig, 2005) (version 2.5–12). For each season, LC50 values for CS and US were compared using a t-test. Additionally, mean LC50 values for different seasons were compared within each group of sites using paired t-tests. To test (i) whether average LC50 values over time differed between CS and US and (ii) whether additional environmental factors affected LC50 values, we constructed a linear mixed effects model with data from all seasons by applying the lme4 package (Bates et al., 2014) (version 1.1–7). Significant fixed effects were identified by applying a forward selection. First, we created separate models with restricted maximum likelihood (REML); each model included only the main effect of one of the following environmental variables: (i) contamination categories based on the SPEAR indices in summer, (ii) relative abundance of G. pulex at the sampling site, (iii) biodiversity (Shannon index), (iv) evenness (Pielou index), (v) conductivity, (vi) pH and (vii) temperature of the stream water. Second, these environmental variables were sorted in order of their significance in one-way models (based on type II Wald χ2 tests) and then sequentially added to a null model (with only the intercept as the fixed effect) to determine if their addition significantly improved the model according to a likelihood ratio test. Finally, the same procedure was repeated with two-way interactions. The final model included the main effects and the interaction of contamination

566

R. Russo et al. / Science of the Total Environment 610–611 (2018) 563–569

and temperature as fixed effects. A random intercept was included for each sampling site to avoid pseudoreplication resulting from repeated sampling of the same sites. Additionally, a random intercept was included for each combination of contamination and sampling season. By considering random effects, we excluded the possibility that the observed increase in sensitivity with temperature in the CS might have resulted only from increased pesticide contamination in the summer, which coincides with higher temperatures. Required homoscedasticity and normal distribution of residuals were confirmed by visual inspection using normal-Q-Q plots and residuals vs. fitted values plots. The fitted values of the final model were extracted by applying the effects package (Fox, 2003) (version 3.0–3). A post hoc analysis of the results was conducted by applying the phia package (De Rosario-Martinez, 2015) (version 0.2–1).

3.2. Environmental factors determining the sensitivity of G. pulex individuals We applied a linear mixed effects analysis to identify relevant factors affecting the sensitivity of crustaceans to esfenvalerate. Among all environmental variables tested as fixed effects (see the Materials and methods section), only the contamination categories (based on the toxic pressure estimated with SPEAR) and stream temperature significantly determined the sensitivity of G. pulex to esfenvalerate (Table S3). The effect of temperature interacted synergistically with the effect of contamination (χ2 = 4.67, d.f. = 1, p b 0.05). At US, we did not observe a significant change in LC50 with increasing temperature (χ2 = 0.30, d.f. = 1, p = 0.590; Fig. 2). By contrast, at CS, the LC50 decreased from 0.88 to 0.31 μg/L when the temperature increased from 6 to 16 °C (χ2 = 21.4, d.f. = 1, p b 0.001; Fig. 2).

3. Results

4. Discussion

3.1. Difference in sensitivity of G. pulex individuals to esfenvalerate

4.1. Increased sensitivity of previously exposed G. pulex individuals

In autumn and spring, when pesticide exposure is low, there was no difference in sensitivity of G. pulex individuals from CS compared with US (Table S2). By contrast, in early summer, when pesticide exposure reaches its maximum in the field (Liess and Von Der Ohe, 2005) (Fig. 1, summer), the LC50 of individuals from US was 2.7-fold higher than that of individuals from CS (t = 6.9192, d.f. = 0.398, p b 0.005). Apparently, the acute sensitivity of G. pulex to esfenvalerate increased in the presence of pesticide application in summer, even though the population density did not differ significantly between CS and US during the same period (results not shown). Additionally, individuals from CS showed a slightly lower LC50 in summer than in autumn (2.2-fold, t = − 3.5546, d.f. = 2, p = 0.071), while the sensitivity of individuals from US did not significantly differ over all seasons (Fig. 1).

We observed that in early summer, when pesticides and especially insecticides are applied most frequently (Liess et al., 1999; Huber et al., 2000), the sensitivity of G. pulex to esfenvalerate increased in agricultural streams compared to that of individuals collected from reference sites. This effect was not found in spring and autumn, when fewer pesticides are applied (Liess et al., 1999; Huber et al., 2000). However, we did not detect a significantly reduced population density in agricultural streams over the investigation time. Migration from uncontaminated areas up- or down-stream, which is common for Gammarus species (Macan and Mackereth, 1957; Minckley, 1964; Hultin, 1971),

Fig. 1. Median 48 h LC50 values of G. pulex collected from contaminated streams (CS, orange) or uncontaminated streams (US, blue) and exposed to esfenvalerate for 1 h in the laboratory. The central rectangle spans the first quartile to the third quartile, the segment inside the rectangle shows the median, and whiskers above and below the box show locations of the minimum and maximum. The three paired values correspond to the seasons of sampling. Significance levels of the 21 observations are displayed; n.s.: p ≥ 0.05, ***: p b 0.001. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 2. LC50 values predicted by the model according to stream temperature. Increasing temperatures at contaminated sites (CS, orange) are associated with higher toxicant sensitivity (lower LC50). Temperature dependency is considerably less at uncontaminated sites (US, blue). Sampling sites and sampling seasons were included as random effects. The mean ± 95% CI is presented. Marginal R2 (fixed effects only) = 0.74; conditional R2 (fixed and random effects) = 0.78. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

R. Russo et al. / Science of the Total Environment 610–611 (2018) 563–569

and increased reproduction from spring to autumn may have compensated the consequences of the pesticide effects on population density. In contrast to our observations, other studies have shown that macroinvertebrates evolve resistance under sequential exposure to toxicants in the field (Clark et al., 2015; Weston et al., 2013). These disparities can be reconciled by considering the differences in ecological scenarios of the respective investigations. Unlike other studies, our investigation used sites close to uncontaminated refuge sections that likely served as sources of immigrating non-resistant individuals (Welton, 1979b). Additionally, refuge sections also preserve species diversity of exposed freshwater communities (Valle et al., 2013; Orlinskiy et al., 2015b), which may have slowed the evolution of pesticide resistance through decreased intraspecific selection pressure (Becker and Liess, 2015; Becker and Liess, 2017). Finally, selection pressure for resistance may have occurred in summer through the reduced performance of sensitive individuals. However, pesticide resistance is generally associated with fitness costs in an uncontaminated environment (Vigneron et al., 2015; Feng et al., 2009). Therefore, resistant individuals in agricultural streams are subjected to reverse selection pressure in autumn and winter when no pesticide exposure is expected. This reverse selection may have counterbalanced the selection for resistance in summer under the specific condition of the sites investigated. Accordingly, in this study, G. pulex from contaminated streams were weakened in the field and resulted in higher sensitivity to the exposure to esfenvalerate in the laboratory. 4.2. Temperature, an environmental stress further increasing the sensitivity of previously exposed individuals Of several environmental parameters investigated, only the temperature in streams synergistically magnified the effect of pesticide exposure on G. pulex. Previous studies have shown that the sensitivity of aquatic organisms to toxicants can be generally increased by the presence of additional environmental stressors (Liess et al., 2016; Liess et al., 2013; Liess, 2002; Coors and De Meester, 2008; Enserink et al., 1990; Relyea and Mills, 2001), including temperature stress (Ostenauer and Köhler, 2008; Tak et al., 2015). This effect was observed, for example, in the crustaceans Hyalella azteca and Daphnia magna exposed to the fungicide chlorothalonil and the insecticide malathion, respectively (Willming et al., 2013); the two contaminants exerted stronger lethal and sublethal effects when exposed to fluctuating temperatures in a non-optimal range compared with organisms exposed to a constant temperature regime. Similarly, Lydy et al. (Lydy et al., 1999) observed amplified effects of the insecticides methyl-parathion and chlorpyrifos on the diptera Chironomus tentans with increasing temperatures. In our study, water temperature in summer was on average 2 °C higher at CS than at US; notably, temperature ranged from 5.7 °C to 16 °C across all seasons and sites. Such temperatures may have affected crustaceans even though they were well below the upper thermal limit for G. pulex (approximately 25 °C) (Sutcliffe et al., 1981). When temperature increases, the metabolic rate of organisms increases; however, concurrently oxygen concentration in water is reduced. Accordingly, the enhanced metabolic activity is not supported by a concomitant increase in oxygen availability, exerting stress on organisms (Khan and Khan, 2008). Recently, Verberk et al. (Verberk et al., 2016) investigated the role of the interacting effects of oxygenation and warming in freshwater systems at ecologically relevant temperatures. They showed that an increase of 2 °C over the ambient temperature in the field resulted in delayed effects on aquatic macroinvertebrates due to the interacting effects of warming and hypoxia. Therefore, even mild environmental stress, such as a slightly increased temperature that is still well within physiological limits, may synergistically increase the toxicological sensitivity of organisms (Fig. 2, orange line). By contrast, when crustaceans were not exposed to pesticides in the field (Fig. 2, blue line), the stress of temperature alone did not result in detectable effects. Only in summer,

567

when both stressors reached their maximum, did we observe a significant increase in the sensitivity of G. pulex to esfenvalerate. 5. Conclusions The common scenario of sequential exposure (Liess et al., 1999; Ashauer et al., 2016) to pesticides within one generation may affect the performance of freshwater macroinvertebrates. Moreover, moderate environmental stress well below the upper physiological limit may magnify the effect of pesticide exposure in the field context. Therefore, a realistic assessment of safe concentrations must consider the effects of sequential pesticide exposure within one generation and its interactions with additional environmental stressors. For the environmental scenario investigated here an additional assessment factor of 2.7 would apply. However, we expect that other scenarios representing additional stressor combinations may require different assessment factors. Author contributions ML conceived the research question. ML and RR developed the design of the investigation. RR performed the field work and laboratory tests. RR, JMB and ML analyzed and interpreted the results. The manuscript was written and approved by all authors. Funding sources This work was supported by Helmholtz long-range strategic research funding (POF III). The funding source was not involved in the study design, in the collection, analysis and interpretation of data, in the writing of the article; nor in the decision to submit the article for publication. Acknowledgment We thank Alena Lepilova for her support in the collection and identification of test animals. Additionally, we thank Helmholtz long-range strategic research funding (POF III) for financial support. Appendix A. Supplementary data A single file (PDF) of supporting information is provided. Further information regarding pesticide application in the year 2015, sampling sites, statistical analyses and datasets is included. Supplementary data associated with this article can be found in the online version, at http://dx.doi.org/10.1016/j.scitotenv.2017.08.073. References Abel, P.D., 1980. Toxicity of Gamma-hexachlorcyclohexane (Lindane) to Gammarus pulex; mortality in relation to concentration and duration of exposure. Freshw. Biol. 10 (3): 251–259. http://dx.doi.org/10.1111/j.1365-2427.1980.tb01200.x. Agatz, A., Ashauer, R., Brown, C.D., 2014. Imidacloprid perturbs feeding of Gammarus puilex at environmentally relevant concentrations. Environ. Toxicol. Chem. 33 (3), 648–653. Ashauer, R., Boxall, A.B.A., Brown, C.D., 2007. Modeling combined effects of pulsed exposure to Carbaryl and Chlorpyrifos on Gammarus Pulex. Environ. Sci. Technol. 41 (15): 5535–5541. http://dx.doi.org/10.1021/es070283w. Ashauer, R., O'Connor, I., Hintermeister, A., Escher, B.I., 2015. Death dilemma and organism recovery in ecotoxicology. Environ. Sci. Technol. 49 (16):10136–10146. http:// dx.doi.org/10.1021/acs.est.5b03079. Ashauer, R., Albert, C., Augustine, S., Cedergreen, N., Charles, S., Ducrot, V., Focks, A., Gabsi, F., Gergs, A., Goussen, B., Jager, T., Kramer, N.I., Nyman, A.-M., Poulsen, V., Reichenberger, S., Schäfer, R.B., Van den Brink, P.J., Veltman, K., Vogel, V., Zimmer, E.I., Preuss, T.G., 2016. Modelling survival: exposure pattern, species sensitivity and uncertainty. Sci. Rep. 6:29178. http://dx.doi.org/10.1038/srep29178. Ashauer, R., O'Connor, I., Escher, B.I., 2017. Toxic mixture in time – the sequence makes the poison. Environ. Sci. Technol. 51 (5):3084–3092. http://dx.doi.org/10.1021/ acs.est.6b06163. Barry, M.J., Logan, D.C., 1998. The use of temporary pond microcosms for aquatic toxicity testing: direct and indirect effects of endosulfan on community structure. Aquat. Toxicol. 41 (1–2), 101–124.

568

R. Russo et al. / Science of the Total Environment 610–611 (2018) 563–569

Bates, D., Maechler, M., Bolker, B., Walker, S., 2014. lme4: Linear Mixed-effects Models Using Eigen and S4. (R package version 1.1–7). http://CRAN.R-project.org/package=lme4. Becker, J.M., Liess, M., 2015. Biotic interactions govern genetic adaptation to toxicants. Proc. R. Soc. B Biol. Sci. 282. http://dx.doi.org/10.1098/rspb.2015.0071. Becker, J.M., Liess, M., 2017. Species diversity hinders adaptation to toxicants. Environ. Sci. Technol. http://dx.doi.org/10.1021/acs.est.7b02440. Beketov, M.A., Liess, M., 2005. Acute contamination with esfenvalerate and food limitation: chronic effects on the mayfly, Cloeon dipterum. Environ. Toxicol. Chem. 24 (5): 1281–1286. http://dx.doi.org/10.1897/04-256R1.1. Beketov, M.A., Liess, M., 2008. Acute and delayed effects of the neonicotinoid insecticide thiacloprid on seven freshwater arthropods. Environ. Toxicol. Chem. 27 (2): 461–470. http://dx.doi.org/10.1897/07-322R.1. Beketov, M.A., Foit, K., Schäfer, R.B., Schriever, C.A., Sacchi, A., Capri, E., Biggs, J., Wells, C., Liess, M., 2009. SPEAR indicates pesticide effects in streams – comparative use of species- and family-level biomonitoring data. Environ. Pollut. 157 (6), 1841–1848. Beketov, M.A., Kefford, B.J., Schäfer, R.B., Liess, M., 2013. Pesticides reduce regional biodiversity of stream invertebrates. PNAS 110 (27):11039–11043. http://dx.doi.org/ 10.1073/pnas.1305618110. Bowman, M.C., Oiler, W.L., Cairns, T., Gosnell, A.B., Oliver, K.H., 1981. Stressed bioassay systems for rapid screening of pesticide residues. Part I: Evaluation of bioassay systems. Arch. Environ. Contam. Toxicol. 10 (1), 9–24. Brosed, M., Lamothe, S., Chauvet, E., 2016. Litter breakdown for ecosystem integrity assessment also applies to streams affected by pesticides. Hydrobiologia 773 (1): 87–102. http://dx.doi.org/10.1007/s10750-016-2681-2. Cairns, J.J., Heath, G.A., Parker, C.B., 1975. Temperature influence on chemical toxicity to aquatic organisms. J. Water Pollut. Control Fed. 47 (2), 267–280. Clark, S.L., Ogle, R.S., Gantner, A., Hall, J.L.W., Mitchell, G., Giddings, J., McCoole, M., Dobbs, M., Henry, K., Valenti, T., 2015. Comparative sensitivity of field and laboratory populations of Hyalella azteca to the pyrethroid insecticides bifenthrin and cypermethrin. Environ. Toxicol. Chem. 34 (10):2250–2262. http://dx.doi.org/10.1002/etc.2907. Coors, A., De Meester, L., 2008. Synergistic, antagonistic and additive effects of multiple stressors: predation threat, parasitism and pesticide exposure in Daphnia magna. J. Appl. Ecol. 45 (6):1820–1828. http://dx.doi.org/10.1111/j.13652664.2008.01566.x. Dangles, O., Gessner, M.O., Guerold, F., Chauvet, E., 2004. Impacts of stream acidification on litter breakdown: implications for assessing ecosystem functioning. J. Appl. Ecol. 41:365–378. http://dx.doi.org/10.1111/j.0021-8901.2004.00888.x. De Rosario-Martinez, H., 2015. phia: Post-Hoc Interaction Analysis. (R package version 0.2–0). http://CRAN.R-project.org/package=phia. Enserink, L., Luttmer, W., Maas-Diepeveen, H., 1990. Reproductive strategy of Daphnia magna affects the sensitivity of its progeny in acute toxicity tests. Aquat. Toxicol. 17 (1), 15–25. European Food Safety Authority (EFSA), Parma, Italy, 2013. Guidance on tiered risk assessment for plant protection products for aquatic organisms in edge-of-field surface waters. EFSA J. 11 (7), 3290. Feng, Y.T., Wu, Q.J., Xu, B.Y., Wang, S.L., Chang, X.L., Xie, W., Zhang, Y.J., 2009. Fitness costs and morphological change of laboratory-selected thiamethoxam resistance in the Btype Bemisia tabaci (Hemiptera: Aleyrodidae). Jpn. J. Appl. Entomol. 133 (6):466–472. http://dx.doi.org/10.1111/j.1439-0418.2009.01383.x. Floyd, E., Geist, J.P., Werner, I., 2008. Acute, sublethal exposure to a pyrethroid insecticide alters behavior, growth, and predation risk in larvae of the fathead minnow (Pimephales promelas). Environ. Toxicol. Chem. 27 (8):1780–1787. http://dx.doi.org/ 10.1897/07-448.1. Fox, J., 2003. Effect displays in R for generalised linear models. J. Stat. Softw. 8 (15):1–27. http://dx.doi.org/10.18637/jss.v008.i15. Handy, R.D., 1994. Intermittent exposure to aquatic pollutants: assessment, toxicity and sublethal responses in fish and invertebrates. Comp. Biochem. Physiol. C 107 (2), 171–184. Harwood, A.D., You, J., Lydy, M.J., 2009. Temperature as a toxicity identification evaluation tool for pyrethroid insecticides: toxicokinetic confirmation. Environ. Toxicol. Chem. 28 (5):1051–1058. http://dx.doi.org/10.1897/08-291.1. Hill, I.R., 1985. Effects on nontarget organisms in terrestrial and aquatic environments. In: Leahey, J.P. (Ed.), The Pyrethroid Insecticides. Taylor and Francis, pp. 151–262. Huber, A., Bach, M., Frede, H.G., 2000. Pollution of surface waters with pesticides in Germany: modeling non-point source inputs. Agric. Ecosyst. Environ. 80 (3), 191–204. Hultin, L., 1971. Upstream movements of Gammarus pulex pulex (Amphipoda) in a South Swedish Stream. Oikos 22 (3):329–347. http://dx.doi.org/10.2307/3543856. ISO 10695:2000, 2000. Determination of Selected Organic Nitrogen and Phosphorus Compounds – Gas Chromatographic Methods. http://www.iso.org/iso/catalogue_ detail.htm?csnumber=18783. Jordaan, M.S., Reinecke, S.A., Reinecke, J.A., 2012. Acute and sublethal effects of sequential exposure to the pesticide azinphos-methyl on juvenile earthworms (Eisenia andrei). Ecotoxicology 2 (1):649–661. http://dx.doi.org/10.1007/s10646-011-0821-z. Khan, M.A.Q., Khan, M.A., 2008. Effect of temperature on waterflea Daphnia magna (Crustacea:Cladocera). Nature Precedings http://hdl.handle.net/10101/npre.2008.1909.1. Krog, J., 1954. The influence of seasonal environmental changes upon the metabolism, lethal temperature and rate of heart beat of Gammarus limnaeus (smith) taken from an Alaskan lake. Biol. Bull. 107 (3):397–410. http://dx.doi.org/10.2307/ 1538588. Liess, M., 2002. Population response to toxicants is altered by intraspecific interaction. Environ. Toxicol. Chem. 21 (1):138–142. http://dx.doi.org/10.1002/etc.5620210120. Liess, M., Von Der Ohe, P.C., 2005. Analyzing effects of pesticides on invertebrate communities in streams. Environ. Toxicol. Chem. 24 (4):954–965. http://dx.doi.org/10.1987/ 03-652.1.

Liess, M., Schulz, R., Liess, M.H.-D., Rother, B., Kreuzig, R., 1999. Determination of insecticide contamination in agricultural headwater streams. Water Res. 33 (1), 239–247. Liess, M., Foit, K., Becker, A., Hassold, E., Dolciotti, I., Kattwinkel, M., Duquesne, S., 2013. Culmination of low-dose pesticide effects. Environ. Sci. Technol. 47 (15): 8862–8868. http://dx.doi.org/10.1021/es401346d. Liess, M., Foit, K., Knillmann, S., Schäfer, R.B., Liess, H.D., 2016. Predicting the synergy of multiple stress effects. Sci. Rep. 6:32965. http://dx.doi.org/10.1038/srep32965. Lydy, M.J., Belden, J.B., Ternes, M.A., 1999. Effects of temperature on the toxicity of mparathion, chlorpyrifos, and pentachlorobenzene to Chironomus tentans. Arch. Environ. Contam. Toxicol. 37 (4), 542–547. Macan, T.T., Mackereth, J.C., 1957. Notes on Gammarus puilex in the English Lake District. Hydrobiologia 9 (1), 1–12. Minckley, W.L., 1964. Upstream movements of Gammarus (Amphipoda) in Doe Run, Meade County, Kentucky. Ecology 45 (1):195–197. http://dx.doi.org/10.2307/ 1937129. Münze, R., Orlinskiy, P., Gunold, R., Paschke, A., Kaske, O., Beketov, M.A., Hundt, M., Bauer, C., Schuurmann, G., Moder, M., Liess, M., 2015. Pesticide impact on aquatic invertebrates identified with Chemcatcher® passive samplers and the SPEAR (pesticides) index. Sci. Total Environ. 537:69–80. http://dx.doi.org/10.1016/ j.scitotenv.2015.07.012. Münze, R., Hannemann, C., Orlinskiy, P., Gunold, R., Paschke, A., Foit, K., Becker, J., Kaske, O., Paulsson, E., Peterson, M., Jernstedt, H., Kreuger, J., Schüürmann, G., Liess, M., 2017. Pesticides from wastewater treatment plant effluents affect invertebrate communities. Sci. Total Environ. 599-600:387–399. http://dx.doi.org/10.1016/ j.scitotenv.2017.03.008. Nyman, A.M., Hintermeister, A., Schirmer, K., Ashauer, R., 2013. The insecticide imidacloprid causes mortality of the freshwater amphipod Gammarus piulex by interfering with feeding behavior. PLoS One 8 (5). http://dx.doi.org/10.1371/ journal.pone.0062472. OECD, 1998. Test No. 211: Daphnia magna Reproduction Test. Guidelines for Testing of Chemicals. OECD Publishing, Paris. OECD, 2002. Test no. 423: acute oral toxicity - acute toxic class method. OECD Guidelines for Testing of Chemicals, Section 4. OECD Publishing, Paris. U.S. environmental protection agency, 1999. Office of pesticide programs. Determination of the appropriate FQPA safety factor(s) for use in the tolerance-setting process. https://www.epa.gov/pesticide-science-and-assessing-pesticide-risks/determination-appropriate-fqpa-safety-factors. Orlinskiy, P., Münze, R., Beketov, M., Gunold, R., Paschke, A., Knillmann, S., Liess, M., 2015a. Forested headwaters mitigate pesticide effects on macroinvertebrate communities in streams: mechanisms and quantification. Sci. Total Environ. 524–525:115–123. http://dx.doi.org/10.1016/jscitotenv.2015.03.143. Orlinskiy, P., Münze, R., Beketov, M., Gunold, R., Paschke, A., Knillmann, S., Liess, M., 2015b. Forested headwaters mitigate pesticide effects on macroinvertebrate communities in streams: mechanisms and quantification. Sci. Total Environ. 524–525: 115–123. http://dx.doi.org/10.1016/j.scitotenv.2015.03.143. Ostenauer, R., Köhler, H.R., 2008. Temperature-dependent effects of the pesticides thiacloprid and diazinon on the embryonic development of zebrafish (Danio rerio). Aquat. Toxicol. 86 (4):485–494. http://dx.doi.org/10.1016/j.aquatox.2007.12.013. Rasmussen, J.J., Nørum, U., Jerris, M.R., Wiberg-Larsen, P., Kristensen, E.A., Friberg, N., 2013. Pesticide impacts on predator-prey interactions across two levels of organisation. Aquat. Toxicol. 140-141, 340–345. Reinecke, S.A., Reinecke, J.A., 2005. The impact of organophosphate pesticides in orchards on earthworms in the Western Cape, South Africa. Ecotox. Environ. Saf. 66 (2), 244–251. Relyea, R.A., Mills, N., 2001. Predator-induced stress makes the pesticide carbaryl more deadly to gray treefrog tadpoles (Hyla versicolor). PNAS 98 (5):2491–2496. http:// dx.doi.org/10.1073/pnas.031076198. Ritz, C., Streibig, J.C., 2005. Bioassay analysis using R. J. Stat. Softw. 12 (5), 1–22. Schäfer, R.B., Von Der Ohe, P.C., Rasmussen, J., Kefford, B.J., Beketov, M., Schulz, R., Liess, M., 2012. Thresholds for the effects of pesticides on invertebrate communities and leaf breakdown in stream ecosystems. Environ. Sci. Technol. 46 (9):5134–5142. http://dx.doi.org/10.1021/es2039882. Schulz, R., Liess, M., 2001. Toxicity of aqueous-phase and suspended particle-associated fenvalerate: chronic effects after pulse-dosed exposure of Limnephilus lunatus (trichoptera). Environ. Toxicol. Chem. 20 (1), 185–190. Sutcliffe, D.W., Carrick, T.R., Willoughby, L.G., 1981. Effects of diet, body size, age and temperature on growth rates in the amphipod Gammarus pulex. Freshw. Biol. 11: 183–214. http://dx.doi.org/10.111/j.1365-2427.1981.tb01252.x. Tak, C.L.E., Karraker, N.E., Mei, Y.L.K., 2015. Temperature-dependent acute toxicity of methomyl pesticide on larvae of 3 asian amphibian species. Environ. Toxicol. Chem. 9999:1–6. http://dx.doi.org/10.1002/etc.3061. Valle, I.C., Buss, D.F., Baptista, D.F., 2013. The influence of connectivity in forest patches, and riparian vegetation width on stream macroinvertebrate fauna. Braz. J. Biol. 73 (2):231–238. http://dx.doi.org/10.1590/S1519-69842013000200002. Verberk, W.C.E.P., Durance, I., Vaughan, I.P., Ormerod, S.J., 2016. Field and laboratory studies reveal interacting effects of stream oxygenation and warming on aquatic ectotherms. Glob. Chang. Biol. 22:1769–1778. http://dx.doi.org/10.1111/gcb.13240. Vigneron, A., Geffard, O., Coquery, M., François, A., Quéau, H., Chaumot, A., 2015. Evolution of cadmium tolerance and associated costs in a Gammarus fossarum population inhabiting a low-level contaminated stream. Ecotoxicology 24 (6):1239–1249. http://dx.doi.org/10.1007/s10646-015-1491-z. Welton, J.S., 1979a. Life-history and production of the amphipod Gammarus pulex in a Dorset chalk stream. Freshw. Biol. 9, 263–275. Welton, J.S., 1979b. Life-history and production of the amphipod Gammarus pulex in a Dorset chalk stream. Freshw. Biol. 9:263–275. http://dx.doi.org/10.1111/j.13652427.1979.tb01508.x.

R. Russo et al. / Science of the Total Environment 610–611 (2018) 563–569 Weston, P.D., Poynton, H.C., Wellborn, G.A., Lydy, M.J., Blalock, B.J., Sepulveda, M.S., Colbourne, J.K., 2013. Multiple origins of pyrethroid insecticide resistance across the species complex of a nontarget aquatic crustacean, Hyalella Azteca. PNAS 110 (41): 16532–16537. http://dx.doi.org/10.1073/pnas.1302023110.

569

Willming, M.M., Qin, G., Maul, J.D., 2013. Effects of environmentally realistic daily temperature variation on pesticide toxicity to aquatic invertebrates. Environ. Toxicol. Chem. 32 (12):2738–2745. http://dx.doi.org/10.1002/etc.2354.