Science of the Total Environment 643 (2018) 1533–1541

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Identification of pesticide exposure-induced metabolic changes in

mosquito larvae
Renato Russo a,b, Sven-Bastiaan Haange c,d, Ulrike Rolle-Kampczyk c, Martin von Bergen c,
Jeremias Martin Becker a,b, Matthias Liess a,b,⁎
a
UFZ, Helmholtz Centre for Environmental Research, Department of System-Ecotoxicology, Permoserstraße 15, 04318 Leipzig, Germany
b
RWTH Aachen University, Institute for Environmental Research (Biology V), Worringerweg 1, 52074 Aachen, Germany
c
UFZ, Helmholtz Centre for Environmental Research, Department of Molecular System Biology, Permoserstraße 15, 04318 Leipzig, Germany
d
University of Leipzig, Institute of Biochemistry, Leipzig, 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

• Pesticides affect ecosystems at concen-

trations below regulatory thresholds.
• We observed metabolic changes at con-
centrations considered safe in risk as-
sessment.
• The insecticide clothianidin induced
energy-related metabolic changes in
mosquitoes.
• Effects were weaker but lasted longer
after exposure to very low concentra-
tions.
• Low concentrations may induce envi-
ronmental effects through increased en-
ergy demand.

a r t i c l e i n f o a b s t r a c t

Article history: The European regulatory framework for pesticides generally applies an assessment factor of up to 100 below the
Received 5 April 2018 acute median lethal concentration (LC50) in laboratory tests to predict the regulatory acceptable concentrations
Received in revised form 22 June 2018 (RACs). However, long-term detrimental effects of pesticides in the environment occur far below the RACs.
Accepted 22 June 2018
Here, we explored the metabolic changes induced by exposure to the neonicotinoid insecticide clothianidin in
Available online xxxx
larvae of the mosquito Culex pipiens. We exposed the test organisms to the insecticide for 24 h and then measured
Editor: Yolanda Picó the levels of 184 metabolites immediately and 48 h after the pulse contamination.
We established a link between the exposure to clothianidin and changes in the level of three specific classes of metab-
Keywords: olites involved in energy metabolism, namely, glycerophospholipids, acylcarnitines and biogenic amines. Remarkably,
Metabolic change exposure to concentrations considered to be safe according to the regulatory framework (2–4 orders of magnitude
Low-concentration exposure lower than the acute LC50), induced longer-term effects than exposure to the highest concentration. These results sug-
Freshwater macroinvertebrates gest that a specific detoxification mechanism was only triggered by the highest concentration.
Neonicotinoids We conclude that even very low insecticide concentrations increase the energy demands of exposed organisms,
which potentially translates into a decline in sensitive species in the field.
© 2018 Elsevier B.V. All rights reserved.

Abbreviations: LC50, Median lethal concentration; RACs, Regulatory acceptable concentrations; PFOS, Perfluorooctanesulfonic acid; LC, Liquid chromatography; FIA, Flow injection
analysis; MS, Mass spectrometry; MRM, Multi-reaction monitoring; ESI, Electrospray ionization; CCD, Colony collapse disorder; ATP, Adenosine triphosphate; RNA, Ribonucleic acid;
LD50, Median lethal dose; HBCD, Hexabromocyclododecane.
⁎ Corresponding author at: UFZ, Helmholtz Centre for Environmental Research, Department of System-Ecotoxicology, Permoserstraße 15, 04318 Leipzig, Germany.
E-mail address: matthias.liess@ufz.de (M. Liess).

https://doi.org/10.1016/j.scitotenv.2018.06.282
0048-9697/© 2018 Elsevier B.V. All rights reserved.
1534 R. Russo et al. / Science of the Total Environment 643 (2018) 1533–1541

1. Introduction 2008). Additionally, Martin-Park et al. (2017) explored the metabolic

profiles of contaminated mosquitoes for the first time and proposed
Within the European risk assessment framework, an assessment fac- that an increase in acylcarnitines is a potential tool for diagnosing pyre-
tor of up to 100 below the acute LC50 (the concentration that is lethal to throid insecticide exposure. The higher sensitivity of metabolomic tech-
50% of test organisms in laboratory standard tests) is commonly applied niques compared to traditional toxicity tests enables the detection of
to predict the regulatory acceptable concentrations (RACs) of pesticides effects at low and environmentally relevant concentrations (Tufi et al.,
in the natural environment (EFSA, 2013). However, the current frame- 2015). Furthermore, the identification of key metabolites as biomarkers
work is clearly not protective enough as several studies on non-target in freshwater invertebrates may contribute to the design of more effec-
freshwater macroinvertebrates have detected the detrimental effects tive toxicological tests without the use of either mammal systems,
of pesticides at concentrations considered to be safe (Beketov et al., which are too expensive and undesirable for ethical reasons (EU,
2013; Liess et al., 2013; Liess and Von Der Ohe, 2005; Münze et al., 2016), or cell culture systems, which do not present the required func-
2017; Nyman et al., 2013; Schäfer et al., 2012). Low pesticide concentra- tional complexity (Jones et al., 2012).
tions, such as RACs, might not have short-term lethal effects. Neverthe- Accordingly, in the present study, we explored the link between in-
less, such concentrations may result in delayed lethal effects (Beketov secticide exposure and metabolic changes in the larvae of a common
and Liess, 2008), interact synergistically with other stressors in the freshwater macroinvertebrate and relevant target of pest control, the
field (Janssens et al., 2017; Liess et al., 2016; Beketov and Liess, 2005; mosquito Culex pipiens. In the laboratory, we aimed to identify key met-
Coors and De Meester, 2008; Relyea and Mills, 2001) and show long- abolic profiles that reveal sub-lethal effects at low concentrations that
term cumulative effects in cases of repeated or chronic exposure potentially lead to delayed/synergistic effects in the field. The metabolic
(Alkassab and Kirchner, 2016; Liess et al., 2013; Russo et al., 2018). In changes in C. pipiens were investigated at high (10,000 ng/L, the same
2013, EU Commission Regulation 284 (EU, 2013a) indicated that resolv- order of magnitude as LC50), medium (100 ng/L) and very low
ing the uncertainties about the effects of pesticides at low concentra- (1 ng/L) concentrations of clothianidin, a neonicotinoid insecticide com-
tions is crucial for refining the risk assessment framework for monly applied in agriculture. The high concentration is of great regula-
pesticides. In this respect, understanding the biochemical processes tory relevance, being at the same order of magnitude as the LC50 while
that underlie pesticide toxicity, particularly at low concentrations, the medium concentration resembled environmentally realistic condi-
may increase the predictive power of toxicity tests. In fact, the identifi- tions (Knillmann et al., 2018). Finally, we applied the very low concen-
cation of responses in key metabolites for use as regulatory endpoints tration to test the experimental hypothesis that metabolic changes can
may allow a prompt detection of detrimental effects even at sub-lethal be detected even below the RACs. We investigated changes at two dif-
concentrations. The increased knowledge of toxicity mechanisms can ferent time points in order to explore the transient nature of metabolic
thus contribute to the design of cost- and time-effective analyses and, changes: the first, directly after the exposure, to investigate the changes
ultimately, to more realistic extrapolations of the laboratory data to induced by the toxicity of clothianidin; the second, approximately at
the field. half of the larval stage (Gerberg et al., 1969), to explore the combination
In recent years, due to the advancement in metabolomic techniques, of the effects of toxicity with the biological response of the larvae. It is
new high-throughput and multi-endpoint methods for analysing meta- pivotal to understand the transient nature of such metabolic changes
bolic changes have become available (Aardema and MacGregor, 2002). in order to ensure their detection and, eventually, apply them as regula-
These techniques have been increasingly applied to characterize the ef- tory endpoint in the future.
fects of toxicants on organisms (Mckelvie et al., 2009; Marx-Stoelting
et al., 2015; Tralau et al., 2015) and to identify potential biomarkers of 2. Materials and methods
exposure to specific compounds (Lin et al., 2006; Robertson, 2005).
The link between the changes in the levels of key natural metabolites 2.1. Study design
and the exposure to a specific toxicant is relatively well established for
humans. An increase in mercapturic acid, for instance, has been linked We exposed newborn larvae of the mosquito Culex pipiens pipiens to
to exposure to atrazine (Lucas et al., 1993), while an increase in 1- three different concentrations of clothianidin (1 ng/L, 100 ng/L and
hydroxypyrene has been linked to exposure to polycyclic aromatic hy- 10,000 ng/L) for 24 h. The metabolite contents were quantified directly
drocarbons (Jongeneelen, 2001). Recently, Zhang et al. (2014) proposed after the 24 h of exposure and 48 h after the end of the pulse exposure
using increases in serine and testosterone as biomarkers of arsenic ex- by LC-MS-MS.
posure. In rats, the fluctuations of various metabolites have been linked
to exposure to hydrazine (Nicholls et al., 2001) and γ radiation (Lanz 2.2. Study species
et al., 2009), and Wu et al. (2013) observed decreased levels of amino
acids and increased levels of betaine and fumarate in clams exposed to The mosquito Culex pipiens (Linnaeus, 1758 - Insecta: Diptera) is a
arsenic. potential vector of diseases such as West Nile fever, Rift Valley fever
Similarly, various potential biomarkers of toxicant exposure have and Elephantiasis (Chandre et al., 1998; Kumar and Pillai, 2011; Paul
been developed for earthworms in the past decades: an increase in his- et al., 2005; Rodriguez et al., 1993; Scott et al., 2015). The larvae of
tidine, for instance, has been proposed as a biomarker of exposure to this species are common in ephemeral ponds and are therefore fre-
copper (Gibb et al., 1997). More recently, fluctuations in the levels of quently exposed to various insecticides either through vector control
various metabolites have been identified as potential indicators of measures or through run-off from agricultural fields. We used the bio-
PFOS exposure (Lankadurai et al., 2013). Mckelvie et al. (2009) explored type molestus (subsequently referred to as simply C. pipiens), which
in the use of increasing concentrations of alanine as a potential bio- has adapted to subterranean habitats, because it is relatively easy to cul-
marker of organochlorine insecticide exposure. ture. Unlike other mosquitoes, C. pipiens can lay its first egg raft without
However, few studies to date have investigated the metabolic re- a blood meal (Clements, 1999) and mates in small cages without
sponses of freshwater invertebrates after exposure to various sub- swarming, which makes this species suitable for laboratory culture.
stances (Jones et al., 2012; Nagato et al., 2013) despite the relevance The mosquito culture was obtained from the Federal Environment
of this taxonomic group for the risk assessment of pesticides. Recent ev- Agency, UBA, Berlin, Germany, and was grown in a rearing chamber at
idence has suggested that alanine and lysine play key roles in the toxic- 25 °C with 60% humidity and a 16:8 light:dark regime. The egg rafts
ities of copper, arsenic and lithium in Daphnia (Nagato et al., 2013), and were collected in plastic trays containing tap water. The hatched larvae
vitellogenin has been identified as a biomarker of exposure to were cultured in tap water in white plastic trays that were constantly
oestrogenic compounds in aquatic environments (Matozzo et al., aerated using an aquarium pump. The larvae were fed with a 1:1
 R. Russo et al. / Science of the Total Environment 643 (2018) 1533–1541 1535

(weight ratio) mixture of ground dog biscuits (Hd-H biscuits, ssniff On day 2 (directly after the 24 h of exposure), the mortality was al-
Spezialdiäten GmbH, Soest, Germany) and stinging nettle powder ways below 5% at each concentration level (Fig. S2). While such low
(Folia urticae, Caesar & Loretz GmbH, Hilden, Germany) until pupation mortality was expected for the lowest and the medium concentrations,
(approximately 10 days). The pupae were placed into glass trays, trans- this seemed unexpected for the highest concentration. However, the
ferred into the cages and covered with a mesh to prevent adult mos- LC50 calculation referred to mortality measured after 48 h of exposure,
quito oviposition until emergence (24–72 h). The emerged while in the present study we exposed the mosquito larvae to
mosquitoes were released to mate inside the cages and fed with clothianidin only for 24 h. In fact, when comparing our results with
sponges soaked in glucose-, fructose- and honey-saturated solutions. the LC50 calculation (Table S9), the mortality after 24 h of exposure at
Female mosquitoes placed their egg rafts into open trays filled with 10,000 ng/L appeared comparable. Sixty surviving larvae from each of
water, which were subsequently transferred into plastic trays outside the four different contamination levels were frozen in 1.5-mL Eppendorf
the cages to restart the cycle. tubes at −80 °C for the metabolic analysis. The amount of larvae needed
to achieve enough biomass to allow metabolomics measurements were
2.3. Insecticide estimated based on calculations of average dry weight of larvae at dif-
ferent developmental stages (Table S10). The remaining larvae were
The test organisms were exposed to the neonicotinoid insecticide placed in 2-L glass vessels filled with 1 L of fresh (non-contaminated)
clothianidin. Neonicotinoid insecticides were commercialized in 1991 medium and fed with 1.2 mg/larvae of a mixture of ground dog biscuits
and have dominated the global agricultural insecticide market since (Hd-H biscuits, ssniff Spezialdiäten GmbH, Soest, Germany) and sting-
2000 (Simon-Delso et al., 2015). Clothianidin [(E)-1-(2-chloro-1,3- ing nettle powder (Folia urticae, Caesar & Loretz GmbH, Hilden,
thiazol-5-ylmethyl)-3-methyl-2-nitroguanidine] is one of the latest Germany) at a final weight ratio of 1:1 and a final concentration of
members of neonicotinoid insecticides. This compound was invented 12 mg/mL.
and developed by Sumitomo Chemical Takeda Agro Company, Ltd. On day 4 (48 h after the insecticide exposure), 20 larvae per condi-
and has been effectively used against Diptera, Coleoptera and Lepidop- tion were frozen in 1.5-mL Eppendorf tubes at −80 °C for the metabolic
tera pests. The chemical structure of clothianidin is characterized by the analysis. We repeated the experiment nine times over a span of five
presence of a thiazolyl ring. Clothianidin and other neonicotinoids are of months; however, due to measurement issues, we could only consider
increasing concern (Commission Regulation (EU) No. 485, EU, 2013b) seven of the nine replicates for statistical analysis. All the analytical
because evidence suggests that these compounds impact non- measurements were performed together at the end of the last experi-
targeted, beneficial insects (Mullin et al., 2005; Pecenka and Lundgren, mental replication.
2015), especially honeybees (Brandt et al., 2016; Di Prisco et al., 2013;
Jeschke and Nauen, 2008; Laurino et al., 2011). Due to its chemical 2.5. Sample preparation for LC-MS-MS measurement
and physical properties, clothianidin has the potential to reach surface
waters (APVMA, 2007) and has often been detected in agricultural The sample tubes were centrifuged for 15 min (15,000 rpm at 5 °C),
streams (Whiting et al., 2014; Münze et al., 2017) at concentrations of and up to 500 μL of medium was removed. Subsequently, the samples
up to 1000 ng/L (Knillmann et al., 2018). were (i) dried in a vacuum centrifuge (VAQ at 45 °C for 4 h), (ii) diluted
Clothianidin was obtained as granulated powder (weight ratio 1:1) in 1 mL of a mix of methanol/acetonitrile (1:1), (iii) treated with ultra-
from DANTOP® (Spiess-Urania Chemical GmbH, Germany) and pre- sound under cooling with ice for 1 h, and (iv) centrifuged again
pared by diluting 0.125 g of the powder in 0.5 L of deionized water (to (15,000 rpm at 5 °C). Prior to the analysis, the supernatant was dried
a final concentration of 0.125 g clothianidin/L) and then stirring the so- again in a nitrogen evaporator (1 h at 4 bar) and then placed in 100 μL
lution for 12 h. During this time, the solution was protected against of methanol.
photodegradation by covering it with aluminium foil. The three test so-
lutions were obtained from serial dilutions prepared by mixing the 2.6. Metabolomic measurement by MS-MS
stock solution with appropriate quantities of deionized water (for
more details on the preparation of test solutions, see Table S7). Since With metabolomics having only been recently applied to aquatic in-
the lowest concentration of clothianidin applied in our study was out vertebrates, information on the links between pesticide exposure and
of the detection range of most standard methods for this substance, related changes in metabolite profiles is sparse. As such, a sound and re-
we performed analytical measurements only for (i) the last dilution producible extraction method has not been developed. Therefore, we
step before obtaining the test concentrations and (ii) the medium test used a commonly applied metabolomics kit (Absolute IDQ p180,
concentration. On average, the measured concentrations deviated Biocrates LIFE Science AG, Innsbruck, Austria). The kit has been designed
from the nominal concentrations by ±3.5% (Table S8). Therefore, all re- to cover metabolic pathways that are known to be of central biological
sults reported in the subsequent sections refer to the nominal relevance – e.g., acylcarnitines are involved in mitochondrial function
concentrations. (Imam et al., 2018; Kenéz et al., 2016; Moser et al., 2015), phosphatidyl-
In preliminary tests, the LC50 for C. pipiens was 14.85 μg/L (Fig. S1), cholines influence lipoprotein structure (Jackson et al., 1976), and bio-
which is comparable with data in the literature on the toxicity of genic amines are involved in cell cycle controls (Heby, 1981) and
neonicotinoid to aquatic macroinvertebrates (PPDB website; Beketov detoxification (Frieling et al., 2012). The kit identifies and quantifies
and Liess, 2008; Mo et al., 2002). 184 metabolites from 6 compound classes (Table 1) and provides highly
robust, quantitative metabolomics data with excellent accuracy and
2.4. Acute toxicity tests precision. Its outstanding inter-instrument, inter-laboratory, as well as
long-term reproducibility has been proven by an independent, interna-
On day 1, up to 24-h-old C. pipiens larvae were placed in 2-L glass tional ring trial (Siskos et al., 2017).
vessels filled with 1 L of M4 Elendt medium (OECD guideline No. 211, The analyses were carried out as previously described (Kenéz et al.,
1998, annex 2) for the exposure to clothianidin for 24 h. Test tempera- 2016; Oberbach et al., 2011). In brief, a flow injection tandem mass
ture was 18 ± 1 °C. Each vessel contained 20 mosquito larvae. We ap- spectrometry method (FIA-MS/MS) was employed for the unipolar me-
plied four contamination levels: control group (non-exposed), 1 ng/L tabolites, and an LC-MS/MS method was employed for the polar metab-
(4 orders of magnitude below the LC50), 100 ng/L (2 orders of magni- olites. The assay required only very small amounts of substrate (10 μL).
tude below the LC50) and 10,000 ng/L (the same order of magnitude The quantification of analytes utilized stable isotope-labelled or chemi-
as the LC50). Five glass vessels (up to a total of 100 larvae) were pre- cally homologous internal standards. Quality controls (QCs) were in-
pared for each concentration level. cluded for three different concentration levels. For calibration, the kit
1536 R. Russo et al. / Science of the Total Environment 643 (2018) 1533–1541

contained a calibrator mix with seven different concentrations. The parametric tests. In the plots, we report the effects of pesticide exposure
measurements were carried out with electrospray ionization (ESI) in as the metabolite concentration of the exposure group relative to that of
multi-reaction monitoring (MRM) mode to ensure high specificity and the non-exposed control group.
sensitivity. For the FIA measurements, two injections were needed. In
total, 158 MRM pairs were measured in positive ion mode (including 3. Results
13 internal standards (ISs)), and two MRM pairs were measured in neg-
ative ionization mode (including one IS). In the LC-MS/MS method, 65 3.1. Average metabolite content
MRM pairs were measured only in the positive mode (including 25
Iss). Therefore, only one injection was necessary for the LC part. The We found that all the applied clothianidin concentrations had signif-
kit needed the following additional chemicals: water, Millipore; ITC, icant effects on the average content across all metabolites (Fig. 1). On
Fluka (derivatisation of amino acids); pyridine, Fluka (p.a.); methanol, day 2, directly after the 24-h contamination period (Fig. 1a), exposure
Merck; LiChrosolv for LC/MS; Acetonitril, Merck; LiChrosolv for LC/MS, to 10,000 ng/L resulted in a 27% reduction in the overall metabolite con-
Fluka; and formic acid, Fluka. tent (t = −9.4; df = 2392 following Satterthwaite approximation; p b
The kit has been used for several biofluids and tissue samples in var- 0.001), while exposure to 100 ng/L resulted in only a 13% reduction (t
ious species, ranging from human to yeast (Ferrario et al., 2016; = −4.2; df = 2392; p b 0.001), and exposure to 1 ng/L resulted in a
Papathanassiu et al., 2017; Rußmayer et al., 2015), but to our knowl- 17% reduction (t = −5.6; df = 2393; p b 0.001). Thus, the highest pes-
edge, using this kit for mosquito larvae is new. ticide concentration resulted in stronger effects than the medium con-
centration (χ2 = 28.1; df = 1; p b 0.001) or the lowest concentration
2.7. Data analysis (χ2 = 14.8; df = 1; p b 0.001). In contrast, on day 4 (48 h after exposure,
Fig. 1b), the highest pesticide concentration did not result in significant
Statistical analyses were conducted with RStudio for Mac (version changes in the overall metabolite content, while the medium and low-
1.0.153). The dataset used consisted of 5152 (184 × 4 × 7 × 2) observa- est concentrations resulted in reductions of 16% (t = −4.6; df =
tions resulting from 184 metabolites that were measured at four differ- 2411; p b 0.001) and 12% (t = −2.9; df = 2402; p = 0.005),
ent contamination levels over seven experimental replications on day 2 respectively.
and day 4 (Tables S1–S6).
To investigate which pesticide concentrations affected the 3.2. Metabolite classes
(i) metabolite content as a whole and (ii) the contents of each metabo-
lite class on day 2 and day 4, we constructed linear mixed-effects Directly after pesticide exposure (day 2), we observed significant
models with the lme4 (Bates et al., 2014; version 1.1–7) and lmerTest changes in the average metabolite contents within three (out of six)
packages (Kuznetsova et al., 2017). To explore the effects of the pesti- metabolic classes, namely, biogenic amines, glycerophospholipids and
cide on the metabolite content, we included only the pesticide concen- acylcarnitines (Fig. 2a). In particular, on day 2, exposure to
trations as a fixed factor; the effect of individual metabolites, nested 10,000 ng/L reduced the contents of biogenic amines (t = −3.0; df =
within the classes, and the effect of experimental replications were in- 240; p = 0.006), glycerophospholipids (t = −14.0; df = 1145; p b
cluded as random intercepts. When investigating the effects of the pes- 0.001), and acylcarnitines (t = −6.9; df = 513; p b 0.001). Also expo-
ticide on the average metabolite content within each metabolite class, sure to 100 ng/L marginally reduced the biogenic amine content (t =
we applied similar models as above, with the only difference being −2.0; df = 239; p = 0.084) and significantly reduced the
that the metabolite classes were not included in the random intercept glycerophospholipid content (t = −6.5; df = 1146; p b 0.001) but in-
because a separate model was constructed for each class. To compare creased the acylcarnitine content (t = 3.3; df = 513; p = 0.003). Simi-
the effects of different pesticide concentrations in a post hoc analysis, larly, exposure to 1 ng/L significantly reduced the biogenic amine
we specified custom contrasts between the pesticide concentrations content (t = −2.5; df = 240; p = 0.038) and glycerophospholipid con-
that matched our specific hypotheses using the phia package (De tent (t = −7.0; df = 1145; p b 0.001) but marginally increased the
Rosario-Martinez, 2015; version 0.2–0). Finally, to explore the effects acylcarnitine content (t = 2.0; df = 513; p = 0.089). The highest pesti-
of the pesticide on each individual metabolite, we compared the con- cide concentration decreased the glycerophospholipid content more
centrations of each metabolite in pesticide-treated larvae and in the strongly than the medium (χ2 = 54.3; df = 1; p b 0.001) and the lowest
control group using paired t-tests with the seven experimental replica- concentrations (χ2 = 23.1; df = 1; p b 0.001). However, the decrease in
tions included as the grouping factor. the biogenic amine content did not vary significantly between the dif-
To avoid type I error inflation by multiple comparisons, all p values ferent pesticide concentrations. The decrease in acylcarnitines at the
were adjusted to the number of classes, the number of metabolites highest pesticide concentration differed significantly from the increase
within a given class, or the number of custom contrasts used in a post in acylcarnitines at the lowest (χ2 = 62.51; df = 1; p b 0.001) and me-
hoc test by applying the fdr correction. The metabolite concentrations dium (χ2 = 85.31, df = 1, p b 0.001) concentrations; however, the
were log transformed prior to the analyses to improve homoscedasticity changes in the acylcarnitine content were generally small (b10%) com-
and to normalize the distribution of the residuals, as is required by pared to those of the other metabolite classes.
However, two days after pesticide exposure (day 4), only the
glycerophospholipid and acylcarnitine contents were affected, and this
Table 1 result was observed only after exposure to the low and medium pesti-
List of compounds measured by mass spectrometry. The analysis was carried out by a flow
injection tandem mass spectrometry method (FIA-MS/MS) for the unipolar metabolites,
cide concentrations (Fig. 2b). Decreased glycerophospholipid contents
and an LC-MS/MS method was used for the polar metabolites. were observed after exposure to both 100 ng/L (t = −9.4; df = 1163;
p b 0.001) and 1 ng/L (t = −5.3; df = 1163; p b 0.001) of clothianidin,
Compound class Subclasses No. of
and the effect was stronger at 100 ng/L (χ2 = 11.55; df = 1, p b 0.001).
compounds
Notably, the glycerophospholipid content tended to increase (even
Acylcarnitines 40
though not significantly) after the exposure to 10,000 ng/L. Similarly,
Amino acids Proteinogenic, citrulline, ornithine 19
Glycerophospholipids Phosphatidylcholines, 76, 14 the content of acylcarnitines was reduced after exposure to both
lysophosphatidylcholines 100 ng/L (t = −4.9; df = 513; p b 0.001) and 1 ng/L (t = −6.6; df =
Sphingolipids Sphingomyelines 15 513; p b 0.001) clothianidin, while tended to increase after the exposure
Biogenic amines 19 to 10,000 ng/L. The reduction was marginally stronger at 1 ng/L (χ2 =
Hexoses 1
3.17; df = 1, p = 0.075). A detailed list of all measured factors of change
 R. Russo et al. / Science of the Total Environment 643 (2018) 1533–1541 1537

Fig. 1. Percent change of average metabolites levels at different clothianidin concentrations. The relative overall metabolite content (percent of the metabolite content of the non-exposed
control group) after exposure to 1 ng/L, 100 ng/L and 10,000 ng/L clothianidin for 24 h measured (a) directly after exposure and (b) 48 h hours later. The analysis is based on a linear mixed-
effects model with 5152 observations. The relative changes are shown with their associated 95% confidence intervals. The asterisks below the bars report significant deviations from the
control group, while significant differences between doses of exposure are displayed on the top, where “⁎⁎” indicates p b 0.01 and “⁎⁎⁎” indicates p b 0.001.

Fig. 2. Percent change of metabolite classes level at different clothianidin concentrations. The relative changes in the contents of the different metabolite classes (percent of the metabolite
content of the non-exposed control group) after exposure to 1 ng/L, 100 ng/L and 10,000 ng/L clothianidin for 24 h measured (a) directly after exposure and (b) 48 h later. The analysis is
based on linear mixed-effects models with 5152 observations. Only classes with significant changes are reported. The relative changes are shown with their associated 95% confidence
intervals. The asterisks below the bars report significant deviations from the control group, while significant differences between the exposure doses are displayed on the top, where
“.” denotes 0.05 b p ≤ 0.1; “⁎” denotes p ≤ 0.05; and “⁎⁎” denotes p b 0.01 and “⁎⁎⁎” p b 0.001.
1538 R. Russo et al. / Science of the Total Environment 643 (2018) 1533–1541

for all metabolite classes at each of the three pesticide concentrations on and terrestrial invertebrates (Maute et al., 2017). Remarkably, we
day 2 and day 4 is provided in the supplementary material (Table S11). could detect metabolic changes even at a concentration 4 orders of mag-
nitude lower than the acute LC50, which did not cause additional mor-
3.3. Single metabolites tality in exposed larvae as compared to the control group (Fig. S2). This
result highlights the urgency of improving the understanding of the
After adjusting the p values for the numerous comparisons per- mechanisms that underlie the toxicity of contaminants at very low
formed, we revealed significant changes in only three of the 184 metab- doses that are considered safe according to governmental risk assess-
olites measured. These three metabolites belonged to the same classes ments (EFSA, 2013).
identified above, namely, glycerophospholipids, acylcarnitines, and bio-
genic amines. On day 2 (directly after the 24-h exposure to 4.2. Effects on different metabolite classes
clothianidin), we observed a 26% decrease (t = −8.2; df = 6; p =
0.002) of the glycerophospholipid phosphatidylcholine diacyl C26:0 Three different classes of metabolites, namely, glycerophospholipids,
(PC aa C26:0) after exposure to clothianidin at 1 ng/L (Fig. 3a). In con- acylcarnitines and biogenic amines, responded significantly to insecticide
trast, on day 4, we observed a marginally significant increase of the exposure.
two biogenic amines histamine (8%; t = 3.8; df = 6; p = 0.086) and tau- Glycerophospholipids decreased immediately after exposure. Both
rine (16%; t = 4.0; df = 6; p = 0.086) after exposure to clothianidin at down- and up-regulation of glycerophospholipids in rats have been pre-
10,000 ng/L (Fig. 3b). A detailed list of all measured metabolite contents viously linked to exposure to high doses (the same order of magnitude
at each of the three pesticide concentrations on day 2 and day 4 is pro- as the acute LD50 – PPDB website) of the phenylpyrazole insecticide
vided in the supplementary information (Table S1–S6). fipronil (Moser et al., 2015). In mosquitoes, glycerophospholipids are
known to be a major component of cellular membranes (Ecker and
4. Discussion Liebisch, 2014). Glycerophospholipids also act as modulators of energy
metabolism, the mitochondrial electron transport system and of signal-
4.1. Effects of clothianidin exposure on the overall metabolite content ling pathways (Atella and Shahabuddin, 2002; Soares et al., 2015) in-
volved in growth, survival and proliferation (Kerr and Colucci, 2011).
High pesticide doses exhibited strong but short-term effects on gen- Both fipronil and clothianidin target the central nervous system of in-
eral metabolite contents, while much lower doses resulted in weaker sects through the hyperexcitation of nerves and muscles (Simon-Delso
but longer-term effects. Pulse exposure to low concentrations of pesti- et al., 2015). Therefore, the observed reduction in glycerophospholipids
cides is already known to have sub-lethal but long-term effects on fresh- in mosquito larvae may result from an insecticide-induced disturbance
water (Liess et al., 2013; Beketov et al., 2008; Wieczorek et al., 2018) of neural cell membranes. However, we observed that the highest

Fig. 3. Percent change of single metabolites level at different clothianidin concentrations. The relative changes in the contents of the individual metabolites (percent of the metabolite
content of the non-exposed control group) after exposure to 1 ng/L, 100 ng/L and 10,000 ng/L clothianidin for 24 h measured (a) directly after exposure and (b) 48 h hours later. This
analysis is based on paired t-tests that have been adjusted for multiple comparisons with 5152 observations. Only the contents of those metabolites showing significant changes were
reported for the three different concentration treatments. The relative changes are shown with their associated 95% confidence intervals. The asterisks below the bars report significant
deviations from the control group, where “.” denotes 0.05 b p ≤ 0.1 and “⁎” denotes p ≤ 0.05. The baseline concentrations in the non-exposed control group were as follows: 1.77 μM (phos-
phatidylcholine diacyl C26:0 on day 2), 0.62 μM (histamine on day 4), and 2.37 μM (taurine on day 4).
 R. Russo et al. / Science of the Total Environment 643 (2018) 1533–1541 1539

pesticide concentration only reduced the glycerophospholipid content pesticides (Williamson and Wright, 2013) such as neonicotinoids. In
immediately after exposure, while the smaller decrease caused by particular, more insights should be gained to explore the transient na-
lower pesticide concentrations lasted for at least 48 h. We have ture of changes measured in the present investigation.
interpreted our outcomes as follows: given that the observed changes
in metabolite levels stemmed from a combination of both (i) the toxicity 4.3. Effects on individual metabolites
mechanisms of clothianidin and (ii) the biological response of the larvae
after exposure, we can assume that different doses of the insecticide ac- Finally, we explored the possibility of identifying individual metabo-
tivated different defence mechanisms, resulting in different levels of lites as potential indicators of neonicotinoid exposure. Our outcomes
metabolites. For example, Shi et al. (2018) observed different dose- suggest that a few individual metabolites responded more strongly to
dependent responses of earthworms exposed to various concentrations the pesticide exposure than most other metabolites from the same clas-
of HBCD, a brominated flame retardant. The response to low doses in ses. We identified a significant reduction in the glycerophospholipid
earthworms mainly involved the breakdown of proteins for energy pro- phosphatidylcholine diacyl C26:0 directly after exposure as well as a
duction; in fact, greatly increased amino acid and adenosine triphos- marginally significant increase in the two biogenic amines histamine
phate [ATP] levels were detected. In contrast, exposure to high doses and taurine 48 h later.
also enhanced the expression of genes for detoxifying enzymes; in addi- The increase in histamine and taurine two days after insecticide ex-
tion to a mild increase in amino acid and ATP levels, a large increase in posure was unexpected since biogenic amines did not show significant
RNA coding for detoxifying enzymes was detected. Similarly, in our changes at the same tested time when analysed as whole class. How-
study, we can hypothesize that after the Culex pipiens larvae were ex- ever, over the 441 levels (21 biogenic amines ∗ 7 experimental
posed to low insecticide concentrations (≤100 ng/L), only basic com- repetitions ∗ 3 clothianidin concentrations) measured two days after in-
pensatory mechanisms (increased energy production) were activated, secticide exposure, 57% were above, and 43% below, the control level.
which resulted in a long-lasting decrease in the overall metabolite con- Therefore, it is likely that the two single metabolites alone showed sig-
tent and particularly in glycerophospholipid levels. However, more spe- nificant increase that was masked when the whole group of 21 biogenic
cific defence mechanisms – including detoxification and the restoration amines was analysed. Such increase may result from the roles of these
of membranes through the modulation of gene expression and the up- metabolites as signal modulators in the activation of detoxification
regulation (even though not significant) of the glycerophospholipid mechanisms. However, information in the literature on the biological
production - may have been involved in the response to the high insec- roles of these metabolites in aquatic invertebrates is scarce. Therefore,
ticide concentration, resulting in shorter-term effects. we can suggest that phosphatidylcholine diacyl C26:0, taurine and his-
The acylcarnitine levels initially increased after insecticide exposure tamine are only putative biomarkers of neonicotinoid exposure. Further
at the lowest and the medium concentrations but decreased at the studies should investigate the consistency of changes in metabolic pro-
highest insecticide concentration. In contrast, 48 h later, the files across different contaminants, doses, species, and time intervals in
acylcarnitine levels decreased following exposure to the lowest and order to (i) clarify the role of individual metabolites in the toxicity of
the medium concentration. Recently, Martin-Park et al. (2017) associ- neonicotinoids and the subsequent biological response and to (ii) en-
ated exposure to a pyrethroid insecticide with a decrease in sure that the identified biomarkers can be successfully applied.
acylcarnitine levels and indicated that this result was due to the in-
creased energy demand for the production of detoxifying enzymes.
5. Conclusions
Acylcarnitines represent a transport stage of fatty acids before beta-
oxidation for energy production in mitochondria; therefore,
Taken together, our outcomes showed that very low concentrations
acylcarnitines are well-established biomarkers for the early diagnosis
of the neonicotinoid insecticide clothianidin, which did not induce mor-
of metabolic disorders in mammals (Rodrıguez-Sanchez et al., 2015;
tality in the mosquito Culex pipiens, showed metabolic effects that indi-
Roschinger et al., 2000). However, recent investigations of mammalian
cated an increased energy demand. Even though no impact is expected
brains have also suggested that acylcarnitines play a role in modulating
at such low concentrations according to the European regulatory risk
neurotransmission through the cerebral synthesis of acetylcholine
assessment, this increased energy demand can potentially translate
(Jones et al., 2010; Lemhonwah et al., 2008) and in altering and stabiliz-
into reduced fitness and may explain changes in the freshwater macro-
ing the membrane composition (Jones et al., 2010). These additional
invertebrate community under field conditions. We suggest
functions may explain the short-term increase in acylcarnitines after ex-
acylcarnitines, glyerophospholipids and biogenic amines as potential
posure to low levels of neonicotinoid insecticides, which target the nic-
regulatory endpoints to detect effects even at low sublethal concentra-
otinic acetylcholine receptors located in the cell membranes of insect
tions. However, most metabolic changes lasted for 48 h only after expo-
nervous systems (Sparks and Nauen, 2015). At a later stage, or after ex-
sure to low insecticide concentrations. We conclude that the transient
posure to higher insecticide concentrations that may induce specific de-
changes in the metabolic profile reflect dose-dependent biological re-
toxification and restoration pathways, the decrease in acylcarnitines
sponses. Very low insecticide concentrations may disturb metabolism
suggests an increased energy demand.
in a non-specific way that increases energy demand for an extended
Toxicant-induced changes in the contents of biogenic amines have
time, while higher concentrations may trigger a specific detoxification
never been investigated in insects. However, it is known that biogenic
response that results in stronger but short-lived metabolic changes.
amines act as neurotransmitters, neuromodulators and neurohormones
(Evans, 1980). Additionally, these compounds are involved in various
functions such as reproduction (Gruntenko and Rauschenbach, 2008), Author contributions
responsiveness to olfactory stimuli (Mercer and Menzel, 1982) and ori-
entation (Casagrand and Ritzmann, 1992). Clothianidin has raised many ML, RR and MvB conceived the research question. URK and RR devel-
concerns due to its potential effects on colony collapse disorder (CCD - oped the design of the investigation. RR performed the laboratory tests.
Di Prisco et al., 2013; Brandt et al., 2016; Laurino et al., 2011) of honey- RR, JMB, ML and SBH analysed and interpreted the results. The manu-
bees. One of the primary symptoms of CCD is the failure of the bees to script was written by RR, with contributions of ML and JMB.
return to the hive. Thus, the neonicotinoid-induced reduction of bio-
genic amines, as measured in the present study, may play a relevant Funding sources
role in the impairment of olfactory functions and the orientation of ex-
posed organisms. The reduction of biogenic amines should be investi- Helmholtz long-range strategic research funding (POF III). The
gated further as a potential indicator of exposure to cholinergic funding source was not involved in the study design, in the collection,
1540 R. Russo et al. / Science of the Total Environment 643 (2018) 1533–1541

analysis and interpretation of data, in the writing of the article; nor in EU, 2016. European commission scientific conference: "non-animal approaches - the way
forward". http://ec.europa.eu/environment/chemicals/lab_animals/events_en.htm
the decision to submit the article for publication. (Accessed on February 2018).
European Food Safety Authority (EFSA), Parma, Italy, 2013. Guidance on tiered risk as-
Acknowledgment sessment for plant protection products for aquatic organisms in edge-of-field surface
waters. EFSA J. 11 (7), 3290.
Evans, P., 1980. Biogenic amines in the insect nervous system. Adv. Ins. Physiol. 15,
We thank the Helmholtz long-range strategic research funding (POF 317–473.
III) for the financial support. Ferrario, M., Cambiaghi, A., Brunelli, L., Giordano, S., Caironi, P., Guatteri, L., Raimondi, F.,
Gattinoni, L., Latini, R., Masson, S., Ristangno, G., Pastorelli, R., 2016. Mortality predic-
tion in patients with severe septic shock: a pilot study using a target metabolomics
Appendix A. Supplementary data approach. Sci. Rep. 6. https://doi.org/10.1038/srep20391.
Frieling, H., Leitmeier, V., Haschemi-Nassab, M., Kornhuber, J., Rhein, M., Bleich, S.,
Hillemacher, T., 2012. Reduced plasma levels of asymmetric di-methylarginine
Two files of supplementary material are provided: the file named
(ADMA) in patients with alcohol dependence normalize during withdrawal. Eur.
“Data” includes information on the level measured in each sample for Neuropsychopharmacol. 22 (11):836–840. https://doi.org/10.1016/j.
every metabolite, divided into classes (Tables S1 – S6); the file name euroneuro.2012.03.010.
Gerberg, E.J., Hopkins, T.M., Gentry, J.W., 1969. Mass rearing of Culex pipiens L. Mosq.
“Supplementary info” include data on mortality during the toxicity
News 29 (3), 382–385.
tests, LC50 calculations, preparation of test solutions, analytical mea- Gibb, J.O.T., Svendsen, C., Weeks, J.M., Nicholson, J.K., 1997. 1H NMR spectroscopic inves-
surements of clothianidin and details on dry weight of mosquitoes lar- tigations of tissue metabolite biomarker response to Cu II exposure in terrestrial in-
vae and significant metabolic changes measured. Supplementary data vertebrates: identification of free histidine as a novel biomarker of exposure to
copper in earthworms. Biomarkers 2 (5):295–302. https://doi.org/10.1080/
to this article can be found online at https://doi.org/10.1016/j. 135475097231526.
scitotenv.2018.06.282. Gruntenko, N.E., Rauschenbach, I.Y., 2008. Interplay of JH, 20E and biogenic amines under
normal and stress conditions and its effect on reproduction. J. Insect Physiol. 54 (6):
902–908. https://doi.org/10.1016/j.jinsphys.2008.04.004.
References Heby, O., 1981. Role of polyamines in the control of cell proliferation and differentiation.
Differentiation 19 (3):1–20. https://doi.org/10.1111/j.1432-0436.1981.tb01123.x.
Aardema, M.J., MacGregor, J.T., 2002. Toxicology and genetic toxicology in the new era of Imam, S.Z., He, Z., Cuevas, E., Rosas-Hernandez, H., Lantz, S.M., Sarkar, S., Raymick, J.,
“toxicogenomics”: imoact of “-omics” technologies. Mutat. Res. 499, 13–25. Robinson, B., Hanig, J.P., Herr, D., MacMillan, D., Smith, A., Liachenko, S., Ferguson,
Alkassab, A.T., Kirchner, W.H., 2016. Impacts of chronic sublethal exposure to clothianidin S., O'Callaghan, J., Miller, D., Somps, C., Pardo, I.D., Slikker, W.J., Pierson, B.J., Roberts,
on winter honeybees. Ecotoxicology 25, 1000–1010. R., Gong, B., Tong, W., Aschner, M., Kallman, J.M., Calligaro, D., Paule, M.G., 2018.
APVMA, Australian Pesticides and Veterinary Medicines Authority, 2007. Evaluation of Changes in the metabolome and microRNA levels in biological fluids might represent
the New Active CLOTHIANIDIN in the Products Sumitomo Shield Systemic Insecticide, biomarkers of neurotoxicity: a trimethyltin study. Exp. Biol. Med. 243 (3):228–236.
Sumitomo Samurai Systemic Insecticide, Sumitomo Stealth Systemic Insecticide https://doi.org/10.1177/1535370217739859.
(Canberra, Australia). Jackson, R.L., Morrisett, J.D., Gotto, J.A.M., 1976. Lipoprotein structure and metabolism.
Atella, G.C., Shahabuddin, M., 2002. Differential partitioning of maternal fatty acid and Physiol. Rev. 56 (2).
phospholipid in neonate mosquito larvae. J. Exp. Biol. 205, 3623–3630. Janssens, L., Tüzün, N., Stoks, R., 2017. Testing the time-scale dependence of delayed inter-
Bates, D., Maechler, M., Bolker, B., Walker, S., 2014. lme4: linear mixed-effects models actions: a heat wave during the egg stage shapes how a pesticide interacts with a suc-
using Eigen and S4. (R package version 1.1-7. http://CRAN.R-project.org/package= cessive heat wave in the larval stage. Environ. Pollut. 230:351–359. https://doi.org/
lme4. 10.1016/j.envpol.2017.06.082.
Beketov, M.A., Liess, M., 2005. Acute contamination with esfenvalerate and food limita- Jeschke, P., Nauen, R., 2008. Neonicotinoids—from zero to hero in insecticide chemistry.
tion: chronic effects on the mayfly, Cleon dipterum. Environ. Toxicol. Chem. 24 (5): Pest Manag. Sci. 6 (4), 1084–1098.
1281–1286. https://doi.org/10.1897/04-256R1.1. Jones, L.L., McDonald, D.A., Borum, P.R., 2010. Acylcarnitines: role in brain. Prog. Lipid Res.
Beketov, M.A., Liess, M., 2008. Acute and delayed effects of the neonicotinoid insecticide 49, 61–75.
thiacloprid on seven freshwater arthropods. Environ. Toxicol. Chem. 27 (2): Jones, O.A.H., Swain, S.C., Svendsen, C., Griffin, J.L., Sturzenbaum, S.R., Spurgeon, D.J., 2012.
461–470. https://doi.org/10.1897/07-322R.1. Potential new method of mixture effects testing using metabolomics and
Beketov, M.A., Schäfer, R.B., Marwitz, A., Paschke, A., Liess, M., 2008. Long-term stream in- Caenorhabditis elegans. J. Proteome Res. 11, 1446–1453.
vertebrate community alterations induced by the insecticide thiacloprid: effect con- Jongeneelen, F.J., 2001. Benchmark guideline for urinary 1-hydroxypyrene as biomarker
centrations and recovery dynamics. Sci. Total Environ. 405 (1–3):96–108. https:// of occupational exposure to polycyclic aromatic hydrocarbons. Ann. Occup. Hyg. 45
doi.org/10.1016/j.scitotenv.2008.07.001. (1), 3–13.
Beketov, M.A., Kefford, B.J., Schäfer, R.B., Liess, M., 2013. Pesticides reduce regional biodi- Kenéz, A., Dänicke, S., Rolle-Kampczyk, U., von Bergen, M., Huber, K., 2016. A metabolo-
versity of stream invertebrates. PNAS 110 (27):11039–11043. https://doi.org/ mics approach to characterize phenotypes of metabolic transition from late preg-
10.1073/pnas.1305618110. nancy to early lactation in dairy cows. Metabolomics 12:165. https://doi.org/
Brandt, A., Gorenflo, A., Siede, R., Meixner, M., Büchler, R., 2016. The neonicotinoids 10.1007/s11306-016-1112-8.
thiacloprid, imidacloprid, and clothianidin affect the immunocompetence of honey Kerr, W.G., Colucci, F., 2011. Inositol phospholipid signaling and the biology of natural
bees (Apis mellifera L.). J. Insect Physiol. 86:40–47. https://doi.org/10.1016/j. killer cells. J. Innate Immun. 3 (3), 249–257.
jinsphys.2016.01.001. Knillmann, S., Orlinskiy, P., Kaske, O., Foit, K., Liess, M., 2018. Indication of pesticide effects
Casagrand, J.L., Ritzmann, R.E., 1992. Biogenic amines modulate synaptic transmission be- and recolonization in streams. Sci. Total Environ. 630, 1619–1627.
tween identified giant interneurons and thoracic interneurons in the escape system Kumar, S., Pillai, M., 2011. Correlation between the reproductive potential and the pyre-
of the cockroach. J. Neurobiol. 23 (6), 644–655. throid resistance in an Indian strain of filarial vector, Culex quinquefasciatus Say (Dip-
Chandre, F., Darriet, F., Darder, M., Cuany, A., Doannio, J.M.C., Pasteur, N., Guillet, P., 1998. tera: Culicidae). Bull. Entomol. Res. 101 (01), 25–31.
Pyrethroid resistance in Culex quinquefasciatus from West Africa. Med. Vet. Entomol. Kuznetsova, A., Brockhoff, P.B., Christensen, R.H.B., 2017. lmerTest: tests in linear mixed
12 (4), 359–366. effects models. (R package version 2.0-30). https://CRAN.R-project.org/package=
Clements, A.N., 1999. The biology of mosquitoes. Sensory Reception and Behaviour. Vol. lmerTest 82 (13). https://doi.org/10.18637/jss.v082.i13.
2. CABI Publishing, Wallingford, p. 752. Lankadurai, B.P., Furdui, V.I., Reiner, E.J., Simpson, A.J., Simpson, M.J., 2013. 1H NMR-based
Coors, A., De Meester, L., 2008. Synergistic, antagonistic and additive effects of multiple metabolomic analysis of sub-lethal perfluorooctane sulfonate exposure to the earth-
stressors: predation threat, parasitism and pesticide exposure in Daphnia magna. worm, Eisenia fetida, in soil. Meta 3:718–740. https://doi.org/10.3390/
J. Appl. Ecol. 45 (6):1820–1828. https://doi.org/10.1111/j.1365-2664.2008.01566.x. metabo3030718.
De Rosario-Martinez, H., 2015. phia: post-hoc interaction analysis. (R package version Lanz, C., Patterson, A.D., Slavík, J., Krausz, K.W., Ledermann, M., Gonzalez, F.J., Idle, J.R.,
0.2-0). http://CRAN.R-project.org/package=phia. 2009. Radiation metabolomics. 3. Biomarker discovery in the urine of gamma-
Di Prisco, G., Cavaliere, V., Annoscia, D., Varricchio, P., Caprio, E., Nazzi, F., Gargiulo, G., irradiated rats using a simplified metabolomics protocol of gas chromatography-
Pennacchio, F., 2013. Neonicotinoid clothianidin adversely affects insect immunity mass spectrometry combined with random forests machine learning algorithm.
and promotes replication of a viral pathogen in honey bees. PNAS 110 (46): Radiat. Res. 172 (2), 198–212.
18466–18471. https://doi.org/10.1073/pnas.1314923110. Laurino, D., Porporato, M., Patetta, P., Manino, A., 2011. Toxicity of neonicotinoid insecti-
Ecker, J., Liebisch, G., 2014. Application of stable isotopes to investigate the metabolism of cides to honey bees: laboratory tests. Bull. Insectol. 64 (1), 107–113.
fatty acids, glycerophospholipid and sphingolipid species. Prog. Lipid Res. 54, 14–31. Lemhonwah, A.M., Hawkins, C.E., Tam, C., Wong, J., Mai, L., Tein, I., 2008. Expression pat-
EU, 2013a. Commission Regulation (EU) No 284/2013 setting out the data requirements terns of the organic cation/carnitine transporter family in adult murine brain. Brain
for plant protection products, in accordance with Regulation (EC) No 1107/2009 of Dev. 39 (1), 31–42.
the European Parliament and of the Council concerning the placing of plant protec- Liess, M., Von Der Ohe, P.C., 2005. Analyzing effects of pesticides on invertebrate commu-
tion products on the market. Off. J. Eur. Union L 93, 85–150. nities in streams. Environ. Toxicol. Chem. 24 (4):954–965. https://doi.org/10.1987/
EU, 2013b. Commission Regulation (EU) No 485/2013 amending Implementing Regula- 03-652.1.
tion (EU) No 540/2011, as regards the conditions of approval of the active substances Liess, M., Foit, K., Becker, A., Hassold, E., Dolciotti, I., Kattwinkel, M., Duquesne, S., 2013.
clothianidin, thiamethoxam and imidacloprid, and prohibiting the use and sale of Culmination of low-dose pesticide effects. Environ. Sci. Technol. 47 (15):
seeds treated with plant protection products containing those active substances. 8862–8868. https://doi.org/10.1021/es401346d.
Off. J. Eur. Union L 139, 12–26.
 R. Russo et al. / Science of the Total Environment 643 (2018) 1533–1541 1541

Liess, M., Foit, K., Knillmann, S., Schäfer, R.B., Liess, H.D., 2016. Predicting the synergy of Rodriguez, M., Ortiz, E., Bisset, J.A., Hemingway, J., Saledo, E., 1993. Changes in malathion
multiple stress effects. Sci. Rep. 6, 32965. https://doi.org/10.1038/srep32965. and pyrethroid resistance after cypermethrin selection of Culex quinquefasciatus field
Lin, C.Y., Viant, M.R., Tjeerdema, R.S., 2006. Metabolomics: methodologies and applica- populations of Cuba. Med. Vet. Entomol. 7 (2), 117–121.
tions in the environmental sciences. J. Pest. Sci. 31, 245–251. Rodrıguez-Sanchez, I.P., Treviño-Alvarado, V.M., Torres-Sepulveda, M.R., Lopez-Saldaña,
Lucas, A.D., Jones, A.D., Goodrow, M.H., Saiz, S.G., Blewett, C., Seiber, J.N., Hammock, B.D., L.A., Ponce-Garcıa, G., Lopez-Uriarte, G.A., et al., 2015. Reference values for amino
1993. Determination of atrazine metabolites in human urine: development of a bio- acids and acylcarnitines in peripheral blood in Quarter horses and American Minia-
marker of exposure. Chem. Res. Toxicol. 6, 107–116. ture horses. Acta Vet. Scand. 57 (1), 1–4.
Martin-Park, A., Gomez-Govea, M.A., Lopez-Monroy, B., Treviño-Alvarado, V.M., Torres- Roschinger, W., Muntau, A.C., Duran, M., Dorland, L., IJlst, L., Wanders, R.J., Roscher, A.A.,
Sepúlveda, M.D., López-Uriarte, G.A., Villanueva-Segura, O.K., Ruiz-Herrera, M.D., 2000. Carnitine-acylcarnitine translocase deficiency: metabolic consequences of an
Martiniez-Fierro, M.L., Delgado-Enciso, I., Flores-Suárez, A.E., White, G.S., Martínez impaired mitochondrial carnitine cycle. Clin. Chim. Acta 298 (1–2), 55–68.
de Villareal, L.E., Ponce-Garcia, G., Black IV, W.C., Rodríguez-Sanchez, I.P., 2017. Pro- Rußmayer, H., Buchetics, M., Gruber, C., Valli, M., Grillitsch, K., Modarres, G., Guerrasio, R.,
files of amino acids and acylcarnitines related with insecticide exposure in Culex Klavins, K., Neubauer, S., Drexler, H., Steiger, M., Troyer, C., Al Chalabi, A., Krebiehl, G.,
quinquefasciatus (Say). PLoS One 12 (1). https://doi.org/10.1371/journal. Sonntag, D., Zellnig, G., Daum, G., Graf, A.B., Altmann, F., Koellensperger, G., Hann, S.,
pone.0169514. Sauer, M., Mattanovich, D., Gasser, B., 2015. Systems-level organization of yeast
Marx-Stoelting, P., Braeuning, A., Buhrke, T., Lampen, A., Niemann, L., Oelgeschlaeger, M., methylotrophic lifestyle. BMC Biol. 13 (80). https://doi.org/10.1186/s12915-015-
Rieke, S., Schmidt, F., Heise, T., Pfeil, R., Solecki, R., 2015. Application of omics data in 0186-5.
regulatory toxicology: report of an international BfR expert workshop. Arch. Toxicol. Russo, R., Becker, J.M., Liess, M., 2018. Sequential exposure to low levels of pesticides and
89 (11), 2177–2184. temperature stress increase toxicological sensitivity of crustaceans. Sci. Total Environ.
Matozzo, V., Gagné, F., Marin, M.G., Ricciardi, F., Blais, C., 2008. Vitellogenin as a biomarker 610–611:563–569. https://doi.org/10.1016/j.scitotenv.2017.08.073.
of exposure to estrogenic compounds in aquatic invertebrates: a review. Environ. Int. Schäfer, R.B., Von Der Ohe, P.C., Rasmussen, J., Kefford, B.J., Beketov, M., Schulz, R., Liess,
34, 531–545. M., 2012. Thresholds for the effects of pesticides on invertebrate communities and
Maute, K., French, K., Story, P., Bull, C.M., Hose, G.C., 2017. Short and long-term impacts of leaf breakdown in stream ecosystems. Environ. Sci. Technol. 46 (9):5134–5142.
ultra-low-volume pesticide and biopesticide applications for locust control on non- https://doi.org/10.1021/es2039882.
target arid zone arthropods. Agric. Ecosyst. Environ. 240, 233–243. Scott, J.G., Yoshimizu, M.H., Kasai, S., 2015. Pyrethroid resistance in Culex pipiens mosqui-
Mckelvie, R.J., Yuk, J., Xu, J., Simpson, A.J., Simpson, M.J., 2009. 1H NMR and GC/MS meta- toes. Pestic. Biochem. Physiol. 120, 68–76.
bolomics of earthworm responses to sub-lethal DDT and endosulfan exposure. Meta- Shi, Y., Xu, X., Chen, J., Liang, R., Zheng, X., Shi, Y., Wang, Y., 2018. Antioxidant gene expres-
bolomics 5:84–94. https://doi.org/10.1007/s11306-008-0122-6. sion and metabolic responses of earthworms (Eisenia fetida) after exposure to various
Mercer, R.A., Menzel, R., 1982. The effects of biogenic amines on conditioned and uncon- concentrations of hexabromocyclododecane. Environ. Pollut. 232, 245–251.
ditioned responses to olfactory stimuli in the honeybee Apis mellifera. J. Comp. Phys- Simon-Delso, N., Amaral-Rogers, V., Belzunces, L.P., Bonmatin, J.M., Chagnon, M., Downs,
iol. 145 (3), 363–368. C., Furlan, L., Gibbons, D.W., Giorio, C., Girolami, V., Goulson, D., Kreutzweiser, D.P.,
Mo, J.-C., Yang, T.-C., Cheng, T.-C., Song, X.-g., 2002. Lethal and sublethal effects of Krupke, C.H., Liess, M., Long, E., McField, M., Mineau, P., Mitchell, E.A.D., Morrissey,
acetamiprid on the larvae of Culex pipiens pallens. J. Insect Sci. 9 (3):45–49. https:// C.A., Noome, D.A., Pisa, L., Settele, J., Stark, J.D., Tapparo, A., Van Dyck, H., Van
doi.org/10.1111/j.1744-7917.2002.tb00153.x. Praagh, J., Van der Sluijs, J.P., Whitehorn, P.R., Wiemers, M., 2015. Systemic insecti-
Moser, V.C., Stewart, N., Freeborn, D.L., Crooks, J., MacMillan, D.K., Hedge, J.M., Wood, C.E., cides (neonicotinoids and fipronil): trends, uses, mode of action and metabolites. En-
McMahen, R.L., Strynar, M.J., Herr, D.W., 2015. Assessment of serum biomarkers in viron. Sci. Pollut. Res. Int. 22, 5–34.
rats after exposure to pesticides of different chemical classes. Toxicol. Appl. Siskos, A.P., Jain, P., Römisch-Margl, W., Bennett, M., Achaintre, D., Asad, Y., Marney, L.,
Pharmacol. 282 (2), 161–174. Richardson, L., Koulman, A., Griffin, J.L., Raynaud, F., Scalbert, A., Adamski, J., Prehn,
Mullin, C.A., Saunders II, M.C., Leslie, T.W., Biddinger, D.J., Fleischer, S.J., 2005. Toxic and C., Keun, H.C., 2017. Interlaboratory reproducibility of a targeted metabolomics plat-
behavioral effects to carabidae of seed treatments used on Cry3Bb1- and Cry1Ab/c- form for analysis of human serum and plasma. Anal. Chem. 89 (1):656–665. https://
protected corn. Environ. Entomol. 34 (6), 1626–1636. doi.org/10.1021/acs.analchem.6b02930.
Münze, R., Hannemann, C., Orlinskiy, P., Gunold, R., Paschke, A., Foit, K., Becker, J., Kaske, Soares, J.B.R.C., Gaviraghi, A., Oliveira, M.F., 2015. Mitochondrial physiology in the major
O., Paulsson, E., Peterson, M., Jernstedt, H., Kreuger, J., Schüürmann, G., Liess, M., arbovirus vector Aedes aegypti: substrate preferences and sexual differences define
2017. Pesticides from wastewater treatment plant effluents affect invertebrate com- respiratory capacity and superoxide production. PLoS One 10 (3). https://doi.org/
munities. Sci. Total Environ. 599-600:387–399. https://doi.org/10.1016/j. 10.1371/journal.pone.0120600.
scitotenv.2017.03.008. Sparks, T.C., Nauen, R., 2015. IRAC: mode of action classification and insecticide resistance
Nagato, E.G., D'eon, J.C., Lankadurai, B.P., Poirier, D.G., Reiner, E.J., Simpson, A.J., Simpson, management. Pestic. Biochem. Physiol. 121, 122–128.
M.J., 2013. (1)H NMR-based metabolomics investigation of Daphnia magna responses Tralau, T., Oelgeschläger, M., Gürtler, R., Heinemeyer, G., Herzler, M., Höfer, T., Itter, H.,
to sub-lethal exposure to arsenic, copper and lithium. Chemosphere 93 (2):331–337. Kuhl, T., Lange, N., Lorenz, N., Müller-Graf, C., Pabel, U., Pirow, R., Ritz, V., Schafft, H.,
https://doi.org/10.1016/j.chemosphere.2013.04.085. Schneider, H., Schulz, T., Schumacher, D., Zellmer, S., Fleur-Böl, G., Greiner, M.,
Nicholls, A.W., Holmes, E., Lindon, J.C., Shockcor, J.P., Farrant, R.D., Haselden, J.N., Lahrssen-Wiederholt, M., Lampen, A., Luch, A., Schönfelder, G., Solecki, R.,
Damment, S.J.P., Waterfield, C.J., Nicholson, J.K., 2001. Metabonomic investigations Wittkowski, R., Hensel, A., 2015. Regulatory toxicology in the twenty-first century:
into hydrazine toxicity in the rat. Chem. Res. Toxicol. 14, 975–987. challenges, perspectives and possible solutions. Arch. Toxicol. 89:823–850. https://
Nyman, A.M., Hintermeister, A., Schirmer, K., Ashauer, R., 2013. The insecticide doi.org/10.1007/s00204-015-1510-0.
imidacloprid causes mortality of the freshwater amphipod Gammarus pulex by inter- Tufi, S., Stel, J.M., de Boer, J., Lamoree, M.H., Leonard, P.E.G., 2015. Metabolomics to explore
fering with feeding behavior. PLoS One 8 (5). https://doi.org/10.1371/journal. imidacloprid-induced toxicity in the central nervous system of the freshwater snail
pone.0062472. Lymnaea stagnalis. Environ. Sci. Technol. 49:14529–14536. https://doi.org/10.1021/
Oberbach, A., Blüher, M., Wirth, H., Till, H., Kovacs, P., Kullnick, Y., Schlichting, N., Tomm, acs.est.5b03282.
J.M., Rolle-Kampczyk, U., Murugaiyan, J., Binder, H., Dietrich, A., von Bergen, M., 2011. Whiting, S.A., Strain, K.E., Campbell, L.A., Young, B.G., Lydy, M.J., 2014. A multi-year field
Combined proteomic and metabolomic profiling of serum reveals association of the study to evaluate the environmental fate and agronomic effects of insecticide mix-
complement system with obesity and identifies novel markers of body fat mass tures. Sci. Total Environ. 497-98, 534–542.
changes. J. Proteome Res. 10 (10):4769–4788. https://doi.org/10.1021/pr2005555. Wieczorek, M.V., Bakanov, N., Bilancia, D., Szöcs, E., Stehle, S., Bundschuh, M., Schulz, R.,
OECD, 1998. Test no. 211: Daphnia magna reproduction test. Guidelines for Testing of 2018. Structural and functional effects of a short-term pyrethroid pulse exposure
Chemicals. OECD Publishing, Paris. on invertebrates in outdoor stream mesocosms. Sci. Total Environ. 610-611, 810–819.
Papathanassiu, A.E., Ko, J.-H., Imprialou, M., Bagnati, M., Srivastava, P.K., Vu, H.A., Cucchi, Williamson, S.M., Wright, G.A., 2013. Exposure to multiple cholinergic pesticides impairs
D., McAdoo, S.P., Ananieva, E.A., Mauro, C., Behmoaras, J., 2017. BCAT1 controls met- olfactory learning and memory in honeybees. J. Exp. Biol. 216:1799–1807. https://doi.
abolic reprogramming in activated human macrophages and is associated with in- org/10.1242/jeb.083931.
flammatory diseases. Nat. Commun. 8, 16040. https://doi.org/10.1038/ Wu, H., Zhang, X., Wang, Q., Li, L., Ji, C., Liu, X., Zhao, J., Yin, X., 2013. A metabolomic inves-
ncomms16040. tigation on arsenic-induced toxicological effects in the clam Ruditapes philippinarum
Paul, A., Harrington, L.C., Zhang, L., Scott, J.G., 2005. Insecticide resistance in Culex pipiens under different salinities. Ecotoxicol. Environ. Saf. 90:1–6. https://doi.org/10.1016/j.
from New York. J. Am. Mosq. Control Assoc. 21 (3):305–309. https://doi.org/10.2987/ ecoenv.2012.02.022.
8756-971X(2005)21[305:IRICPF]2.0.CO;2. Zhang, J., Shen, H., Xu, W., Xia, Y., Barr, D.B., Mu, X., Wang, X., Liu, L., Huang, Q., Tian, M.,
Pecenka, J.R., Lundgren, J.G., 2015. Non-target effects of clothianidin on monarch butter- 2014. Urinary metabolomics revealed arsenic internal dose-related metabolic alter-
flies. Ann. Sci. Nat. 102:19. https://doi.org/10.1007/s00114-015-1270-y. ations: a proof-of-concept study in a Chinese male cohort. Environ. Sci. Technol. 48
Relyea, R.A., Mills, N., 2001. Predator-induced stress makes the pesticide carbaryl more (20), 12265–12274.
deadly to gray treefrog tadpoles (Hyla versicolor). PNAS 98 (5):2491–2496. https://
doi.org/10.1073/pnas.031076198.
Robertson, D.G., 2005. Metabonomics in toxicology: a review. Toxicol. Sci. 85, 809–822.