Neurotoxicology and Teratology 44 (2014) 62–69

Contents lists available at ScienceDirect

Neurotoxicology and Teratology

j ourna l homepage: www.e lsev ie r .com/ locate /neutera

Subchronic atrazine exposure changes defensive behaviour profileand disrupts brain acetylcholinesterase activity of zebrafish

Ademir J. Schmidel a,1, Karla L. Assmann a,1, Chariane C. Werlang a,1, Kanandra T. Bertoncello a,Francini Francescon a, Cassiano L. Rambo a, Gabriela M. Beltrame a, Daiane Calegari a, Cibele B. Batista b,Rachel E. Blaser c, Walter A. Roman Júnior a, Greicy M.M. Conterato a, Angelo L. Piato a, Leila Zanatta a,Jacir Dal Magro a, Denis B. Rosemberg a,d,e,⁎a Programa de Pós-graduação emCiências Ambientais, Área de Ciências Naturais e Exatas, Universidade Comunitária da Região de Chapecó, Avenida Senador Attílio Fontana, 591E, 89809–000 Chapecó,SC, Brazilb Programa de Pós-graduação em Zoologia, Faculdade de Biociências, Pontifícia Universidade Católica do Rio Grande do Sul, Avenida Ipiranga, 6681, 90619–900, Porto Alegre, RS, Brazilc Department of Psychology, University of San Diego, 5998 Alcala Park, San Diego, CA 92110, USAd Instituto Nacional de Ciência e Tecnologia em Excitotoxicidade e Neuroproteção (INCT-EN), 90035–003, Porto Alegre, RS, Brazile Zebrafish Neuroscience Research Consortium (ZNRC)

⁎ Corresponding author at: Programa de Pós-graduaçãDepartamento de Bioquímica e Biologia Molecular, CentrUniversidade Federal de Santa Maria. Avenida Roraima,RS, Brazil. Tel./fax: +55 49 33218179.

E-mail address: dbrosemberg@gmail.com (D.B. Rosem1 These authors contributed equally to this work.

http://dx.doi.org/10.1016/j.ntt.2014.05.0060892-0362/© 2014 Elsevier Inc. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 24 March 2014Received in revised form 22 May 2014Accepted 23 May 2014Available online 2 June 2014

Keywords:AtrazineZebrafishShoaling behaviourExploratory activityAcetylcholinesterase

Animal behaviour is the interaction between environment and an individual organism,which also can be influencedby its neighbours. Variations in environmental conditions, as those caused by contaminants, may lead to neuro-chemical impairments altering the pattern of thebehavioural repertoire of the species. Atrazine (ATZ) is anherbicidewidely used in agriculture that is frequently detected in surfacewater, affecting non-target species. The zebrafish is avaluablemodel organism to assess behavioural andneurochemical effects of different contaminants since it presentsa robust behavioural repertoire and also all major neurotransmitter systems described for mammalian species. Thegoal of this study was to evaluate the effects of subchronic ATZ exposure in defensive behaviours of zebrafish(shoaling, thigmotaxis, and depth preference) using the split depth tank. Furthermore, to investigate a putativerole of cholinergic signalling on ATZ-mediated effects, we tested whether this herbicide alters acetylcholinesterase(AChE) activity in brain and muscle preparations. Fish were exposed to ATZ for 14 days and the following groupswere tested: control (0.2% acetone) and ATZ (10 and 1000 μg/L). The behaviour of four animals in the same tankwas recorded for 6 min and biological samples were prepared. Our results showed that 1000 μg/L ATZ significantlyincreased the inter-fish distance, as well as the nearest and farthest neighbour distances. This group also presentedan increase in the shoal area with decreased social interaction. No significant differences were detected for thenumber of animals in the shallow area, latency to enter the shallow and time spent in shallow and deep areas ofthe apparatus, but the ATZ 1000 group spent significantly more time near the walls. Although ATZ did not affectmuscular AChE, it significantly reduced AChE activity in brain. Exposure to 10 μg/L ATZ did not affect behaviour orAChE activity. These data suggest that ATZ impairs defensive behaviours of zebrafish, which could be related to itsaction on brain cholinergic neurotransmission. Moreover, the use of the split depth tank could be an alternativestrategy to assess group behaviour and depth preference after exposure to chemical compounds.

© 2014 Elsevier Inc. All rights reserved.

1. Introduction

Animal behaviour is the interaction between an organism and its sur-rounding environment. The “perception” of environmental characteris-tics is a function of the sensory neurons, in which the stimuli are further

o em Bioquímica Toxicológica,o de Ciências Naturais e Exatas,1000, 97105–900, Santa Maria,

berg).

processed in the central nervous system (CNS). A sequential “response”is triggered by the activation of motor neurons leading to modificationsin the structure of the behaviour (Fonio et al., 2012). Changes in thisperception–response circuitry may affect social preferences, intraspecificinteractions, and also the responses of organisms when they are chal-lenged with potentially aversive situations (Sih et al., 2012). Thus, theevolution of the complexity of neural systems has resulted in sophisticat-ed neuronal responses that allow a fast adaptation to environmentalchanges in face of different ecological contexts (Hayden et al., 2011).

The use of chemical compounds in the environment has drasticallyincreased worldwide in the last few decades. Pesticides/herbicides areamong the most toxic substances due to their high chemical stability,

63A.J. Schmidel et al. / Neurotoxicology and Teratology 44 (2014) 62–69

resistance to metabolism, and their facility in penetrating cellular envi-ronments due to their lipophilic properties (Chen et al., 2007; El-Amraniet al., 2012). In this context, a problem that arises from the widespreaduse of chemical compounds is the bioaccumulation of pollutants, affect-ing the ecosystem stability, the food chain, and also animal health (Bolleet al., 2004).

Atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazine, ATZ)is one of themost used herbicides throughout the world. The compoundis usually synthesized from cyanuric chloride in the presence ofethylamine and isopropyl amine and the mechanism of action is associ-ated with the inhibition of the plastoquinone-binding protein of plants(Pavlovic et al., 2008). ATZ is applied to attack grassy weeds increasingthe yield of crops, such as sorghum, maize and sugarcane (Corvi et al.,2012; Jin et al., 2010). Despite its common use in agriculture, ATZ mayreach the surface water affecting non-target organisms. Studies demon-strate that ATZ exposure induces toxic effects in animal species leadingto the impairment of embryo development, endocrine disruption, andchanges in neurochemical parameters (Dooley et al., 2013; El-Amraniet al., 2012; Jin et al., 2014). Although it has been postulated that theoxidant properties of ATZ are associated with its deleterious effects onbiological systems (Abarikwu et al., 2013; Elia et al., 2002), there is agrowing necessity to validate other experimental models in order tobetter understand the actions of subchronic ATZ exposure on behaviourand CNS physiology.

The zebrafish (Danio rerio) is a suitable model organism to evaluatethe effects of environmental contaminants (Froehlicher et al., 2009;Hill et al., 2005). One advantage of this species is the rapid absorptionof chemical compounds directly added to the tank water that canreach a variety of tissues, including the brain (Linney et al., 2004;Oliveira et al., 2013). Second, the zebrafish genome has been completelysequenced, facilitating analysis of the interplay between genes andcomplex behaviours, and allowing the development of transgenic bio-markers (Howe et al., 2013; Kalueff et al., 2014). Finally, key neurotrans-mitter systems and brain structures have already been characterized inzebrafish, allowing researchers to examine potential neurotoxic effectsof chemicals in depth (see revision from Rico et al., 2011).

The close relationship between mammalian and zebrafish CNS func-tions has allowed the characterisation of behavioural paradigms analo-gous to those used with rodents (revised in Kalueff et al., 2014).Similarly to other teleosts, zebrafish display a strong tendency to swimclose to conspecifics (Engeszer et al., 2007; Gerlai et al., 2000). This socialbehaviour, also known as shoaling, develops with age and is associatedwith several ecological benefits such as protection from predators,increased foraging ability, improvedmating opportunities, and decreasednovelty stress (Buske and Gerlai, 2011; Pagnussat et al., 2013; Spenceet al., 2008). A recent glossary of zebrafish behaviour has proposed thedescription of the terminologies relevant to several key behaviouraldomains including defensive behaviours (e.g. diving response, shoaling,and thigmotaxis) (Kalueff et al., 2013). This same study argued that theconcomitant evaluation of these parameters could be an interestingstrategy to better understand the relationship of different behaviours aswell as potential biomarkers related to behavioural phenotypes.

Blaser and Goldsteinholm (2012) have designed an apparatus toassess motivational aspects that could drive the diving response inzebrafish, in which the split depth configuration allowed animals toswim freely between a deep and a shallow compartment of the tank.This apparatus was used to conclude that escape from the surface, rath-er than approach to the substrate, motivates the diving responsein zebrafish, as well as to propose dissociation between anxiety-likebehaviours measured by the novel tank and the light–dark tests(Blaser and Rosemberg, 2012;Maximino et al., 2012). Based on the con-served nature of defensive behaviours in vertebrates and consideringthat environmental contaminants could influence ecologically relevantbehavioural patterns, the goal of this study was to investigate whethersubchronic ATZ exposure affects defensive behaviours (e.g. shoaling,thigmotaxis, and depth preference) of zebrafish using the split depth

tank. Moreover, since cholinergic signalling plays a key role in differentbehavioural processes (Picciotto et al., 2012) and that acetylcholinester-ase (AChE) is a potential target of pesticides/herbicides, we evaluatedthe effects of ATZ on AChE activity from brain and muscle preparations.

2. Methods

2.1. Animals

Subjects were 128 adult (4–6 months-old) zebrafish (Danio rerio)from heterogeneous wild-type stock (short-fin phenotype), of mixedgender in a 50:50 male:female ratio, obtained from a local commercialdistributor (Mundo Animal, SC). Animals were kept in the laboratoryfacility for 14 days before the experiments in 40-L tanks at a maximumdensity of four fish per litre. All tanks were filled with non-chlorinatedwater and maintained under constant mechanical, biological, and chem-icalfiltration. The temperature of thewaterwas held at 27±1 ºC,where-as the pH and conductivity were set at 7.0–8.0 and 1500–1600 μS/cm,respectively. The fish were kept on a 14/10 light/dark photoperiod cycle(lights on at 7:00 am) provided by ceiling-mounted fluorescent lighttubes and fed with a commercial flake fish food (alcon BASIC®, Alcon,Brazil) thrice daily. Animals were maintained in accordance with theNational Institute of HealthGuide for Care andUse of Laboratory Animals.The housing conditions and protocolswere approved by the by the EthicsCommittee of Unochapecó (CEUA, #013/13).

2.2. Reagents

Atrazine (CAS number: 1912-24-9), Trizma Base, bovine serum albu-min, acetylthiocholine iodide, and 5,5′-dithiobis-2-nitrobenzoic acid(DTNB) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Allother reagents used were from analytical grade.

2.3. Subchronic atrazine exposure

Before ATZ exposure, zebrafish were transferred to rectangular 3-Ltanks in groups of 12 animals per tank. Fish were exposed for 14 daysto ATZ at the concentrations of 10 and 1000 μg/L (ATZ 10 and ATZ 1000groups, respectively). Despite the U.S. Environmental Protection Agencyconsiders an atrazine concentration of ≤3 ppb in drinking water safefor consumption (Weber et al., 2013), previous data reported that bothtime of exposure and ATZ concentrations chosen induce deleterious ef-fects on biochemical and physiological parameters of zebrafish (Blahováet al., 2013; Dong et al., 2009; Jin et al., 2010). Two stock solutions ofATZ were prepared by dissolving the compound in acetone as vehicle. Acontrol groupwas raised inwater containing 0.2% acetone in the absenceof ATZ. In order tomaintain a similar concentration of the compounds, allsolutions were changed daily during the exposure period and animalswere kept in the same physical conditions of water described for theirhome tanks.

2.4. Behavioural tests

2.4.1. ApparatusThe behavioural profile of zebrafish group was assessed using the

split depth tank as previously described (Blaser and Goldsteinholm,2012; Blaser and Rosemberg, 2012). The apparatus consists of a rectan-gular transparent tank (20 × 15 × 20 cm; length × width × depth)whichwas placed on a stable surfacewith all environmental distractionskept to aminimum. In order to allow a complete exploration of the tank,we used the ‘split’ configuration, in which one side of the aquarium wasset to a depth of 10 cm (shallow area) and the other side set to a depth of15 cm (deep area) with Plexiglas partitions. Importantly, the fish wasable to freely swim into both compartments of the tank. The apparatuswas illuminated by fluorescent lamps from the ceiling (approximately800 lx at the water level). A webcam (Vtrex X6000®) was placed above

64 A.J. Schmidel et al. / Neurotoxicology and Teratology 44 (2014) 62–69

the tank, being connected to a laptop to register the location of thezebrafish.

2.4.2. Measurement of group behaviourFollowing the exposure period, four animals were simultaneously

placed in the test tank and their behaviours were recorded during a sin-gle test of 6 min. After each trial, the tank was cleaned and the waterwas replaced by clean home system water. Additionally, we rotatedthe tank by 180° to eliminate possible spatial effects on fish behaviour.A total of 128 animalswere tested in the behavioural apparatus (N=8).

The videos were further analysed by three trained observers (inter-rater reliability N0.85) blind to the experiment. The behavioural param-eters measured were the proximity between the animals (inter-fishdistance, nearest neighbour distance, and farthest neighbour distance),dispersion of the group (shoal area), and thigmotaxis (an index thatindicates the proximity of fish to the tank walls of the respectivearea). To associate the measures of inter-fish distance and shoal area,we also estimated a social interaction index based on the proximity offish at amaximum distance of 3 body lengths (around 6 cm). This strat-egy allowed us to determine particular nuclei of interaction during thetest period, in which the value “6” corresponds to a complete cohesion(single group) and “0” indicates the lowest cohesion value (animalswidely spread in the apparatus).

All parameters were manually assessed using screenshots madeevery 15 s over the test period (with a total of 25 screenshots pershoal). The experimental protocols were similar to those described byGreen et al. (2012), which demonstrated that the representative groupbehaviour of zebrafish can be easily assessed by manual recordingsusing four animals concomitantly in a test apparatus during a 6-mintrial. Initially, the videos were exported as animated images (.gif files)at a frame rate of 20 frames per second. The respective frames thatcorresponded to each screenshot were further calibrated to the size ofthe tank. The analyses were performed using the ImageJ 1.46 softwarefor Windows®.

2.4.3. Quantification of fish locationThe exploratory activity of the zebrafish group in the split depth tank

was assessed bymeasuring the fish location during the trial (number offish in shallow area, aswell as the latency for the group to enter the shal-low, and the time spent by the group in each compartment). Similarly tothe protocol described above, these parameters were assessed by threetrained “blind” observers. The number of animals in each areawasman-ually scored for each snapshot, while the latency and time spent in shal-low and deep areas were measured with a stopwatch using the videorecordings.

2.4.4. Determination of AChE activityAfter the behavioural experiments, fish were anaesthetised in cold

water at 4 ºC and euthanized by decapitation. The whole brain andmuscle from caudal peduncle were dissected out and a pool of threezebrafish samples was used to prepare an independent homogenatefraction for each tissue. Samples were gently homogenized on ice in 60volumes (v/w) of Tris–citrate buffer (50 mM Tris, 2 mM EDTA, 2 mMEGTA, pH 7.4, adjusted with citric acid) in a motor driven Teflon-glasshomogenizer. The hydrolysis of acetylthiocholine iodide (ACSChI, finalconcentration of 0.8mM)was assessed in 2-mL assay solutions contain-ing 100 mM phosphate buffer, pH 7.5, and 1.0 mM DTNB (Ellman et al.,1961). The experiments were performed during a period of 2–3 minusing a protein amount of 5–10 μg protein in order to ensure the linear-ity of the reactions. Before adding the substrate, samples containingprotein and reaction medium mentioned above were preincubated for10 min at 25 ºC. The hydrolysis of ACSChI was monitored by the forma-tion of thiolate dianion of DTNB at 412 nm in intervals of 30 s. Controlswithout the homogenate preparationwere performed in order to deter-mine the non-enzymatic hydrolysis of substrate. AChE activity wasexpressed as micromole of thiocholine (SCh) released per hour per

milligram of protein. All experiments (N = 6 for brain and muscle)were performed in triplicate.

2.4.5. Determination of total protein contentTheprotein contentwas quantified according to themethodprevious-

ly described by Peterson (1977) using bovine serum albumin as standard.

2.4.6. Statistical analysesThe normal distribution and homogeneity of the data were con-

firmed by Kolmogorov–Smirnov and Bartlett's tests, respectively. Thenumber of interactions in the group aswell as the number of fish in shal-low during the test were analysed by Friedman test, followed by Dunn'smultiple comparison test as post hoc. The area under curve (AUC) wasdetermined to express the social interaction and exploration in shallow,respectively, being represented as arbitrary units. Endpoint results wereexpressed as means± standard error of means (S.E.M.) and analysed byStudent's t test (comparison of water and acetone groups), one-wayanalysis of variance (ANOVA) or two-way ANOVA when the time spentby the group in each compartment was assessed. All other behaviouralprofiles evaluated across the trial were statistically compared by repeat-ed measures ANOVA and the graphs show the representative mean ofthe groups. The Newman–Keuls multiple comparisons test was usedwhen appropriate and the significance level was set at P ≤ 0.05.

3. Results

3.1. Behaviour of zebrafish group in the split depth tank

First, we evaluated whether the split depth tank and the ImageJ 1.46software for Windows® could be used for assessing the group behav-iour of zebrafish. In general, the results showed that zebrafish swaminto groups during the 6 min of test, displaying a strong preference forthe deep area with a subsequent tendency for exploring the shallowcompartment as reported previously (Blaser and Goldsteinholm,2012). Moreover, pilot experiments (N = 8) indicated that the overallbehavioural patterns of fish exposed to home tank water did not signif-icantly differ from those observed in animals exposed to 0.2% acetone(Supplementary Fig. 1). Thus, further experiments were performedusing 0.2% acetone as control group.

3.2. Parameters of social behaviour after ATZ exposure

Subchronic ATZ exposure induced different effects on zebrafish be-haviour depending on the concentration tested. Fig. 1 shows the proxim-ity of fish in the split depth apparatus. One way ANOVA revealed that1000 μg/L ATZ significantly increased the inter-fish distance (F2,21 =3.859, P b 0.05), as well as the nearest neighbour distance (F2,21 =5.220, P b 0.02), and the farthest neighbour distance (F2,21 = 3.909, P b

0.05). Similarly, a 3 × 6 (treatment × time) repeated measures ANOVAindicated that 1000 μg/L ATZ showed increased values for the inter-fishdistance (F2,21 = 3.858, P b 0.05), nearest neighbour distance (F2,21 =3.817, P b 0.05), and farthest neighbour distance (F2,21 = 3.861, P b

0.05) in comparison to those of control and 10 μg/ATZ, as demonstratedby the representative profile of groups (Supplementary Fig. 2).

We further investigated whether the ATZ could affect the distribu-tion of the fish in the test apparatus. A significant increase of the shoalareawas detected for 1000 μg/L ATZ groupwhen compared to both con-trol and ATZ 10 μg/L (F2,21 = 7.156, P b 0.005) (Fig. 2). The analysisacross time revealed significant effects of interaction (F48,504 = 1.388,P b 0.05) and treatment (F2,21 = 7.155, P b 0.005), in which animalsexposed to 1000 μg/L ATZ group displayed a decreased cohesion acrossthe trial (Supplementary Fig. 3).

The number of interactions between fish was affected by ATZ expo-sure (Fig. 3). At 1000 μg/L, ATZ significantly decreased the numberinteractions across the trial (Friedman test, P b 0.0001; Dunn's multiplecomparison test, P b 0.0001) (Fig. 3A). Corroborating these findings, the

Fig. 1. Effects of ATZ on the proximity of fish in the split depth tank. The figure shows the results of the inter-fish distance, nearest neighbour distance, and farthest neighbour distance in asingle trial of 6 min for control, ATZ 10, and ATZ 1000 groups. Data were represented as means ± S.E.M. and analysed by one-way ANOVA followed by Newman–Keuls multiplecomparisons test. Different letters indicate statistical differences. Statistical significance was set at P ≤ 0.05 (N = 8).

65A.J. Schmidel et al. / Neurotoxicology and Teratology 44 (2014) 62–69

AUC data showed a significant decrease in social interaction caused by1000 μg/L ATZ (F2,21 = 3.920, P b 0.05) (Fig. 3B).

3.3. Effects of ATZ on thigmotaxis and on depth preference

Fig. 4 presents the thigmotaxis index of zebrafish after subchronicATZ exposure. Results showed that the ATZ 1000 group explored thetank walls significantly more than control and ATZ 10 (F2,21 = 5.626,P b 0.02). Moreover, a 3 × 6 repeated measures ANOVA revealed amain effect of group (F2,21 = 4.841, P b 0.02), demonstrating that ATZ1000 displayed increased thigmotaxis during the 6-min trial (Supple-mentary Fig. 4). Additionally, Pearson correlation analyses showed asignificant negative relationship between the inter-fish distance andthigmotaxis for ATZ 1000 group (rp= −0.7176, n = 8, P b 0.05).

Next, we evaluatedwhether ATZ alters the exploratory profile of fishin the split depth tank. As shown in Fig. 5, the exposure protocol did notsignificantly change the number of animals in shallow area during thetrial (Friedman test, P N 0.05) (Fig. 5A), nor the exploration of thezebrafish group in the respective compartment (Fig. 5B). AlthoughATZ exposure did not alter the latency to enter the shallow nor thetime spent by the four fish in both shallow and deep areas (Fig. 6),a two-way ANOVA revealed a main effect of tank area (F1,42 = 112.5,P b 0.0001), in which the animals spent more time in deep compart-ment (Fig. 6B).

3.4. Brain AChE activity is altered after ATZ exposure

The effects of subchronic ATZ exposure on AChE activity were evalu-ated in brain and muscle preparations of zebrafish (Fig. 7). Our data

Fig. 2. Dispersion of the shoal in the split depth tank. Endpoint analyses of the shoal area forcontrol, ATZ 10, and ATZ 1000 groups. Data were represented as means ± S.E.M. andanalysed by one-way ANOVA followed by Newman–Keulsmultiple comparisons test. Differ-ent letters indicate statistical differences. Statistical significance was set at P≤ 0.05 (N=8).

demonstrated that 1000 μg/L ATZ produced a significant decrease inbrain AChE activity in comparison to the values of control and ATZ 10groups (F2,15 = 4.691, P b 0.03), whereas ATZ did not change muscularAChE. Importantly, AChE activity from animals exposed to non-chlorinated water did not reveal significant differences in comparisonto the enzyme activity measured in 0.2% acetone group (data notshown).

4. Discussion

Behavioural analyses have a great potential to investigate the neuro-chemical effects of chemical compounds since they may serve as abioindicators of the physicalmanifestation of CNSphysiology integratedwith the environment (Clotfelter et al., 2004). Specifically, a wide rangeof behaviours from non-target organismsmay be altered after exposureto pollutants, such as locomotion, aggression, cognitive abilities, anddefensive behaviours (Zala and Penn, 2004). In the current study, wereport for the first time that the split depth tank may be used to assessdifferent defensive behaviours of a zebrafish group concomitantly,such as shoaling, thigmotaxis, and depth preference using a simplemanual analysis. Moreover, we show that subchronic ATZ exposureimpairs social behaviour, changes the exploratory pattern of fish, anddecreases AChE activity in brain. These novel findings suggest that thebehaviouralmeasures affected by ATZ in zebrafish are somehow associ-ated with its influence on neurochemical parameters.

It is conceivable that changes in social behaviour may have seriousconsequences not only for individual fitness, but for entire populations(Clotfelter et al., 2004). As many other teleosts, zebrafish form shoalswhen placed in a novel tank, a behaviour that matures with age (Buskeand Gerlai, 2012; Engeszer et al., 2007; Gerlai et al., 2000). This strategyallows fast adaptation to the environment and the shoal represents afunctional unity that ensures protection and safety (Pagnussat et al.,2013). Unlike rodents, the locomotor activity of zebrafish occurs inthree dimensions, a factor that complicates the analysis of this behaviour.Recent protocols have shown that social behaviour of zebrafish may bemeasured using automated analyses, which ensures fast and reliablemedium/high-throughput recording (see Maaswinkel et al., 2013a,b).Green et al. (2012) reported that themanual quantification of behaviour-al endpoint data from four zebrafish in the same test tank produces sim-ilar results to those obtained by video-tracking. Although automatedanalyses may provide amore detailed quantification of the three dimen-sions of behaviour, they are usually expensive and restricted to particularresearch groups. To circumvent this problem,we used an apparatuswithtwo compartments differing in depth, which allows depth preferenceand thigmotaxis to be recorded from above the tank as previouslydescribed (Blaser and Goldsteinholm, 2012; Blaser and Rosemberg,2012). Additionally, the use of ImageJ software® provided useful mea-sures of shoal behaviour (endpoint and across time analyses), which

Fig. 3.Effects ofATZon social cohesionof zebrafish. (A)Numberof groupedfish (until 6 cmofdistance) across the trial. The representative areas under curve are represented in the figure.Datawere expressed asmedian± interquartile range andanalysed by Friedman test, follow-ed by Dunn's multiple comparison test. (B) Social interaction value calculated using the areaunder curve of each quadruplet. Endpoint data were expressed as means ± S.E.M. andanalysed by one-way ANOVA followed by Newman–Keulsmultiple comparisons test. Differ-ent letters indicate statistical differences at a P≤ 0.05 level (N= 8).

Fig. 4.Positionof thefish in relation to thewalls of the split depth tank. Theendpoint analysesof thigmotaxis, calculated as the distance ratio (centre:periphery), are depicted for control,ATZ 10, and ATZ 1000 groups. Data were represented as means ± S.E.M. and analysed byone-way ANOVA followed by Newman–Keuls multiple comparisons test. Different lettersindicate statistical differences. Statistical significance was set at P ≤ 0.05 (N= 8).

66 A.J. Schmidel et al. / Neurotoxicology and Teratology 44 (2014) 62–69

was affected by ATZ, an environmental contaminant that may inducedeleterious effects in aquatic biota.

Our results showed that 1000 μg/LATZ significantly reduced the prox-imity between fish, as measured by the inter-fish distance, the nearestneighbour distance, and the farthest neighbour distance. Furthermore,since this group showed an increase in shoal area, the animals weremore dispersed in the split depth tank. Concomitant with changes insocial interaction, the shoal explored both shallow and deep areas in asimilarmanner, but nearer the tankwalls (Fig. 8). Previous studies dem-onstrated that when individual zebrafish are placed in novel environ-ments, they rapidly swim to the bottom, likely due to aversion tosurface (Blaser and Goldsteinholm, 2012; Egan et al., 2009; Levin et al.,2007; Rosemberg et al., 2011). Additionally, habituation to novelty alsohas been related to increased exploration to the centre of a test appara-tus across the trial, in which the fish generally use the periphery as key“safe” points for their exploration (Stewart et al., 2010). A recent studyfrom Pagnussat et al. (2013) demonstrated thatwhen tested individual-ly, zebrafish have higherwhole-body cortisol levels and their behaviourswere more variable due to disruption of their natural shoal tendency.Thus, the exploratory activity in deep and peripheral portions of thetank, together with group formation, constitutes a wide spectrum of be-havioural responses known as defensive behaviours, which improve thefitness of the species in its environment (Kalueff et al., 2013). Althoughthe exact relationship between the effects of ATZ on both social behav-iour and thigmotaxis still requires further investigation, our data strong-ly suggest that ATZ altered the exploratory strategy of the zebrafishgroup by changing the pattern of two important defensive behaviours.Studies related to the influence of ATZ in modulating other behaviouralresponses, such as locomotor parameters, could revealwhether changeson motor patterns are associated to the effects observed on defensivebehaviours.

As important mechanisms of protection, defensive responses havebeenwidelymodelled in animals to assess the neurobiology of differentneurochemical disorders (Ballesteros et al., 2014). Acetylcholine is asignallingmolecule that acts both in the CNS and in neuromuscular syn-apses via nicotinic and muscarinic receptors (Picciotto et al., 2012).AChE is the enzyme responsible for cleaving extracellular acetylcholineinto choline and acetate regulating the cholinergic neurotransmission.Since acetylcholine-mediated signalling is involved in the modulationof several behavioural and cognitive processes (Gold et al., 2011) andboth brain and muscular AChE may be a target of chemical contami-nants, we also investigated the effects of subchronic ATZ exposure onAChE activity from zebrafish. Although the results did not show signifi-cant changes on muscular AChE, exposure to 1000 μg/L ATZ decreasedAChE activity in brain preparations. These data suggest that the effects

Fig. 5. Exploration of zebrafish in the shallow area of the split depth tank. (A) Number offish in the shallow area during the test and the representative areas under curve are rep-resented in thefigure. Data were expressed asmedian± interquartile range and analysedby Friedman test. (B) Exploration in shallow calculated using the area under curve of eachquadruplet. Endpoint data were expressed as means ± S.E.M. and analysed by one-wayANOVA followed by Newman–Keuls multiple comparisons test. Different letters indicatestatistical differences at a P ≤ 0.05 level (N = 8).

Fig. 6. Latency to enter the shallow area and time spent in each compartment of the appa-ratus. The graphs show the time at which the entire group (four animals) entered in theshallow area for the first time during the trial (A) and time spent by the entire group(four animals) in both shallow and deep areas of the split depth tank (B). Data wereexpressed as means ± S.E.M. and analysed by one-way ANOVA and two-way ANOVA,respectively, considering P ≤ 0.05 as significant (N = 8).

Fig. 7. Effects of ATZ on AChE activity from brain and muscle of zebrafish. Data wereexpressed as means ± S.E.M. and analysed by one-way ANOVA followed by Newman–Keuls multiple comparisons test. Different letters indicate statistical differences at aP ≤ 0.05 level (N = 6).

67A.J. Schmidel et al. / Neurotoxicology and Teratology 44 (2014) 62–69

Fig. 8. Ethograms of zebrafish group in the split depth tank after ATZ exposure. Thefiguredepicts the representativemain behaviours during the trial and the effects of ATZ onAChE activityfrom brain andmuscle of zebrafish. The shallow and deep areas are represented as dark and lit grey colours, respectively. The fish are shown as small blue dots and the tankwalls near theanimals have also beenmarked. Note themain differences of behaviours from ATZ 1000 groupwhen compared to those of control and ATZ 10 groups, such as increased inter-fish distance(red arrows) and shoal area (yellow marks). The similar frequency of transitions between deep and shallow areas observed in the experimental groups has been demonstrated (bluearrows).

68 A.J. Schmidel et al. / Neurotoxicology and Teratology 44 (2014) 62–69

of ATZ could be associated with alterations of the cholinergic system inthe CNS, which could reflect in the behavioural patterns measured.

The relationship of the cholinergic systemwith social stress, resilience,and anxiety- and depression-like behaviours has already been demon-strated (Mineur et al., 2013). In this regard, a dysregulation of cholinergicneurotransmissionmay induce negative symptoms of schizophrenia (e.g.social deficit) in animal models (Savage et al., 2011). Since modulation ofdopaminergic, glutamatergic, and serotonergic systems is involved inschizophrenia-related phenotypes, it is plausible that changes in defen-sive behaviours may also be linked to variations on the functionality ofdifferent neurotransmitter systems. Similarly to what occurs in mam-mals, evidence suggests an association between cholinergic signallingwith other neurotransmitters in zebrafish. For example, a recent reportshowed that treatmentwithMK-801 and scopolamine impairedmemoryperformance of zebrafish in a Y-maze task (Cognato et al., 2012) and theNMDA receptor blockade induced hyperlocomotion and impaired shoalcohesion in zebrafish (Maaswinkel et al., 2013a; Seibt et al., 2010). Previ-ous studies also demonstrated that maturation of shoal behaviour inzebrafishmaybe associatedwith increased levels of dopamine, serotonin,and their metabolites in whole-brain extracts (Mahabir et al., 2013) andthat antagonism of dopamine D1 receptor significantly decreases socialpreference (Scerbina et al., 2012). Considering the effects of ATZ, a recentreport showed that short-term ATZ exposure causes behavioural deficitsby disrupting monoaminergic systems in mice, demonstrating that ATZalters the functionality of neurotransmitter systems that influence socialbehaviour (Lin et al., 2010). Our data point the first evidence of theinvolvement of cholinergic signalling in shoal formation of zebrafish.However, the precise molecular mechanisms underlying defensivebehaviour functions in zebrafish involving cholinergic signalling andother neurotransmitter systems following ATZ exposure still remain tobe elucidated in future studies.

In summary, the present report used the split depth tank tomeasuredefensive behaviours of zebrafish after subchronic ATZexposure. Repre-sentative ethograms revealed that the higher concentration of ATZdisrupted social behaviour, increased thigmotaxis, and impaired brainAChE activity of zebrafish (see Fig. 8). These findings indicate that ATZcould have substantial effects on non-target organisms, impairingdefensive behaviours and potentially disrupting social interactions in

aquatic ecosystems by its putative role in modulating the cholinergicsignalling. More generally, these results indicate that zebrafish may bean interesting model organism for further ethotoxicological studies.Our data, combined with molecular and histological tools, may alsoreveal novel mechanisms of brain function related to defensive behav-iour in vertebrates.

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.ntt.2014.05.006.

Conflict of interest statement

The authors have declared that no conflict of interest exists.

Acknowledgements

This study was supported by Conselho Nacional de Pesquisa eTecnologia (CNPq), Coordenação de Aperfeiçoamento de Pessoal deNível Superior (CAPES), Fundação de Amparo à Pesquisa de SantaCatarina (FAPESC), and Unochapecó fellowship grants. A.J.S. andD.C. were recipients of fellowship from Fundação Universitária doDesenvolvimento do Oeste (FUNDESTE, Edital nº105/REITORIA/2012).K.L.A. was a recipient of fellowship from CNPq (Modalidade PIBITI).K.T.B. was a recipient of fellowship from FAPESC (Modalidade Artigo171). G.M.B. was a recipient of fellowship from CAPES. The fundershad no role in study design, data collection and analysis, decision topublish, or preparation of the manuscript.

References

Abarikwu SO, Pant AB, Farombi EO. Quercetin decreases steroidogenic enzyme activity,NF-κB expression, and oxidative stress in cultured Leydig cells exposed to atrazine.Mol Cell Biochem 2013;373:19–28.

Lin Z, Dodd CA, Filipov NM. Short-term atrazine exposure causes behavioral deficits anddisrupts monoaminergic systems in male C57BL/6 mice. Neurotoxicol Teratol 2010;39:26–35.

Ballesteros CI, de Oliveira Galvão B, Maisonette S, Landeira-Fernandez J. Effect of dorsaland ventral hippocampal lesions on contextual fear conditioning and unconditioneddefensive behavior induced by electrical stimulation of the dorsal periaqueductalgray. PLoS ONE 2014;9:e83342.

69A.J. Schmidel et al. / Neurotoxicology and Teratology 44 (2014) 62–69

Blahová J, Plhalová L, Hostovský M, Divišová L, Dobšíková R, Mikulíková I, et al. Oxidativestress responses in zebrafish Danio rerio after subchronic exposure to atrazine. FoodChem Toxicol 2013;61:82–5.

Blaser RE, Goldsteinholm K. Depth preference in zebrafish, Danio rerio: control by surfaceand substrate cues. Anim Behav 2012;83:953–9.

Blaser RE, Rosemberg DB. Measures of anxiety in zebrafish (Danio rerio): dissociation ofblack/white preference and novel tank test. PLoS ONE 2012;7:e36931.

Bolle P, Mastrangelo S, Tucci P, Evandri MG. Clastogenicity of atrazine assessed with theAllium cepa test. Environ Mol Mutagen 2004;43:137–41.

Buske C, Gerlai R. Early embryonic ethanol exposure impairs shoaling and the dopaminergicand serotoninergic systems in adult zebrafish. Neurotoxicol Teratol 2011;33:698–707.

Buske C, Gerlai R. Maturation of shoaling behavior is accompanied by changes in thedopaminergic and serotoninergic systems in zebrafish. Dev Psychobiol 2012;54:28–35.

Chen D, Huang X, Liu L, Shi N. Deltamethrin induces mitochondrial membrane permeabilityand altered expression of cytochrome C in rat brain. J Appl Toxicol 2007;27:368–72.

Clotfelter ED, O'Neal DM, Gaudioso JM, Casto JM, Parker-Renga IM, Snajdr EA, et al.Consequences of elevating plasma testosterone in females of a socially monogamoussongbird: evidence of constraints on male evolution? Horm Behav 2004;46:171–8.

Cognato Gde P, Bortolotto JW, Blazina AR, Christoff RR, Lara DR, Vianna MR, et al. Y-Mazememory task in zebrafish (Danio rerio): the role of glutamatergic and cholinergicsystems on the acquisition and consolidation periods. Neurobiol Learn Mem 2012;98:321–8.

Corvi MM, Stanley KA, Peterson TS, Kent ML, Feist SW, La Du JK, et al. Investigating the im-pact of chronic atrazine exposure on sexual development in zebrafish. Birth DefectsRes B Dev Reprod Toxicol 2012;95:276–88.

Dong X, Zhu L, Wang J, Wang J, Xie H, Hou X, et al. Effects of atrazine on cytochrome P450enzymes of zebrafish (Danio rerio). Chemosphere 2009;77:404–12.

Dooley GP, Tjalkens RB, Hanneman WH. The atrazine metabolite diaminochlorotriazinesuppresses LH release frommurine LβT2 cells by suppressing GnRH-induced intracel-lular calcium transients. Toxicol Res 2013;2:180–6.

Egan RJ, Bergner CL, Hart PC, Cachat JM, Canavello PR, Elegante MF, et al. Understandingbehavioral and physiological phenotypes of stress and anxiety in zebrafish. BehavBrain Res 2009;205:38–44.

El-Amrani S, Pena-Abaurrea M, Sanz-Landaluze J, Ramos L, Guinea J, Cámara C.Bioconcentration of pesticides in zebrafish eleutheroembryos (Danio rerio). Sci TotalEnviron 2012;425:184–90.

Elia AC, Waller WT, Norton SJ. Biochemical responses of bluegill sunfish (Lepomismacrochirus, Rafinesque) to atrazine induced oxidative stress. Bull Environ ContamToxicol 2002;68:809–16.

Ellman GL, Courtney KD, Andres Jr V, Feather-Stone RM. A new and rapid colorimetricdetermination of acetylcholinesterase activity. Biochem Pharmacol 1961;7:88–95.

Engeszer RE, Barbiano LA, Ryan MJ, Parichy DM. Timing and plasticity of shoalingbehaviour in the zebrafish, Danio rerio. Anim Behav 2007;74:1269–75.

Fonio E, Golani I, Benjamini Y. Measuring behavior of animal models: faults and remedies.Nat Methods 2012;9:1167–70.

FroehlicherM, Liedtke A, Groh KJ, Neuhauss SC, Segner H, Eggen RI. Zebrafish (Danio rerio)neuromast: promising biological endpoint linking developmental and toxicologicalstudies. Aquat Toxicol 2009;95:307–19.

Gerlai R, LahavM, Guo S, Rosenthal A. Drinks like a fish: zebra fish (Danio rerio) as a behaviorgenetic model to study alcohol effects. Pharmacol Biochem Behav 2000;67:773–82.

Gold PE, Countryman RA, Dukala D, Chang Q. Acetylcholine release in the hippocampusand prelimbic cortex during acquisition of a socially transmitted food preference.Neurobiol Learn Mem 2011;96:498–503.

Green J, Collins C, Kyzar EJ, PhamM, Roth A, Gaikwad S, et al. Automated high-throughputneurophenotyping of zebrafish social behavior. J NeurosciMethods 2012;210:266–71.

Hayden BY, Pearson JM, Platt ML. Neuronal basis of sequential foraging decisions in apatchy environment. Nat Neurosci 2011;14:933–9.

Hill AJ, Teraoka H, Heideman W, Peterson RE. Zebrafish as a model vertebrate for investi-gating chemical toxicity. Toxicol Sci 2005;86:6–19.

Howe K, Clark MD, Torroja CF, Torrance J, Berthelot C, Muffato M, et al. The zebrafishreference genome sequence and its relationship to the human genome. Nature2013;496:498–503.

Jin Y, Lin X,MiaoW,Wu T, ShenH, Chen S, et al. Chronic exposure of mice to environmen-tal endocrine-disrupting chemicals disturbs their energy metabolism. Toxicol Lett2014;225:392–400.

Jin Y, Zhang X, Shu L, Chen L, Sun L, Qian H, et al. Oxidative stress response and gene ex-pression with atrazine exposure in adult female zebrafish (Danio rerio). Chemosphere2010;78:846–52.

Kalueff AV, Gebhardt M, Stewart AM, Cachat JM, Brimmer M, Chawla JS, et al. Towards acomprehensive catalog of zebrafish behavior 1.0 and beyond. Zebrafish 2013;10:70–86.

Kalueff AV, Stewart AM, Gerlai R. Zebrafish as an emerging model for studying complexbrain disorders. Trends Pharmacol Sci 2014;35:63–75.

Levin ED, Bencan Z, Cerutti DT. Anxiolytic effects of nicotine in zebrafish. Physiol Behav2007;90:54–8.

Linney E, Upchurch L, Donerly S. Zebrafish as a neurotoxicological model. NeurotoxicolTeratol 2004;26:709–18.

Maaswinkel H, Zhu L, Weng W. Using an automated 3D-tracking system to recordindividual and shoals of adult zebrafish. J Vis Exp 2013a;5:50681.

Maaswinkel H, Le X, He L, Zhu L, Weng W. Dissociating the effects of habituation, blackwalls, buspirone and ethanol on anxiety-like behavioral responses in shoalingzebrafish. A 3D approach to social behavior. Pharmacol Biochem Behav 2013b;108:16–27.

Mahabir S, Chatterjee D, Buske C, Gerlai R. Maturation of shoaling in two zebrafish strains:a behavioral and neurochemical analysis. Behav Brain Res 2013;247:1–8.

Maximino C, Benzecry R, Oliveira KRM, Batista EJO, Herculano AM, Rosemberg DB, et al. Acomparison of the light/dark and novel tank tests in zebrafish. Behaviour 2012;149:1099–123.

Mineur YS, Obayemi A, Wigestrand MB, Fote GM, Calarco CA, Li AM, et al. Cholinergicsignaling in the hippocampus regulates social stress resilience and anxiety- anddepression-like behavior. Proc Natl Acad Sci U S A 2013;110:3573–8.

Oliveira TA, Koakoski G, Kreutz LC, Ferreira D, da Rosa JG, de Abreu MS, et al. Alcoholimpairs predation risk response and communication in zebrafish. PLoS ONE 2013;8(10):e75780.

Pagnussat N, Piato AL, Schaefer IC, Blank M, Tamborski AR, Guerim LD, et al. One for alland all for one: the importance of shoaling on behavioral and stress responses inzebrafish. Zebrafish 2013;10:338–42.

Pavlovic D, Vrbnicanin S, Bozic D, Fischer AJ. Morphophysiological traits and atrazinesensitivity in Chenopodium album L. Pest Manag Sci 2008;64:101–7.

Peterson GL. A simplification of the protein assay method of Lowry et al. which is moregenerally applicable. Anal Biochem 1977;83:346–56.

Picciotto MR, Higley MJ, Mineur YS. Acetylcholine as a neuromodulator: cholinergicsignaling shapes nervous system function and behavior. Neuron 2012;76:116–29.

Rico EP, Rosemberg DB, Seibt KJ, Capiotti KM, Da Silva RS, Bonan CD. Zebrafishneurotransmitter systems as potential pharmacological and toxicological targets.Neurotoxicol Teratol 2011;33:608–17.

Rosemberg DB, Rico EP, Mussulini BH, Piato AL, Calcagnotto ME, Bonan CD, et al. Differ-ences in spatio-temporal behavior of zebrafish in the open tank paradigm after ashort-period confinement into dark and bright environments. PLoS ONE 2011;6(5):e19397.

Savage S, Kehr J, Olson L, Mattsson A. Impaired social interaction and enhanced sensitivityto phencyclidine-induced deficits in novel object recognition in rats with corticalcholinergic denervation. Neuroscience 2011;195:60–9.

Scerbina T, Chatterjee D, Gerlai R. Dopamine receptor antagonism disrupts social prefer-ence in zebrafish: a strain comparison study. Amino Acids 2012;43:2059–72.

Seibt KJ, Oliveira Rda L, Zimmermann FF, Capiotti KM, Bogo MR, Ghisleni G, et al. Antipsy-chotic drugs prevent the motor hyperactivity induced by psychotomimetic MK-801in zebrafish (Danio rerio). Behav Brain Res 2010;214:417–22.

Sih A, Cote J, Evans M, Fogarty S, Pruitt J. Ecological implications of behaviouralsyndromes. Ecol Lett 2012;15:278–89.

Spence R, Gerlach G, Lawrence C, Smith C. The behaviour and ecology of the zebrafish,Danio rerio. Biol Rev Camb Philos Soc 2008;83:13–34.

Stewart A, Cachat J, Wong K, Gaikwad S, Gilder T, DiLeo J, et al. Homebase behavior ofzebrafish in novelty-based paradigms. Behav Processes 2010;85:198–203.

Weber GJ, Sepúlveda MS, Peterson SM, Lewis SS, Freeman JL. Transcriptome alterationsfollowing developmental atrazine exposure in zebrafish are associated with disrup-tion of neuroendocrine and reproductive system function, cell cycle, and carcinogen-esis. Toxicol Sci 2013;132:458–66.

Zala SM, Penn DJ. Abnormal behaviours induced by chemical pollution: a review of theevidence and new challenges. Anim Behav 2004;68:649–64.