Fish Physiol Biochem https://doi.org/10.1007/s10695-023-01250-0 RESEARCH Evaluation of the effects of exogenous cortisol manipulation and the glucocorticoid antagonist, RU486, on the exploratory tendency of bluegill sunfish (Lepomis macrochirus) Annaïs Carbajal · Michael J. Lawrence · Kathleen M. Gilmour · Manel Lopez‑Bejar · Steven J. Cooke Received: 9 February 2023 / Accepted: 2 October 2023 © The Author(s) 2023 Abstract In teleost fishes, activation of the hypo- following treatment and then were subjected to a thalamic-pituitary-interrenal axis leads to an eleva- Z-maze trial to characterize the exploratory behav- tion of circulating cortisol levels as a primary stress iour. Cortisol treatment had no measurable effect on response. While acute elevation of cortisol is gener- the exploratory behaviour of bluegill sunfish. Despite ally beneficial, long-term elevation, a common char- presenting a higher probability of refuge emergence, acteristic of chronic stress, may lead to detrimental fish treated with cortisol combined with RU486 effects on health and physiological performance in behaved similarly to cortisol-treated and control fishes. Some stress-mediated behavioural shifts, such groups. While these results suggest that cortisol may as variation along the shy-boldness axis in fish, may not be involved in the mechanisms controlling bold- influence individual fitness. The present study evalu- ness, the influence of cortisol elevation across longer ated the role of cortisol and its mechanisms of action time periods plus validation in different contexts will in the exploratory behaviour of the bluegill sunfish be necessary to confirm this conclusion. (Lepomis macrochirus). Fish were implanted with cocoa butter alone (sham treatment), or cocoa butter Keywords Teleost · Stress · Behaviour · containing cortisol, or cortisol and the glucocorticoid Endocrinology · Glucocorticoid receptor · Boldness receptor antagonist, RU486. A control (untreated) group was also used. Animals were held for 48 h A. Carbajal (*) · M. Lopez‑Bejar M. J. Lawrence Department of Animal Health and Anatomy, Department of Biological Sciences, University Veterinary Faculty, UniversitatAutònoma de Barcelona, of Manitoba, Winnipeg, MB, Canada 08193 Bellaterra, Barcelona, Spain e-mail: anais.carbajal@uab.cat K. M. Gilmour Department of Biology, University of Ottawa, Ottawa, ON, M. Lopez‑Bejar Canada e-mail: manel.lopez.bejar@uab.cat e-mail: kathleen.gilmour@uottawa.ca M. J. Lawrence · S. J. Cooke M. Lopez‑Bejar Fish Ecology and Conservation Physiology Laboratory, College of Veterinary Medicine, Western University Department of Biology, Carleton University, Ottawa, ON, of Health Sciences, Pomona, CA, USA Canada e-mail: m_lawrence27@live.ca S. J. Cooke e-mail: StevenCooke@cunet.carleton.ca Vol.: (0123456789) 13 Fish Physiol Biochem Introduction arises from studies using the receptor antagonist, RU486 (Roy and Rai 2009; Espinoza et al. 2017). Glucocorticoids (GC), the end-product of the hypo- Most published studies have focused on the involve- thalamic-pituitary-interrenal (HPI) axis, are a main ment of GR in modulating fish physiology (Vijayan component of the physiological stress response in and Leatherland 1992; Shaw et al. 2007; Teles et al. teleost fishes (Moberg and Mench 2000; Wende- 2013), while the involvement of these receptors laar Bonga 1997; Schreck and Tort 2016) as in other in modulating fish behaviour is still unclear (Best vertebrates (Moberg and Mench 2000). Cortisol, the and Vijayan 2018; Lawrence et al. 2019b; Ros et al. principal GC in teleost fishes, stimulates physiologi- 2012). cal processes such as gluconeogenesis to mobilize From an ecological perspective, the influence of energy reserves at the expense of other functions such stress on fish behaviour is relevant to understanding as growth, reproduction, or immune function (Sapol- evolutionary trade-offs in nature. Boldness, defined as sky 2002; Schreck & Tort 2016). In responses to an the tendency of an animal to take risks and be explora- acute stressor, short-term HPI axis activation is asso- tory in novel environments (Lima & Dill 1990), influ- ciated with a suite of adaptive behavioural responses ences individual fitness and plays an important role that favour survival, such as increased foraging behav- in evolutionary processes (Wilson & Godin 2009). iour and promotion of escape (Pankhurst 2011; Ellis Many efforts have been made to find out how fish et al. 2012). In this manner, the stress response ena- adjust this behaviour in different environmental con- bles the organism to cope with a potentially harmful texts. Although many behaviours are in some meas- threat (Barton 2002). However, long-term HPI axis ure heritable (Dingemanse et al. 2012; Ariyomo et al. activation (i.e., via chronic stress) can have detrimen- 2013), they can be shaped by environmental chal- tal effects on health and physiological performance lenges and as such, animals can modify their bold- (Branson 2008; Schreck & Tort 2016). Stressors can ness phenotype in relation to the stress experienced lead to reduced growth and reproduction, compromise (Godin and Smith 1988; Killen et al. 2013). A com- the immune system, and impair the animal’s ability mon practice to evaluate how stress influences fish to cope with environmental challenges (Wendelaar behaviour is to administer exogenous GC; however, Bonga 1997; Le Moal 2007; Fuzzen et al. 2011; Sad- this approach has to date produced diverse results oul and Vijayan 2016). (Sopinka et al. 2015; Crossin et al. 2016). Previous Despite the importance of the stress response work has shown that exogenous cortisol implants can in allowing animals to respond to diverse and often increase (O’Connor et al. 2010), decrease (Øverli unpredictable challenges (Angelier & Wingfield et al. 2002; Algera et al. 2017a) or have no impact 2012), the mechanisms through which cortisol modi- (Lawrence et al. 2018b, 2019a) on locomotor activity fies behavioural patterns remain poorly understood of fish, highlighting the complexity of the interaction in wild teleost fishes. Cortisol may exert effects via between stress and fish behaviour. Research focused two pathways, a fast non-genomic route, activated on the connection between behaviour and energetics by membrane-mediated signalling cascades, and a without GC manipulation showed that fish experienc- slower genomic route which involves the transfer of ing different metabolic challenges, such as hunger or cortisol across the cell membrane to regulate target parasitism, tended to behave in a more risky manner, gene transcription and translation (Moore and Evans spending less time beneath a refuge or increasing 1999; Sørensen et al. 2013; Das et al. 2018). To date, their activity levels (Godin and Smith 1988; Killen two intracellular receptor types have been identified et al. 2013). This finding suggests that variation along in teleost fishes, the glucocorticoid receptor (GR), of the boldness axis could be mediated in part by fluc- which there are two paralogs in most teleost fishes tuations of cortisol as a function of energy metabo- (GR1 and GR2), and the mineralocorticoid receptor lism. However, divergent findings emphasize the need (MR) (Bury et al. 2003; Stolte et al. 2006). Mineralo- to further investigate the relationship between the corticoid receptors have a higher affinity for cortisol HPI axis and fish behaviour. The currently described and are usually activated at lower cortisol concen- effects of cortisol on physiology are mediated pre- trations than GRs (Bury & Sturm 2007). Evidence dominantly through activation of GRs (Mommsen for actions of cortisol mediated through GR binding et al. 1999; Shaw et al. 2007; Teles et al. 2013), but Vol:. (1234567890) 13 Fish Physiol Biochem the contribution of this receptor to modulating fish cortisol administration since this vehicle has proven behaviour needs to be further studied. efficient for rapid and sustained elevation of plasma Therefore, the purpose of the present study was cortisol for several days (Gamperl et al. 1994; Sop- to test the hypothesis that chronic cortisol elevation inka et al. 2015) including for bluegill (McConnachie would alter exploratory behaviour. We predicted that et al. 2012b). The dosage validation did not contain a bluegill sunfish (Lepomis macrochirus) with elevated no-injection control group nor a group implanted with cortisol levels would exhibit increased exploratory cortisol combined with RU486 because we aimed behaviour compared to sham-treated (implanted with to identify the cortisol dosage that would produce a cocoa butter alone) and control (untreated) individu- physiologically-relevant elevation of cortisol levels. als. Fish treated with RU486, which was expected Moreover, we decided not to include the non-treated to block cortisol-mediated effects on fish behaviour, control group because in bluegill the implant admin- were predicted to exhibit an exploratory tendency istration is unlikely to influence stress parameters comparable to that observed in control and sham indi- beyond a brief acute stress response from handling viduals. We used bluegill sunfish as a common model (McConnachie et al. 2012b). To ensure even distribu- in behavioural ecology and animal physiology that is tion of cortisol throughout the cocoa butter, the pro- also ecologically relevant as dominant fish commu- cedure of Hoogenboom et al., (2011) was employed. nity members in many northern temperate lakes in Briefly, crystalline cortisol (hydrocortisone 21-hemi- North America. This work contributes to the grow- succinate; Sigma H4881, Sigma-Aldrich, Oakville, ing body of studies that consider the manifold role of Ontario, Canada) was dissolved in ethanol and mixed corticosteroids (beyond just a stress hormone) in wild with melted cocoa butter. The ethanol was then evap- animals (see MacDougall-Shackleton et al. 2019). orated, and 0.5 ml of the mixture was drawn into a 1 ml syringe. Once the mixture reached ambient tem- perature it was stored at − 20◦ C until its use at which Materials and methods time it was again melted for injection. The use of coelomic implants for hormone delivery is consid- Cortisol time course and dosage validation ered a standard procedure in teleost fishes (Gamperl et al. 1994) that has been used extensively in centrar- To select the dose and time course for cortisol treat- chid fishes (e.g. Algera et al. 2017a; Lawrence et al. ment, a group of 120 bluegill sunfish was cap- 2019b; McConnachie et al. 2012b; Rodgers et al., tured by seine net along the shore of Lake Opini- 2012; Zolderdo et al., 2016). Following administra- con (Ontario, Canada) on May 27, 2017 (water tion of the implant, fish were maintained in the hold- temperature ~ 19 °C). Fish were transported back to ing tanks (~ 212 L) as a single shoal per treatment the Queen’s University Biological Station (QUBS; group. Blood samples were collected at 12, 24, 48 or Elgin, Ontario, Canada) immediately after being 96 h after implantation. Previous research on bluegill caught. Experimental series followed the guide- has used similar timepoints for validation studies and lines of the Canadian Council on Animal Care under found that this sampling strategy enables one to fully administration of the Carleton University Animal assess exogenously-manipulated cortisol dynam- Care Committee (AUP 104281). Animals were ran- ics for this species (McConnachie et al. 2012b). At domly assigned to one of three treatment groups: each time point, 10 individuals of each treatment implanted with cocoa butter alone (sham treatment; group were sampled. Blood was withdrawn by cau- 5 ml cocoa butter/kg BW), implanted with cocoa but- dal puncture using a heparinized (­Na+ heparin, 10, ter containing a low-dose cortisol (25 mg cortisol/kg 000 USP units/ml; Sandoz Canada Inc., Boucherville, BW) and implanted with cocoa butter containing a QC, Canada) 1 ml syringe and 23 G needle (Law- high-dose cortisol (50 mg cortisol/kg BW). Implants rence et al. 2020). To avoid cortisol elevation induced consisted of cocoa butter that was warmed until it by the stress of capture, blood collection occurred was liquid and injected using a 1 ml syringe and an within 3 min of netting the fish (Lawrence et al. 18G needle into the coelom of the fish just posterior 2018c). Whole blood [glucose] was analysed imme- to the pelvic fins (Gamperl et al., 2019; Lawrence diately using a portable point-of-care device (Accu et al. 2018a,b; 2019a, b). Cocoa butter was used for Chek Compact Plus; Hoffman, La Roche Limited, Vol.: (0123456789) 13 Fish Physiol Biochem Mississauga, ON, Canada) previously validated for Experimental animals were randomly selected use in teleost fish (Beecham et al. 2006; Cooke et al. from the holding tanks and assigned to one of the 2008; reviewed in Stoot et al., 2014). The remain- four treatments: control (no implant), implanted with ing blood was centrifuged at 2000 g (Mandel Sci- cocoa butter alone (sham treatment), implanted with entific, Guelph, ON, Canada) for 2 min and plasma cocoa butter containing cortisol (cortisol treatment), was collected and stored at -80 ºC for later analysis or implanted with cocoa butter containing cortisol of plasma cortisol titres. Plasma cortisol levels were and the GR antagonist, RU486 (cortisol + RU486 analysed using a commercial radioimmunoassay kit treatment). To avoid temporal bias, the order of (ImmuChem Cortisol Coated Tube RIA Kit, MP Bio- implantation with each treatment was altered daily medicals, Solon, OH, USA) previously validated for and was conducted in a systematically random fash- use in teleost fishes (Gamperl et al. 1994). All sam- ion. Cocoa butter alone (i.e. sham implant; 5 ml/ ples were analysed in a single assay where intra-assay kg body weight [BW]), or cocoa butter containing variability (% CV) was 7.13%. Fish were euthanized cortisol (hydrocortisone 21-hemisuccinate; 25 mg/ by cerebral percussion and total length, body weight, kg BW), or cortisol combined with RU486 (25 mg and liver mass were recorded. Liver mass was used to cortisol/kg BW + 100 mg RU486/kg BW; Doyon calculate the hepatosomatic index [HSI = (liver mass/ et al. 2006; Lawrence et al. 2017) was injected as body mass) × 100%; Busacker et al. 1990], an index described previously; the cortisol dose was chosen of individual energetic state that has been previously on the basis of pilot experiments (see above). We also applied as a stress indicator (Sopinka et al. 2016). included a no-injection control group that was oth- erwise handled identically to treated fish to account Experimental animals for stress and other factors associated with the proce- dure itself (McConnachie et al. 2012a). Immediately For the behavioural trials, a separate group of blue- after administering the implant, dorsal fin spines were gill sunfish (Lepomis macrochirus Rafinesque, 1819; clipped (with unique patterns) to identify treatment total length = 9.8 ± 0.6 cm; mass = 13.9 ± 2.4 g; groups. Fish were then transferred to a holding tank mean ± SD, N = 100) was captured as mentioned pre- (~ 406 L) under the conditions described above for viously, over a 4-day period (N = 25 fish captured per 48 h until behavioural assays were performed. This day) in mid-June 2017 (after the spring reproductive holding period was used to allow the implanted corti- period for bluegill) from Lake Opinicon. Fish were sol to reach ecologically relevant circulating concen- quickly transported back to the Queen’s University trations and ensure that levels of the hormone were Biological Station. Bluegill were allowed to accli- elevated at a level that was reflective of a stressed mate and recover for a 24 h period after being caught. state. The same experimental tank was used for all The acclimation period was based on previous studies treatment groups to eliminate variability due to the in bluegill by some of our team (e.g., Lawrence 2019; potential tank effect. Moynes et al. 2020) as well as observations by other researchers that this species acclimatizes easily to Behavioural analysis laboratory conditions (Wilson & Godin 2009). Blue- gill were kept in indoor flow-through tanks (~ 212 L) Behavioural assessments of control (n = 19), sham supplied with flowing lake water at ambient tempera- (n = 18), cortisol-treated (n = 18), and cortisol-plus ture (23.4 ± 0.8 ºC) under natural photoperiod. Hun- RU486-treated (n = 21) fish were performed 48 h after ger state can potentially influence boldness behav- implant injection. The treatment exposure period of iours (e.g., Gotceitas and Godin 1991; Godin and 48 h was selected on the basis of pilot experiments Smith 1988; Killen et al. 2011). As such, fish were (see above). Since fish were captured over a 4-day not fed while kept in captivity to prevent any poten- period, behavioural tests were carried out during four tial confounding effects. Previous research on bluegill consecutive days. Every day, a group of 25 fish from has revealed that withholding of food for several days all four experimental groups (control, sham, cortisol does not yield biologically meaningful reductions in and cortisol plus RU486 treated 48 h before) were fish condition (Lawrence et al. 2019c) presumably behaviourally tested. The 4-day experimental period given the transient nature of food resources in nature. allowed the daily assessment of a reduced group of Vol:. (1234567890) 13 Fish Physiol Biochem fish in a short time frame and thus avoid the circa- we recorded the latency to emerge from the refuge dian rhythm be a source of variability in cortisol lev- once the door had been opened, the number of lines els (Dickmeis 2008). One Z-maze, placed next to the crossed (i.e. activity levels), and the time to reach the holding tanks, was used to evaluate exploratory activ- end of the maze for the first time. Additionally, if the ity following methods described in previous studies individual returned to the refuge after emerging, the (Chapman et al. 2010; Lawrence et al. 2018b). The total time spent in the refuge during the 5 min trial maze (40 cm × 50 cm) was configured with 4 corri- also was recorded. The proportion of fish that did not dors, each divided into equal quarters to yield a total exit the refuge, and the proportion that did not reach of 18 zones and had a refuge in one corner (Fig. 1). the end of the maze also were recorded. At the end of During trials, the maze was illuminated with artifi- the trial, fish were euthanized via cerebral percussion cial light from above the tank, and the behaviour of and body weight and total length were measured. the test fish was recorded using a single Go-pro Hero camera (Go-Pro, San Mateo, CA, USA) situated above the maze. The maze was filled to a depth of Statistical analysis 6.6 cm with fresh lake water that was changed after each trial. Statistical analyses were conducted using R soft- At the onset of a trial, a fish was placed into the ware (R-project, Version 3.5.1, R Development Core refuge zone with the help of a hand net and a small Team, University of Auckland, New Zealand), and bucket and was allowed one minute of acclimation the level of significance for all tests was 0.05. Data (Chapman et al. 2010; Wood et al. 2011; Raoult et al. were tested for normality using the Shapiro–Wilks 2012). The experiment began when the door between test and for homogeneity of variance with Levene’s the refuge and the maze was opened remotely by an test. Data were square-root or log transformed to meet observer behind a blind. Fish were given 10 min to assumptions of normality and equal variance where leave the refuge, and if the individual did not leave appropriate. the refuge during this time, the trial was ended. If For the dose and time-course validation, two-way the fish did emerge, the animal was given 5 addi- analysis of variance (ANOVA) followed by post- tional minutes to explore the refuge. During the trial, hoc Tukey’s HSD test, as appropriate, was used to test whether cortisol and blood glucose concentra- tions, and HSI, differed significantly across treatment groups (sham, low-dose and high-dose cortisol) over the sampling times (12, 24, 48 and 96 h). Chi-square tests were used to assess the signifi- cance of associations between treatment group (con- trol, sham, cortisol alone, and cortisol combined with RU468) and 1) the probability of exiting the refuge, and 2) the probability of reaching the end of the maze. Because all individuals treated with cortisol plus RU468 exited the refuge, a second chi-square test was performed to test for differences among control, sham and cortisol treatments. To determine whether there were significant differences among treatment groups in 1) the latency to emerge from the refuge, 2) the number of lines crossed, 3) the time to reach the end of the maze, and 4) the total time spent in the refuge during the 5 min trial, general linear models Fig. 1  Representation of the Z-maze with the refuge box (grey (GLM) were used. Water temperature and fish total rectangle) and 18 zones delimited by dotted lines. After one length were added to the models as linear covariates, minute of acclimation, the gate of the refuge was remotely opened (dashed lines), and each individual had 5 min to as was the interaction of these covariates with treat- explore the maze ment group. Vol.: (0123456789) 13 Fish Physiol Biochem Results Time course and cortisol dose validation Treatment group (F = 35.155; df = 2; p < 0.001) and sampling time (F = 6.645; df = 3; p < 0.001) signifi- cantly affected circulating cortisol concentrations (Fig. 2A), but the interaction of these factors was not significant (F = 0.264; df = 6; p = 0.95). Indi- viduals from the low- and high-dose cortisol treat- ments presented higher cortisol values than sham fish (Tukey HSD, p < 0.001). At 96 h, cortisol values were lower than those of fish sampled at 12 h (Tukey HSD, p < 0.001) and at 24 h (Tukey HSD, p = 0.03). Blood glucose and HSI were influenced by treat- ment (glucose, F = 7.207, df = 2, p < 0.001, Fig. 2B; HSI, F = 3.898, df = 2, p = 0.02, Fig. 2C) but not by sampling time (glucose, F = 1.054, df = 3, p = 0.37; HSI, F = 0.592, df = 3, p = 0.64). The interaction of these factors was not significant (glucose, F = 1.094, df = 6, p = 0.37; HSI, F = 1.891, df = 6, p = 0.09). Sham individuals differed from fish in both the low- (Tukey HSD, p = 0.001) and high-dose (Tukey HSD, p = 0.03) cortisol treatments in blood glucose levels. In HSI, differences were close to significant between sham and low-dose cortisol (Tukey HSD, p = 0.06) and significant between sham and high-dose cortisol treatment groups (Tukey HSD, p = 0.04). Behavioural responses Fig. 2  Time course elevation and dosage validation for blue- A Chi-square test revealed that treatment had a signif- gill sunfish (Lepomis macrochirus). Plasma cortisol concentra- icant effect on the probability to exit the refuge when tions (A), blood glucose concentrations (B) and hepatosomatic index (C) of sham (implanted with cocoa butter alone), low- all the treatments were considered in the analysis dose cortisol (25 mg/kg BW) and high-dose cortisol (50 mg (χ2 = 9.043, df = 3, p = 0.03; Table 1). After excluding cortisol/kg BW) treatments sampled from 12 to 96 h after the cortisol treatment combined with RU486 from the implant administration. Values are reported as means ± stand- test, the chi-square analysis revealed no significant ard deviation (SD; error bars). Capital letters represent signifi- cant effects of time (p < 0.001) and lower case letters represent differences among control, sham and cortisol-treated significant effects of dose (p < 0.05). The interaction of these groups (χ2 = 2.400, df = 3, p = 0.30). No significant two factors was not significant (p = 0.95). At every dose (sham, differences were detected among groups for the prob- low-dose and high-dose cortisol) and time period (12 h, 24 h, ability of reaching the end of the maze (χ2 = 4.5623, 48 h and 96 h), the sample size was 10 individuals df = 3, p = 0.21; Table 1). No differences were detected among treatments Discussion in the latency to exit the refuge (F = 1.070; df = 3; p = 0.37; Fig. 3A), the latency to reach the end of the Overview maze (F = 1.751; df = 3; p = 0.18; Fig. 3B), the total time spent inside the refuge during the 5-min trial Because of the influence of behaviour in evolu- (F = 0.982; df = 3; p = 0.40; Fig. 3C), or the number tionary processes and individual fitness (Wilson & of lines crossed (F = 2.512; df = 3; p = 0.07; Fig. 3D). Godin 2009; Noakes and Jones 2016), we aimed to Vol:. (1234567890) 13 Fish Physiol Biochem Table 1  Distribution of treatment groups (control—not Validation of cortisol implants implanted, sham—implanted with cocoa butter alone, cortisol alone—25 mg/kg BW and cortisol combined with RU486— 25 mg cortisol/kg BW + 100 mg RU486/kg BW) between the The time-course of cortisol elevation achieved using number of individuals (%) that did (Yes) or not (No) exit the the implants was monitored in a separate group of refuge box and the number of individuals (%) that reached bluegill. Significant differences in circulating cor- (Yes) or not (No) the end of the maze tisol levels between fish treated with low and high Treatment cortisol dosages were not detected in the validation Control Sham Cortisol Cortisol + RU486 experiment. Nevertheless, subjects from the behav- ioural assessment were implanted with the low dose Exit Yes 13 (68%) 13 (72%) 16 (89%) 21 (100%) of cortisol (25 mg cortisol/kg BW). This implant was No 6 (32%) 5 (28%) 2 (11%) 0 (0%) chosen to elevate plasma cortisol levels given that End Yes 7 (37%) 8 (44%) 11 (61%) 14 (67%) previous experiments have revealed that the use of No 12 (63%) 10 (56%) 7 (39%) 7 (33%) high (supraphysiological) doses of cortisol can lead to altered behavioural responses which may not be of physiological relevance (McConnachie et al. 2012b; gain insight into the effects of the HPI axis in medi- Lawrence et al. 2018a). Importantly, 48 h after the ating fish behaviour by testing whether exploratory administration of a cocoa butter implant contain- tendency in the bluegill sunfish is influenced by cor- ing the low dose of cortisol (25 mg cortisol/kg BW), tisol. Contrary to our predictions, cortisol treatment circulating hormone levels were similar to those had no measurable effect on exploratory behaviour. detected in bluegill after either repeated daily stress- Despite a higher probability of emergence from the ors (Cook et al. 2012) or a severe acute stress (McCo- refuge, fish treated with cortisol combined with nnachie et al. 2012b). Cortisol levels in cortisol- RU486 behaved similarly to those in all other treat- treated fish remained significantly higher than those ment groups. of the sham group up to 48 h post implant administra- tion, and, consistent with the gluconeogenic activity Fig. 3  Metrics for con- trol—not implanted (white bars; n = 19), sham— implanted with cocoa butter alone (grey bars; n = 18), cortisol alone—25 mg/kg BW (dotted bars; n = 18) and cortisol combined with RU486—25 mg cortisol/ kg BW + 100 mg RU486/ kg BW (black bars; n = 21) bluegill sunfish in the maze trial displaying (A) the latency to exit the refuge, (B) the latency to reach the end of the maze, (C) the total time spent in the refuge box, and (D) the number of lines crossed. Values are shown as mean and error bars are standard deviations Vol.: (0123456789) 13 Fish Physiol Biochem of this hormone (reviewed in Mommsen et al. 1999), effect of cortisol treatment on activity was observed cortisol-treated fish displayed elevated blood glucose in creek chub or pumpkinseed sunfish implanted after administration of the implant. Hence, the results with cortisol (Lawrence et al. 2019a; Nagrodski et al. of the time-course experiment suggest that the hor- 2013). While these findings may indicate that cortisol mone dose used in the Z-maze trial under laboratory is not involved in exploratory behaviour, the complex conditions would yield circulating cortisol levels that relations between GC and other signalling systems were physiologically relevant. suggest that this interpretation should be regarded with caution. For example, when faced with a sec- Influence of cortisol on exploratory tendency ondary stressor (i.e. intruder test), cortisol-treated rainbow trout showed impaired locomotor perfor- Cortisol elevation was predicted to increase the mance, but this effect may not be detected in undis- exploratory tendency of wild bluegill sunfish because turbed conspecifics also treated with cortisol (Øverli considerable evidence suggests that fish experiencing et al. 2002). This scenario has long been observed in energetic and metabolic challenges alter the degree vertebrates with behaviours other than activity and of boldness towards risk-taking behaviours (Krause locomotion (Crossin et al. 2016; Sopinka et al. 2015). et al. 2000; Killen et al. 2011, 2013). Contrary to our Accordingly, context may play an important role in predictions, we found that fish treated with cortisol determining how GCs influence exploratory tendency, for 48 h exhibited a behavioural phenotype compara- and a behavioural impairment may not be observed in ble to that observed in control fish. Fish species dif- all situations. We cannot rule out potential cortisol fer considerably in their ability to cope with stressors increases related to relocation from the holding tank through behavioural and/or physiological responses to the Z-maze, which could have masked treatment (Sopinka et al. 2015; Schreck and Tort 2016). For differences. Unfortunately, circulating cortisol levels example, increased locomotor activity was observed were not assessed in the experimental fish once the in zebrafish larvae treated with cortisol for 24 h behavioural trial was finished. To further understand (Best & Vijayan 2018), and rainbow trout exhibited the role of cortisol in the exploratory tendency of increased activity 1.5 h after being fed cortisol-treated bluegill fish it will be of interest to evaluate cortisol food, which decreased by 48.5 h (Øverli et al. 2002). concentrations in the experimental individuals. In zebrafish stressed without further GC manipula- tion, an effect on the activity was revealed as long as Influence of RU486 on exploratory tendency 14 d after the onset of the stressor (Piato et al. 2011; Manuel et al. 2014). An influence of cortisol may not Fish implanted with cortisol plus RU486 behaved be apparent if the time at which a behavioural assay similarly to cortisol-treated and control groups once is performed is too close to or too long after cortisol in the maze. However, the antagonist-treated indi- administration (Algera et al. 2017a, b; Zolderdo et al., viduals were more likely to emerge from the ref- 2016; Lawrence et al. 2019b). Although in the current uge into a novel environment than fish from other study cortisol levels in bluegill sunfish remained ele- treatments. The GR antagonist RU486 blocks the vated for 48 h, it is possible that this duration of cor- cortisol-mediated negative feedback pathway that tisol exposure was too short to detect effects of cor- inhibits HPI axis activity (Bury et al. 2003; Roy tisol on the behaviour of this species. In some cases, and Rai 2009). As a consequence of GC feedback while elevated cortisol promotes specific physiologi- inhibition, higher plasma cortisol levels often are cal changes, behaviour may remain unaltered as an detected in RU486-treated fish compared to con- adaptive mechanism to avoid potential fitness conse- specifics treated with cortisol alone (Reddy et al. quences (Dunlap et al. 2011; Lawrence et al. 2019b; 1995; Veillette et al. 2007). Cumulative increases in Piato et al. 2011). Consistent with our findings, nega- cortisol exposure could lead to the higher propor- tive results have been published in previous studies. tion of RU486-treated individuals that exit the ref- For instance, no change in exploratory tendency was uge. Existing literature supports the effectiveness detected in cortisol-treated pumpkinseed sunfish in of RU486 in inhibiting cortisol-mediated effects by a study carried out under comparable experimen- GR signalling on fish behavioural and physiologi- tal conditions (Lawrence et al. 2018b). Likewise, no cal traits (DiBattista et al. 2005; Schjolden et al. Vol:. (1234567890) 13 Fish Physiol Biochem 2009; Dunlap et al. 2011). The behavioural effect Authors’ contributions All authors contributed to the study observed in the present study therefore could be conception and design. Material preparation, data collection and analysis were performed by Annaïs Carbajal and Michael modulated via the MR, a pathway reported to have J. Lawrence. The first draft of the manuscript was written by an important role in mediating effects of cortisol Annaïs Carbajal and all authors commented on previous ver- on behaviours in teleosts, particularly, in relation sions of the manuscript. All authors read and approved the final to locomotor activity (Stolte et al. 2008; Takahashi manuscript. and Sakamoto 2013; Sakamoto et al. 2016; Faught Funding Open Access Funding provided by Universitat and Vijayan 2018). Nevertheless, given that only Autonoma de Barcelona. Funding was provided by NSERC one measurable effect was detected in the RU486- and the Canada Research Chairs program. treated group among the several metrics evaluated, the biological relevance of this outcome requires Data availability The datasets generated during and/or ana- careful consideration. Further studies are warranted lysed during the current study are available from the corre- sponding author on reasonable request. to reveal whether MR are involved in the modula- tion of exploratory tendency. Declarations Ethical approval Experimental series followed the guide- lines of the Canadian Council on Animal Care under admin- Conclusion istration of the Carleton University Animal Care Committee (AUP 104281). In conclusion, although cortisol implants elevated Competing interests No, I declare that the authors have no plasma cortisol levels, the experiments used in the competing interests, or other interests that might be perceived present study did not provide evidence that corti- to influence the results and/or discussion reported in this paper. sol shapes exploratory behaviour in bluegill sun- fish. Behavioural assessments performed at different Open Access This article is licensed under a Creative Com- time points after cortisol exposure (> 48 h) will be mons Attribution 4.0 International License, which permits necessary to further characterize the role of cortisol use, sharing, adaptation, distribution and reproduction in any in regulating fish behaviour. Treatment with corti- medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Crea- sol plus RU486 showed that antagonist treatment tive Commons licence, and indicate if changes were made. The increased the probability of emergence from the ref- images or other third party material in this article are included uge, although no differences were detected for the in the article’s Creative Commons licence, unless indicated other metrics evaluated. We speculate that signalling otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your through MR could be involved, but this possibility intended use is not permitted by statutory regulation or exceeds requires experimental support. In the present study, the permitted use, you will need to obtain permission directly cortisol levels were not assessed in individuals used from the copyright holder. To view a copy of this licence, visit for behavioural trials, limiting the interpretation of http://creativecommons.org/licenses/by/4.0/. our results. As with all studies involving experimen- tal manipulation of GCs, it is important to note that cortisol treatment is not the same as cortisol eleva- References tion during a stress response in that the neurosensory Algera DA, Brownscombe JW, Gilmour KM, Lawrence MJ, cascade is bypassed (Sopinka et al. 2015). As such, Zolderdo AJ, Cooke SJ (2017a) Cortisol treatment affects there is need for additional research that considers the locomotor activity and swimming behaviour of male effects of diverse stressors on fish to investigate the smallmouth bass engaged in paternal care: A field study context-specificity of GC effects on behaviour. using acceleration biologgers. Physiol Behav 181:59–68. https://​doi.​org/​10.​1016/j.​physb​eh.​2017.​08.​026 Algera DA, Gutowsky LFG, Zolderdo AJ, Cooke SJ (2017b) Acknowledgements We thank the staff of the Queen’s Uni- Parental care in a stressful world: experimentally elevated versity Biological Station for their logistic support. Funding cortisol and brood size manipulation influence nest suc- was provided by NSERC and the Canada Research Chairs pro- cess probability and nest-tending behavior in a wild tel- gram. Scientific collection permits were kindly provided by the eost fish. Physiol Biochem Zool 90(1):000–000. https://​ Ontario Ministry of Natural Resources and Forestry. doi.​org/​10.​1086/​689678 Vol.: (0123456789) 13 Fish Physiol Biochem Angelier F, Wingfield JC (2012) Importance of the gluco- modified by glucocorticoids in rainbow trout, Oncorhyn- corticoid stress response in a changing world: Theory, chus mykiss. Gen Comp Endocrinol 146:126–135 hypotheses and perspectives. Gen Comp Endocrinol Dunlap KD, Jashari D, Pappas KM (2011) Glucocorticoid 190:118–128 receptor blockade inhibits brain cell addition and aggres- Ariyomo TO, Carter M, Watt PJ (2013) Heritability of boldness sive signaling in electric fish, Apteronotus leptorhynchus. and aggressiveness in the zebrafish. Behav Genet 43:161–167 Horm Behav 60:275–283 Barton BA (2002) Stress in fishes: a diversity of responses with Ellis T, Yildiz HY, López-Olmeda J, Spedicato MT, Tort L, particular reference to changes in circulating corticoster- Øverli Ø, Martins CIM (2012) Cortisol and finfish wel- oids. Integr Comp Biol 42:517–525 fare. Fish Physiol Biochem 38:163–188 Beecham RV, Small BC, Minchew CD (2006) Using Portable Espinoza MB, Aedo JE, Zuloaga R, Valenzuela C, Molina A, Lactate and Glucose Meters for Catfish Research: Accept- Valdés JA (2017) Cortisol Induces Reactive Oxygen Spe- able Alternatives to Established Laboratory Methods? N cies Through a Membrane Glucocorticoid Receptor in Am J Aquac 68:291–295 Rainbow Trout Myotubes. J Cell Biochem 118:718–725 Best C, Vijayan MM (2018) Cortisol elevation post-hatch Faught E, Vijayan MM (2018) The mineralocorticoid receptor affects behavioural performance in zebrafish larvae. Gen is essential for stress axis regulation in zebrafish larvae. Comp Endocrinol 257:220–226 Sci Rep 8:18081 Branson EJ (2008) Fish welfare. Blackwell P. Blackwell Pub- Fuzzen, M, Bernier, NJ, Van Der Kraak G (2011) Stress and lishing Ltd, UK reproduction. In: Norris DO, Lopez KH (eds.) Hormones Bury NR, Sturm A (2007) Evolution of the corticosteroid and Reproduction in Vertebrates. Academic Press, New receptor signalling pathway in fish. Gen Comp Endocrinol York. pp 103–117 153:47–56 Gamperl AK, Vijayan MM, Boutilier RG (1994) Experimen- Bury NR, Sturm A, Le Rouzic P, Lethimonier C, Ducouret B, tal control of stress hormone levels in fishes: techniques Guiguen Y, Robinson-Rechavi M, Laudet V, Rafestin- and applications. Rev Fish Biol Fish 4:215–255 Oblin ME, Prunet P (2003) Evidence for two distinct Godin J-GJ, Smith SA (1988) A fitness cost of foraging in functional glucocorticoid receptors in teleost fish. J Mol the guppy. Nature 333:69–71 Endocrinol 31:141–156 Gotceitas V, Godin J-GJ (1991) Foraging under the risk of Busacker GP, Adelman IR, Goolish EM (1990) Growth. In: predation in juvenile Atlantic salmon (Salmo salar L.): Schreck CB, Moyle PB (eds) Methods for Fish Biology. effects of social status and hunger. Behav Ecol Sociobiol American Fisheries Society, Bethesda, pp 363–387 29:255–261 Chapman BB, Morrell LJ, Krause J (2010) Unpredictability in Hoogenboom MO, Armstrong JD, Miles MS, Burton T, food supply during early life influences boldness in fish. Groothuis TGG, Metcalfe NB (2011) Implantation of Behav Ecol 21:501–506 cocoa butter reduces egg and hatchling size in Salmo Cook KV, Connor CMO, Mcconnachie SH, Gilmour KM, trutta. J Fish Biol 79:587–596 Cooke SJ (2012) Condition dependent intra-individual Killen SS, Marras S, Mckenzie DJ (2011) Fuel, fasting, fear: repeatability of stress-induced cortisol in a freshwater fi Routine metabolic rate and food deprivation exert syner- sh. Comp Biochem Physiol Part A 161:337–343 gistic effects on risk-taking in individual juvenile Euro- Cooke SJ, Suski CD, Danylchuk SE, Danylchuk AJ, Donald- pean sea bass. J Anim Ecol 80:1024–1033 son MR, Pullen C, Bulté G, O’Toole A, Murchie KJ, Kop- Killen SS, Marras S, Metcalfe NB, McKenzie DJ, Domen- pelman JB, Shultz AD, Brooks E, Goldberg TL (2008) ici P (2013) Environmental stressors alter relationships Effects of different capture techniques on the physiologi- between physiology and behaviour. Trends Ecol Evol cal condition of bonefish Albula vulpes evaluated using 28:651–658 field diagnostic tools. J Fish Biol 73:1351–1375 Krause J, Cheng DJS, Kirkman E, Ruxton GD (2000) Spe- Crossin GT, Love OP, Cooke SJ, Williams TD (2016) Gluco- cies-specific patterns of refuge use in fish: The role of corticoid manipulations in free-living animals: Considera- metabolic expenditure and body length. Behaviour tions of dose delivery, life-history context and reproduc- 137:1113–1127 tive state. Funct Ecol 30:116–125 Lawrence MJ (2019) Does chronic stress mediate predator- Das C, Thraya M, Vijayan MM (2018) Nongenomic cortisol prey interactions in wild fish? An experimental approach signaling in fish. Gen Comp Endocrinol 265:121–127 using exogenous cortisol implants. PhD Thesis, Carleton DiBattista JD, Anisman H, Whitehead M, Gilmour KM (2005) University, Ottawa, Canada The effects of cortisol administration on social status and Lawrence MJ, Eliason EJ, Brownscombe JW, Gilmour KM, brain monoaminergic activity in rainbow trout Oncorhyn- Mandelman JW, Cooke SJ (2017) An experimental chus mykiss. J Exp Biol 208:2707–2718 evaluation of the role of the stress axis in mediating Dickmeis T (2008) Glucocorticoids and the circadian clock. predator-prey interactions in wild marine fish. Comp. J Endocrinol 200(1):3–22. https://​doi.​org/​10.​1677/​ Biochem Physiol -Part A Mol Integr Physiol 207:21–29 JOE-​08-​0415 Lawrence MJ, Eliason EJ, Brownscombe JW, Gilmour KM, Dingemanse NJ, Barber I, Wright J, Brommer JE (2012) Quanti- Mandelman JW, Gutowsky LFG, Cooke SJ (2018a) tative genetics of behavioural reaction norms: Genetic cor- Influence of supraphysiological cortisol manipula- relations between personality and behavioural plasticity vary tion on predator avoidance behaviors and physiological across stickleback populations. J Evol Biol 25:485–496 responses to a predation threat in a wild marine teleost Doyon C, Leclair J, Trudeau VL, Moon TW (2006) Corticotro- fish. Integr Zool 13:206–218 pin-releasing factor and neuropeptide Y mRNA levels are Vol:. (1234567890) 13 Fish Physiol Biochem Lawrence MJ, Godin JGJ, Cooke SJ (2018b) Does experi- of Wild Bluegill Captured Inside and Outside of a Long- mental cortisol elevation mediate risk-taking and anti- Standing Aquatic Protected Area. Trans Am Fish Soc predator behaviour in a wild teleost fish? Comp. Bio- 149:174–183 chem Physiol -Part A Mol Integr Physiol 226:75–82 Nagrodski A, Murchie KJ, Stamplecoskie KM, Suski CD, Lawrence MJ, Jain-Schlaepfer S, Zolderdo AJ, Algera DA, Cooke SJ (2013) Effects of an experimental short-term Gilmour KM, Gallagher AJ, Cooke SJ (2018c) Are 3 cortisol challenge on the behaviour of wild creek chub minutes good enough for obtaining baseline physiologi- Semotilus atromaculatus in mesocosm and stream envi- cal samples from teleost fish? Can J Zool 96:774–786 ronments. J Fish Biol 82:1138–1158 Lawrence MJ, Godin J-GJ, Zolderdo AJ, Cooke SJ (2019a) Noakes DLG, Jones KMM (2016) Cognition, Learning, and Chronic plasma cortisol elevation does not promote Behavior. In: Schreck CB, Tort L, Farrell AP, Brauner CJ riskier behavior in a teleost fish: a test of the behavioral (eds) Fish Physiology - Biology of Stress in Fish. Aca- resiliency hypothesis. Integr Org Biol 1 demic Press, Amsterdam, vol 35, pp 334–351 Lawrence MJ, Zolderdo AJ, Godin J-GJ, Mandelman JW, O’Connor CM, Gilmour KM, Arlinghaus R, Hasler CT, Philipp Gilmour KM & Cooke SJ (2019b) Cortisol does not DP, Cooke SJ (2010) Seasonal Carryover Effects follow- increase risk of mortality to predation in juvenile blue- ing the Administration of Cortisol to a Wild Teleost Fish. gill sunfish: A manipulative experimental field study. J Physiol Biochem Zool 83:950–957 Exp Zool Part A Ecol Integr Physiol 1–9 Øverli Ø, Kotzian S, Winberg S (2002) Effects of cortisol Lawrence MJ, Eliason EJ, Zolderdo AJ, Lapointe D, Best C, on aggression and locomotor activity in rainbow trout. Gilmour KM, Cooke SJ (2019c) Cortisol modulates Horm Behav 42:53–61 metabolism and energy mobilization in wild-caught Pankhurst NW (2011) The endocrinology of stress in fish: pumpkinseed (Lepomis gibbosus). Fish Physiol Biochem an environmental perspective. Gen Comp Endocrinol 45:1813–1828 170:265–275 Lawrence MJ, Raby GD, Teffer AK, Jeffries KM, Danylchuk Piato AL, Capiotti KM, Tamborski AR, Oses JP, Barcellos AJ, Eliason EJ, Hasler CT, Clark TD, Cooke SJ (2020) LJG, Bogo MR, Lara DR, Vianna MR, Bonan CD (2011) Best practices for non-lethal blood sampling of fish via the Unpredictable chronic stress model in zebrafish (Danio caudal vasculature. J Fish Biol 97:4–15 rerio): Behavioral and physiological responses. Prog Le Moal M (2007) Historical approach and evolution of the Neuro-Psychopharmacology Biol Psychiatry 35:561–567 stress concept: a personal account. Psychoneuroendo- Raoult V, Brown C, Zuberi A, Williamson JE (2012) Blood crinology 32(Suppl 1):S3–S9. https://​doi.​org/​10.​1016/j.​ cortisol concentrations predict boldness in juvenile psyne​uen.​2007.​03.​019 mulloway (Argyosomus japonicus). J Ethol 30:225–232 Lima SL, Dill LM (1990) Behavioral decisions made under Reddy PK, Vijayan MM, Leatherland JF, Moon TW (1995) the risk of predation: a review and prospectus. Can J Zool Does RU486 modify hormonal responses to handling 4:619–640 stressor and cortisol treatment in fed and fasted rainbow MacDougall-Shackleton SA, Bonier F, Romero LM, Moore trout? J Fish Biol 46:341–359 IT (2019) Glucocorticoids and “Stress” Are Not Synony- Ros AFH, Vullioud P, Bshary R (2012) Treatment with the mous. Integr Org Biol 1:1–8 glucocorticoid antagonist RU486 reduces cooperative Manuel R, Gorissen M, Zethof J, Ebbesson LOE, Van De Vis cleaning visits of a common reef fish, the lined bristle- H, Flik G, Van Den Bos R (2014) Unpredictable chronic tooth. Horm Behav 61:37–43 stress decreases inhibitory avoidance learning in Tuebin- Roy B, Rai U (2009) Genomic and non-genomic effect of gen long-fin zebrafish: Stronger effects in the resting cortisol on phagocytosis in freshwater teleost Channa phase than in the active phase. J Exp Biol 217:3919–3928 punctatus: An in vitro study. Steroids 74:449–455 McConnachie SH, Cook KV, Patterson DA, Gilmour KM, Sadoul B, Vijayan MM (2016) Stress and growth. In: Schreck Hinch SG, Farrell AP, Cooke SJ (2012) Consequences CB, Tort L, Farrell AP, Brauner CJ (eds) Fish Physiol- of acute stress and cortisol manipulation on the physiol- ogy - Biology of Stress in Fish. Academic Press, San ogy, behavior, and reproductive outcome of female Pacific Diego. pp 168–190 salmon on spawning grounds. Horm Behav 62:67–76 Sakamoto T, Yoshiki M, Takahashi H, Yoshida M, Ogino Y, McConnachie SH, O’Connor CM, Gilmour KM, Iwama GK, Ikeuchi T, Nakamachi T, Konno N, Matsuda K, Sakamoto Cooke SJ (2012b) Supraphysiological Cortisol Elevation H (2016) Principal function of mineralocorticoid signal- Alters the Response of Wild Bluegill Sunfish to Subsequent ing suggested by constitutive knockout of the mineralo- Stressors. J. Exp Zool Part A Ecol Genet Physiol 317:321–332 corticoid receptor in medaka fish. Sci Rep 6:1–10 Moberg J, Mench GP (2000) The biology of animal stress. Sapolsky RM (2002) Endocrinology of the stress response. In: Basic principles and implications for animal welfare. New Becker JB, Breedlove SM, Crews D, McCarthy MM (eds) York, CABI Pub, Wallingford, Oxon Behavioral Ecology. MIT Press, Cambridge, pp 409–450 Mommsen TP, Vijayan MM, Moon TW (1999) Cortisol in tel- Schjolden J, Basic D, Winberg S (2009) Aggression in rainbow eosts: dynamics , mechanisms of action, and metabolic trout is inhibited by both MR and GR antagonists. Physiol regulation. Rev Fish Biol Fish 9:211–268 Behav 98:625–630 Moore FL, Evans SJ (1999) Steroid hormones use non-genomic Schreck CB, Tort L (2016) Biology of stress in fish, 2016th mechanisms to control brain functions and behaviors: A edn. Academic Press, San Diego review of evidence. Brain Behav Evol 54:41–50 Shaw JR, Gabor K, Hand E, Lankowski A, Durant L, Thibo- Moynes EN, Lawrence MJ, Zolderdo AJ, Prystay TS, Abrams deau R, Stanton CR, Barnaby R, Coutermarsh B, Karlson AEI, Holder P, Cooke SJ (2020) Comparative Behavior KH, Sato JD, Hamilton JW, Stanton BA (2007) Role of Vol.: (0123456789) 13 Fish Physiol Biochem glucocorticoid receptor in acclimation of killifish ( Fundu- Teles M, Tridico R, Callol A, Fierro-Castro C, Tort L (2013) lus heteroclitus ) to seawater and effects of arsenic. Am J Differential expression of the corticosteroid receptors Physiol Integr Comp Physiol 292:R1052–R1060 GR1, GR2 and MR in rainbow trout organs with slow Sopinka NM, Patterson LD, Redfern JC, Pleizier NK, Belanger release cortisol implants. Comp Biochem Physiol - A Mol CB, Midwood JD, Crossin GT, Cooke SJ (2015) Manipu- Integr Physiol 164:506–511 lating glucocorticoids in wild animals: Basic and applied Veillette PA, Serrano X, Garcia MM, Specker JL (2007) Evi- perspectives. Conserv Physiol 3:1–16 dence for the onset of feedback regulation of cortisol in lar- Sopinka NM, Donaldson MR, O’Connor CM, Suski CD, val summer flounder. Gen Comp Endocrinol 154:105–110 Cooke SJ (2016) Stress indicators in fish. In: Schreck CB, Vijayan MM, Leatherland JF (1992) In vivo effects of the ster- Tort L, Farrell AP, Brauner CJ (eds) Fish Physiology - oid analogue RU486 on some aspects of intermediary and Biology of Stress in Fish. Academic Press, San Diego, vol thyroid metabolism of brook charr, Salvelinus fontinalis. J 11, pp 406–436 Exp Zool 263:265–271 Sørensen C, Johansen IB, Øverli Ø (2013) Neural plasticity WendelaarBonga SE (1997) The Stress Response in Fish. and stress coping in teleost fishes. Gen Comp Endocrinol Physiol Rev 77:591–625 181:25–34 Wilson ADM, Godin JGJ (2009) Boldness and behavioral Stolte EH, Van Kemenade BMLV, Savelkoul HFJ, Flik G syndromes in the bluegill sunfish, Lepomis macrochirus. (2006) Evolution of glucocorticoid receptors with differ- Behav Ecol 20:231–237 ent glucocorticoid sensitivity. J Endocrinol 190:17–28. Wood LS, Desjardins JK, Fernald RD (2011) Effects of stress https://​doi.​org/​10.​1677/​joe.1.​06703 and motivation on performing a spatial task. Neurobiol Stolte EH, de Mazon AF, Leon-Koosterziel KM, Jesiak M, Learn Mem 95:277–285 Bury NR, Sturm A, Savelkoul HFJ, van Kemenade BMLV, Flik G (2008) Corticosteroid receptors involved Publisher’s Note Springer Nature remains neutral with regard in stress regulation in common carp, Cyprinus carpio. J to jurisdictional claims in published maps and institutional Endocrinol 198:403–417 affiliations. Takahashi H, Sakamoto T (2013) The role of “mineralocorticoids” in teleost fish: Relative importance of glucocorticoid signal- ing in the osmoregulation and “central” actions of mineralo- corticoid receptor. Gen Comp Endocrinol 181:223–228 Vol:. (1234567890) 13