Relationship between stress, feeding and plasma ghrelin levels in ...

1 downloads 0 Views 115KB Size Report
in Barton and Iwama 1991; Pickering 1993; Pankhurst and Van Der Kraak, 1997). .... radioimmunoassay kit for human ghrelin (Linco Research Inc, St Charles, Missouri) which is ..... Otto B, Tschop M, Heldwein W, Pfeiffer AF, Diederich S. 2004.
Relationship between stress, feeding and plasma ghrelin levels in rainbow trout Oncorhynchus mykiss. N.W. PANKHURST1,2, H.R. KING3 and S.L. LUDKE1 1

Fish Endocrinology Laboratory, Faculty of Science Engineering and IT, James Cook University, Townsville, Queensland 4811, Australia. 3 Salmon Enterprises of Tasmania, P.O. Box 1 Wayatinah, Tasmania 7140, Australia Correspondence: Ned Pankhurst, Science, Environment, Engineering and Technology Executive, Gold Coast Campus, Griffith University, PMB 50 Gold Coast Mail Centre, Queensland 9726, Australia. Email: [email protected] Abstract Sexually immature rainbow trout were acclimated to small volume (1 m3) holding tanks, then exposed to short term stress to examine the relationship between feeding, stress, plasma ghrelin levels and other plasma stress parameters. Plasma ghrelin levels showed an increase 24 h after a single feed, plasma lactate and glucose levels fell over the same period and plasma cortisol levels were low and unchanging. One hour of confinement stress resulted in elevations of plasma cortisol, glucose and lactate, and depression of plasma ghrelin levels. In a separate experiment, 2 h of confinement stress also depressed feeding immediately after stress concomitant with increases in plasma cortisol, lactate and glucose; however, in this case there was no change in plasma ghrelin concentrations. A repeat of the 2 h confinement experiment using fish that had not been acclimated to small volume holding tanks produced a more marked elevation in plasma cortisol and a stronger suppression of feeding poststress but here also there was no change in plasma ghrelin levels. The results of the present study confirm that feeding in rainbow trout is suppressed by confinement stress but also that the effect is transitory in this domesticated stock. Plasma ghrelin levels appear as in other fishes to be modulated by feeding status, and may be influenced by stress, suggesting an orexigenic role for ghrelin in rainbow trout. Keywords: feeding, stress, rainbow trout, ghrelin. Introduction A persistent problem in the domestication and culture of fish is the inhibitory effect that stress imposed by the conditions of husbandry can have on a range of physiological processes, including health, reproductive status, and growth (reviewed in Barton and Iwama 1991; Pickering 1993; Pankhurst and Van Der Kraak, 1997). Inhibitory effects on growth may arise at least in part from interference with the secretion and action of growth hormone (GH) and insulin-like growth factor I (IGF1), and the catabolic effects of stress-induced increases in plasma cortisol (Pickering 1993; Pankhurst and Van Der Kraak 1997; Dyer et al. 2004). However, it is also becoming clear that stress also directly reduces feeding activity, presumably via its effect in suppression of appetite (Schreck et al. 1997). For example, Atlantic salmon Salmo salar parr exposed to acute daily stress showed reduced food intake (McCormick et al. 1998), and husbandry practices such as vaccination result in

1

sustained reduction in food intake and growth in the same species (Sørum and Damsgård 2004). Control of appetite in fish is complex, with interplay occurring among an array of orexigenic (stimulatory) and anorexigenic (inhibitory) neuroendocrine and peripheral endocrine signals. Significant orexigenic factors appear to be neuropeptide Y (NPY), β-endorphin, orexins, galanin and the gastric peptide ghrelin, whereas corticotropin releasing factor (CRF) and CRF-related peptides, cholecystokinin, gastrin, bombesin, glucagon-like peptide, and melanocyte stimulating hormone all have anorexigenic action (reviewed in Lin et al. 2000; Bernier and Peter 2001; Volkoff et al. 2005). The mechanism by which stress modulates these control processes is not entirely clear. The main effect of stress is to induce an elevation of plasma cortisol through activation of the hypothalamo-pituitary-interrenal axis (Sumpter 1997). Cortisol in turn has equivocal effects on appetite in fish. Gregory and Wood (1998) reported inhibitory effects of chronic cortisol administration on food intake in rainbow trout but the same effects were absent in goldfish Carassius auratus (De Pedro et al. 1997; Bernier et al. 2004). Exogenous cortisol did; however, suppress food intake and result in weight loss and suppression in plasma IGF1 levels in channel catfish Ictalurus punctatus (Peterson and Small 2005). The situation is further complicated by the fact that cortisol has negative feedback effects on CRF expression and release (Bernier et al. 1999; Bernier and Peter 2001a) and CRF as noted earlier is in turn, itself a potent anorexigenic factor. One interpretation of this is that elevated cortisol secretion resulting from stress may actually reduce the anorexigenic effect of CRF (Bernier et al. 2004). Recent studies have shown that the gastric peptide ghrelin as well as acting as a GH secretagogue, appears to have widespread orexigenic activity in both mammals and lower vertebrates (reviewed by Unniappan and Peter, 2005). In fishes, ghrelin is a 19-24 amino-acid peptide with sequences for the ghrelin gene now described for goldfish (Unniappan et al. 2002), tilapia Oreochromis niloticus (Parhar et al. 2003), rainbow trout Oncorhynchus mykiss (Kaiya et al. 2005a), Japanese eel Anguilla japonica (Kaiya et al. 2003) and channel catfish Ictalurus punctatus (Kaiya et al. 2005b). Plasma ghrelin levels are elevated in fasted goldfish and both intracerebroventricular (icv) and intraperitoneal (ip) injection of ghrelin stimulate food intake (Unniappan et al. 2004). Fasting in burbot Lota lota does not appear to generate the same increase in plasma ghrelin (Nieminen et al. 2003) but ghrelin levels are elevated in post-spawning growth periods in the same species (Mustonen et al. 2002). Collectively these data suggest that as in mammals, ghrelin is a significant orexigenic regulator of appetite in fish. The effect of stress on ghrelin secretion or action in fish is currently not known, but in humans, both exogenous corticosteroids and hypercortisolism resulting from disease states, suppress plasma ghrelin levels (Otto et al. 2004). Similarly, increases in plasma ghrelin following adrenalectomy in rats can be reversed by treatment with glucocorticoids (Proulx et al. 2005), suggesting that stress-suppression of ghrelin in fish might also occur. In the present study, the possible effect of feeding and stress on plasma levels of ghrelin-like immunoreactivity (hereafter termed ghrelin) was assessed in rainbow trout using a heterologous assay for human ghrelin. Stress was characterised in terms of previously established values for cortisol, lactate and glucose for this stock of rainbow trout (Thomas et al. 1999). For an initial set of experiments, fish were acclimated to small holding tanks (1m3) to maximise the likelihood that subsequent feeding experiments in small volume arenas were likely to be reflective of nonstressful holding conditions. To address the possibility that acclimation also

2

diminished the subsequent response to confinement, a final experiment was conducted using stock that had not previously been acclimated to small tank holding conditions. Methods Fish Source and Husbandry Fourteen-month old rainbow trout (mean weight 307 g) were sourced from the University of Tasmania, School of Aquaculture (Launceston, Tasmania) cultured stock line and acclimated in 1 m3 tanks at 24 fish/tank for 3 months prior to experiment. Tanks were supplied with recirculating water passed through solids settling and bio-filters with foam fractionation, and water change of ~ 20% of the system volume each week. Heat-chill units on each tank kept water temperature at 12 ± 0.5o C throughout. A series of 3 experiments was conducted using this stock in late austral spring. A final experiment was performed the following winter using 2-year old fish (mean weight 750 g) from the same stock but transferred to 400-l aerated plastic bins 1-2 days before experiment and held at ambient temperature (6o C). Fish were fed daily at 0900 h with a commercial salmon diet (Nutra HP, Skretting, Australia) until the day before experiment unless otherwise noted. Experimental Protocols 1. Effect of feeding on plasma ghrelin levels. Fish were distributed among four 1 m3 tanks at 12 fish/tank, and fed as normal at ~1% body weight at 0900 h, then at 0.5, 1, 3 and 6 h after feeding, 7 fish were removed from each of the tanks in succession (to avoid serial disturbance), anaesthetised in 0.05% 2-phenoxy ethanol (Sigma Aldrich) and bled by caudal puncture using pre-heparinised syringes and 22G needles. At 24 h post-feeding, a final sample was taken using 7 fresh fish combined from 3 of the tanks and that remained undisturbed since the removal of fish at 6 h. Blood was placed on ice for short term storage and transfer to the laboratory where samples were centrifuged at 10000 g for 3 mins at 4o C, plasma removed and treated as described under ‘plasma analyses’. 2. Effect of stress and recovery on plasma constituents Seven fish were sampled from a 1 m3 tank at 0900 h and anaesthetised and bled as before (pre-stress sample). Thirty-five additional fish were netted at the same time and distributed between 2 aerated 40-l bins for 1 h of confinement stress. After 1 h, 7 fish were netted directly from the confinement bin and bled (0 h post-stress), and the remaining fish distributed to 1m3 tanks at 7 fish/tank for recovery. Fish were then sampled as before at 2, 6, 24 and 48 h post-stress. All fish were unfed throughout the experiment. 3. Effect of stress on feeding in acclimated stock Seven fish from a 1 m3 tank were hand fed at 0900 h at 20 pellets/fish (based on previously measured intakes of ~12 pellets/fish under low disturbance conditions) and left for 1-2 mins after the addition of the last pellets. Fish were then netted, anaesthetised and bled (pre-stress sample) and the number of uneaten pellets counted. Another 28 fish were transferred to two 40-l confinement bins (14 fish/bin) for 2h after which they were distributed to 1m3 tanks at 7 fish/tank for recovery. At 0, 1, 2

3

and 4 h post-stress, fish were fed, netted and bled and uneaten pellets in the tank counted. 4. Effect of stress on feeding in non-acclimated stock Stress responses in earlier experiments were relatively modest (see results) suggesting that extended acclimation to 1 m3 tanks may have limited the subsequent response to confinement. In a repeat of experiment 3, fish were transferred from a 4 m3 stock tank directly to 2 aerated 400-l bins the day before experiment. Seven fish in one bin were hand fed (20 pellets/fish) at 0900 h the next morning, allowed to feed then netted and sampled as before (pre-stress sample). Twenty-one fish from the second 400-l bin were transferred to a 40-l confinement bin for 2 h after which 7 fish were transferred back to the larger bin, fed and sampled (0 h post-stress sample). The remaining fish were allowed to recover at 7 fish/bin in the 400-l bins, then fed and sampled at 1 and 3 h post-stress. Plasma Analyses Plasma ghrelin levels were measured with a heterologous assay using a 125Iradioimmunoassay kit for human ghrelin (Linco Research Inc, St Charles, Missouri) which is specific for biologically active (octanoylated) ghrelin. Plasma samples were stabilised immediately by addition of 50µl ml-1 plasma 1N HCl, and 10µl ml-1 plasma of 10mg ml-1 phenylmethylsulfonyl fluoride (Sigma Aldrich) in 100% methanol, prior to freezing for transport to the laboratory. Heparin levels in plasma samples were kept as recommended at less than 5 U ml-1 to avoid false assay positives. The assay was validated using serial dilution of plasma which in all cases gave parallel displacement to that of assay standards. As an additional check, plasma validation was also performed on plasma from Atlantic salmon, and here also, plasma dilutions gave parallel displacement to assay standards (Figure 1). Inter-assay variation was assessed using repeat measurements of quality control standards provided in the kit and was (%CV[n]) 3.5[5]. Plasma cortisol levels were measured by RIA using the reagents and protocol described in Pankhurst and Sharples (1992) for experiments 1-3, and the same protocol but an antibody purchased from Bioclone Australia Pty. Ltd for experiment 4. Plasma was extracted with ethyl acetate prior to assay, solvent extracts dried down in assay tubes then reconstituted with assay buffer. Extraction efficiency was estimated from recovery of 3H-cortisol from aliquots of pooled plasma samples and assay values were corrected accordingly. Inter-assay variability measured using aliquots of a pooled internal standard was (%CV[n]) 10.9[5]. Plasma levels of lactate and glucose were determined spectrophotometrically using diagnostic kits (Trinity Biotech, Wicklow, Ireland, and ThermoElectron, Melbourne, Australia for lactate and glucose respectively) and were validated against standard curves for both assays. Statistics Comparisons between means were made using one-way ANOVA and subsequent Tukeys b tests. Where necessary, data were square root, or log transformed to satisfy the requirements for homogeneity of variances. Feeding data were assessed using a Chi-squared test. All analyses were conducted using the SPSS (version 12.1) statistical package.

4

Results Fish fed a single meal at 0900 h showed a depression in plasma glucose levels at 24 h post-feeding relative to 1 and 3 h, and a steady decline in plasma lactate levels with time after feeding (Figure 2). Plasma cortisol levels were low throughout and showed no change with time since feeding, whereas plasma ghrelin levels were elevated at 24 h post-feeding relative to 0.5 and 6 h (Figure 2). Both plasma glucose and lactate levels rose in fish exposed to 1 h of confinement stress, with a more sustained increase occurring in plasma glucose than lactate (Figure 3). Plasma cortisol levels were significantly elevated immediately following, and 2 h post-stress, but showed evidence of recovery as early as 6 h post-stress (Figure 3). Plasma ghrelin levels were significantly depressed 2 h post-stress relative to the prestress sample (Figure 3). Fish acclimated to 1m3 holding tanks, exposed to 2 h confinement stress then fed showed a significant depression in feeding but with recovery to pre-feeding levels by 1 h post-stress (Table I, experiment 3). Plasma glucose levels were elevated at all times post-stress, and plasma lactate at 0 and 1 h post-stress (Figure 4). Plasma cortisol levels were also elevated immediately post-stress, whereas there was no significant change in plasma ghrelin values (Figure 4). Unacclimated stock exposed to 2 h of confinement showed a very marked depression of feeding at 0 and 1 h post-stress but had resumed feeding by 3 h (Table I, experiment 4). Plasma glucose levels were elevated at 1 h post-stress but there was no change in plasma lactate over time (Figure 5). Plasma cortisol levels were elevated at 0 and 1 h post-stress whereas plasma ghrelin levels did not change (Figure 5). Discussion Rainbow trout in the present study showed plasma ghrelin levels of similar magnitude to those recorded in burbot (Nieminen et al. 2003) and goldfish (Unniappan et al. 2004), with absolute levels appearing to be labile in response to feeding. Ghrelin levels recorded in the present study rose in rainbow trout fasted for 24 h, consistent with the view that ghrelin is an orexigen in fish (Unniappan and Peter 2005) and other vertebrates (Cummings and Shannon 2003). Goldfish displayed a similar relationship between feeding and plasma ghrelin levels with falls in plasma values occurring 1 and 3 h post-feeding, and starvation-induced increases in both ghrelin gene expression and plasma ghrelin levels occurring after 3-7 d of starvation (Unniappan et al. 2004). Fasting for 24h also increased plasma ghrelin levels in Japanese quail Coturnix coturnix japonica (Shousha et al. 2005), and humans show peaks in plasma ghrelin levels before anticipated mealtimes, and decreases after meals (Cummings and Shannon 2003). In contrast to the present study, starvation for 7 d had no effect in preproghrelin expression in stomachs of tilapia (Parhar et al. 2003), and in burbot held at 2o C, a 2 week fast resulted in falls in plasma ghrelin levels. The effect was not present in burbot held at 10o C (Nieminen et al. 2003). Behavioural effects of icv, ip or systemic injection occur within hours in fish (Unniappan et al. 2002, 2004), Japanese quail (Shousha et al. 2005) and humans (Cummings and Shannon 2003), consistent with the view that orexigenic signals indicative of impending meals are likely to be rapidly induced and relatively short-lived (Cummings and Shannon 2003). The failure of the

5

studies by Parhar et al. (2003) and Nieminen et al. (2003) to detect stimulatory effects of food deprivation on plasma ghrelin levels or ghrelin gene expression may have resulted from measurement too long after imposition of the stimulus of removal of food. The results of the present study support the view that plasma ghrelin levels in fish will typically change over relatively short periods in relation to presentation of a meal. Examination of plasma stress parameters in rainbow trout over a 24 h fast in the present study showed very low and unchanging plasma cortisol levels, and plasma glucose and lactate levels that peaked soon after feeding, and then fell as the meal was digested. Plasma levels of cortisol, glucose and lactate were typical of those seen in unstressed fish from this population of rainbow trout (Thomas et al. 1999), indicating that fish had acclimated to the test tanks, and that the observed changes in plasma ghrelin levels were unlikely to have resulted from stress or handling effects. Fish exposed to 1 h of confinement stress showed a moderate but sustained increase in plasma cortisol levels and shorter term increases in plasma glucose and lactate levels. Changes in plasma glucose result both from initial catecholamine-mediated glycogenolysis, followed by more sustained cortisol-mediated gluconeogenesis (reviewed in Begg and Pankhurst, 2004). These patterns are typical of stress responses in this population, albeit with stress-induced increases in plasma cortisol being at the lower end of the range of values recorded in other studies on rainbow trout (Thomas et al. 1999). The apparently dampened stress response may have resulted from prior acclimation of fish to relatively small (1 m3) holding tanks rather than the 4m3 tanks used by Thomas et al. (1999). Plasma ghrelin levels were depressed in concert with peak concentrations of plasma cortisol and glucose in the present study, indicative of possible stresssuppression of ghrelin secretion in rainbow trout. Exposure to 2 h of confinement stress with the additional assessment of effect on feeding generated similar magnitude and duration of increases in plasma glucose and lactate but only a transitory increase in cortisol, and no significant effect on plasma ghrelin levels. Feeding was suppressed but only immediately after confinement and there was very rapid behavioural recovery. A repeat of the experiment using non-acclimated fish generated more sustained and greater increases in plasma cortisol more typical of stress responses in this stock (Thomas et al. 1999), but here also there was no change in plasma ghrelin levels despite a more marked suppression of feeding and slower behavioural recovery. As noted earlier, there is currently no other information on the effect of stress on plasma ghrelin levels in fish, but in both humans and rats, artificial or disease-induced increases in plasma levels of glucocorticoids suppress plasma ghrelin levels with return to normal following removal of the exogenous stimulus, or adrenalectomy (Otto et al. 2004; Proulx et al. 2005). The present study provides possible evidence that the same can occur in fish but the effect does not appear to be as consistent as in higher vertebrates. An additional complication of the present study was variable pre-stress plasma levels of ghrelin. The reason for this is not known but may relate to undetected variation in feeding activity prior to experiment, based on the demonstration in the present study of increased ghrelin levels with increasing time since last feed. Stress consistently suppresses appetite in fish, with subsequent reduction in growth and condition if the stress is sustained or repeated with high enough frequency (Schreck et al. 1997; McCormick et al. 1998; Sloman et al. 2000). The mechanisms involved are not entirely clear; however, administration of exogenous cortisol consistently reduced growth in rainbow trout (Barton et al. 1987; Gregory and Wood

6

1998), and channel catfish Ictalurus punctatus (Davis et al. 1985). In a later study on catfish, cortisol-induced suppression of growth was accompanied by reduced plasma concentrations of IGF1 (Peterson and Small, 2005). In studies where feeding behaviour was measured, cortisol also reduced food intake in both rainbow trout (Gregory and Wood 1998), and channel catfish (Peterson and Small 2005) indicating that the growth and condition effects were not simply the result of catabolic effects of cortisol on metabolism or reductions in plasma IGF1 levels. A contrasting effect was reported from goldfish where a high dose of cortisol delivered in food had no effect on feeding, but a low dose increased food intake (Bernier et al. 2004). This was accompanied by down-regulation of CRF gene expression, and increased NPY gene expression in the preoptic area of the brain. The appetite promoting effects of moderate doses of cortisol were interpreted as a decrease in the anorexogenic CRF signal, and promotion of the orexigenic NPY signal (Bernier et al. 2004). This is supported by earlier studies showing that CRF gene expression in the preoptic area of goldfish brains could be suppressed by treatment with ip cortisol implants (Bernier et al. 1999), and that icv injection of CRF reduced food intake (Bernier and Peter 2001b). The present study confirms the presence of an immunoreactive ghrelin-like compound in the plasma of rainbow trout and shows that levels change in relation to time since feeding. There is equivocal evidence for an inhibitory effect of stress on plasma ghrelin levels despite stress having a suppressive effect on feed intake for short periods in this domesticated stock. This suggests that while ghrelin is probably an orexigen in trout as in other vertebrates, inhibition of feeding in stressed fish is regulated by a range of factors in addition to plasma levels of ghrelin. Acknowledgments Thanks are extended to Ryan Longland for assistance with fish maintenance and handling, to the University of Tasmania for access to fish and holding facilities, to Suraj Unniappan for advice on the ghrelin assay, and to Richard Peter, University of Alberta for helping us articulate the problem. This study was supported by an Australian Research Council Industry Linkage Grant awarded to NWP, and direct funding from Salmon Enterprises of Tasmania. References Barton BA, Iwama GK. 1991. Physiological changes in fish from stress in aquaculture with emphasis on the response and effects of corticosteroids. Ann. Rev. Fish Dis. 1: 3-26. Barton BA, Schreck CB, Barton LD. 1987. Effects of chronic cortisol administration and daily acute stress on growth, physiological conditions, and stress response in juvenile rainbow trout. Dis. Aquat. Org. 2: 173-185. Begg K, Pankhurst NW. 2004. Endocrine and metabolic responses to stress in a laboratory population of the tropical damselfish Acanthochromis polyacanthus. J. Fish Biol. 64: 133-145. Bernier NJ, Peter RE. 2001a. The hypothalamic-pituitary-interrenal axis and the control of food intake in teleost fish. Comp. Biochem. Physiol. 129B: 639644. Bernier NJ, Peter RE. 2001b. Appetite-suppressing effects of urotensin I and corticotrophin-releasing hormone in goldfish (Carassius auratus). Horm. Behav. 73: 248-260.

7

Bernier NJ, Bedard N, Peter RE. 2004. Effects of cortisol on food intake, growth, and forebrain neuropeptide Y and corticotropin-releasing factor gene expression in goldfish. Gen. Comp. Endocrinol. 135: 230-240. Bernier NJ, Lin X-W, Peter RE. 1999. Differential expression of corticotropinreleasing factor (CRF) and urotensin I precursor genes, and evidence of CRF gene expression regulated by cortisol in goldfish brain. Gen. Comp. Endocrinol. 116: 461-477. Cummings DE, Shannon MH. 2003. Roles for ghrelin in the regulation of appetite and body weight. Archiv. Surg. 138: 389-396. Davis KB, Torrance P, Parker NC, Suttle MA. 1985. Growth, body composition and hepatic tyrosine aminotransferase activity in cortisol-fed channel catfish, Ictalurus punctatus Rafinesque. J. Fish Biol. 27: 177-184. De Pedro N, Alonso-Gomez AL, Gancedo B, Valenciano AI, Delgado MJ, AlonsoBedate M. 1997. Effect of α-helical-crf(9-41) on feeding in goldfish: involvement of cortisol and catecholamines. Behav. Neurosci. 111: 398-403. Dyer AR, Upton Z, Stone D, Thomas PM, Soole KL, Higgs, N, Quinn K, Carragher JF. 2004. Development and validation of a radioimmunoassay for fish insulinlike growth factor I (IGF-I) and the effects of aquaculture stressors on circulating IGF-I levels. Gen. Comp. Endocrinol. 135: 268-275. Gregory TR, Wood CM. 1998. The effects of chronic plasma cortisol elevation on the feeding behaviour, growth, competitive ability and swimming performance of juvenile rainbow trout. Physiol. Biochem. Zool. 72: 286-295. Kaiya H, Kojima M, Hosoda H, Riley LG, Hirano T, Grau EG, Kangawa K. 2003. Amidated fish ghrelin purification, cDNA cloning in the Japanese eel and its biological activity. J. Endocrinol. 176: 415-423. Kaiya H, Kojima M, Hosoda H, Moriyama S, Takahashi A, Kawauchi H, Kangawa K. 2005. Peptide purification, complementary deoxyribonucleic acid (DNA) and genomic DNA cloning, and functional characterization of ghrelin in rainbow trout. Endocrinology 144: 5215-5226. Kaiya H, Small BC, Bilodeau AL, Shepherd BS, Kojima M, Hosoda H, Kangawa K. 2005. Purification, cDNA cloning, and characterization of ghrelin in channel catfish, Ictalurus punctatus. Gen. Comp. Endocrinol. 143: 201-210. Lin X-W, Volkoff H, Narnaware Y, Bernier NJ, Peyon P, Peter RE. 2000. Brain regulation of feeding behavior and food intake in fish. Comp. Biochem. Physiol. 126A: 415-434. McCormick SD, Shrimpton JM, Carey JB, O’Dea MF, Sloan KE, Moriyama S, Björnsson BTh. 1998. Repeated acute stress reduces growth rate of Atlantic salmon parr and alters plasma levels of growth hormone, insulin-like growth factor I and cortisol. Aquaculture 168: 221-235. Mustonen A-M, Nieminen P, Hyvärinen H. 2002. Leptin, ghrelin, and energy metabolism of the spawing burbot (Lota lota L.). J. Exp. Zool. 293: 119-126. Nieminen P, Mustonen A-M, Hyvärinen H. 2003. Fasting reduces plasma leptin- and ghrelin-immunoreactive peptide concentrations of the burbot (Lota lota) at 2oC but not at 10oC. Zool. Sci. 20: 1109-1115. Otto B, Tschop M, Heldwein W, Pfeiffer AF, Diederich S. 2004. Endogenous and exogenous glucocorticoids decrease plasma ghrelin in humans. Eur. J. Endocrinol. 151: 113-117. Pankhurst NW, Sharples DF. 1992. Effects of capture and confinement on plasma cortisol levels in the snapper Pagrus auratus. Aust. J. Mar. Freshwat. Res. 43: 345-356.

8

Pankhurst NW, Van Der Kraak G. 1997. Effects of stress on growth and reproduction. In: Iwama GK, Pickering AD, Sumpter JP, Schreck CB, editors. Fish Stress and Health in Aquaculture. Cambridge: Cambridge University Press. pp. 73-93. Parhar IS, Sato H, Sakuma Y. 2003. Ghrelin gene in cichlid fish is modulated by sex and development. Biochem. Biophys. Res. Comm. 305: 169-175. Peterson BC, Small BC. 2005. Effects of exogenous cortisol on the GH/IGF/IGFBP network in channel catfish. Dom. Anim. Endocrinol. 28: 391-404. Pickering AD. 1993. Growth and stress in fish production. Aquaculture 111: 51-63. Proulx K, Vahl TP, Drazen DL, Woods SC, Seeley RJ. 2005. The effect of adrenalectomy on ghrelin secretion and orexigenic action. J. Neuroendocrinol. 17: 445-451. Schreck CB, Olla BL, Davis MW. 1997. Behavioral responses to stress. In: Iwama GK, Pickering AD, Sumpter JP, Schreck CB, editors. Fish Stress and Health in Aquaculture. Cambridge: Cambridge University Press. pp. 145-170. Shousha S, Nakahara K, Kojima M, Miyazato M, Hosoda H, Kangawa K, Murakami N. 2005. Different effects of peripheral and central ghrelin on regulation of food intake in the Japanese quail. Gen. Comp. Endocrinol. 141: 178-183. Sloman KA, Gilmour KM, Taylor AC, Metcalfe NB. 2000. Physiological effects of dominance hierarchies within groups of brown trout, Salmo trutta, held under simulated natural conditions. Fish Physiol. Biochem. 22: 11-20. Sørum U, Damsgård B. 2004. Effects of anaesthetisation and vaccination on feed intake and growth in Atlantic salmon (Salmo salar L.). Aquaculture 232: 333341. Sumpter JP. 1997. The endocrinology of stress. In: Iwama GK, Pickering AD, Sumpter JP, Schreck CB, editors. Fish Stress and Health in Aquaculture. Cambridge: Cambridge University Press. pp. 95-118. Thomas PM, Pankhurst NW Bremner HA. 1999. The effect of stress and exercise on post-mortem biochemistry of Atlantic salmon (Salmo salar) and rainbow trout (Oncorhynchus mykiss). J. Fish Biol. 54: 1177-1196. Unniappan S, Peter RE. 2005. Structure, distribution and physiological functions of ghrelin in fish. Comp. Biochem. Physiol. 140A: 396-408. Unniappan S, Canosa LF, Peter RE. 2004. Orexigenic actions of ghrelin in goldfish: Feeding-induced changes in brain and gut mRNA expression and serum levels, and responses to central and peripheral injections. Neuroendocrinology 79: 100-108. Unniappan S, Lin X-W, Cervini L, Rivier J, Kaiya H, Kangawa K, Peter RE. 2002. Goldfish ghrelin: Molecular characterization of the complementary deoxyribonucleic acid, partial gene structure and evidence for its stimulatory role in food intake. Endocrinology 143: 4143-4146. Volkoff H, Canosa LF, Unniappan S, Cerdá-Reverter JM, Bernier NJ, Kelly SP, Peter RE. 2005. Neuropeptides and the control of food intake in fish. Gen. Comp. Endocrinol. 142: 3-19.

9

Table I. Proportion of feed pellets eaten by fish (n = 7 per treatment) acclimated to 1 m3 holding tanks (experiment 3) or 1 day following transfer from a 4 m3 stock tank (experiment 4) before and after exposure to 2h confinement stress. Treatment stress) Pre0h 1h 2h 3h 4h

(time

post Proportion eaten (%)* Experiment 3 92.7 40.0 84.3 88.6 100

Chi-squared value 35.5, P