Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Corticosteroid receptor dynamics and smolting in hatchery-reared and wild coho salmon (Oncorhynchus kisutch) Shrimpton, James M. 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_spring_phd_shrimpton_james.pdf [ 6.68MB ]
JSON: 831-1.0098828.json
JSON-LD: 831-1.0098828-ld.json
RDF/XML (Pretty): 831-1.0098828-rdf.xml
RDF/JSON: 831-1.0098828-rdf.json
Turtle: 831-1.0098828-turtle.txt
N-Triples: 831-1.0098828-rdf-ntriples.txt
Original Record: 831-1.0098828-source.json
Full Text

Full Text

Corticosteroid Receptor Dynamics and Smolting inHatchery-Reared and Wild Coho Salmon(Oncorhynchus kisutch)byJames Mark ShrimptonB.Sc. (Honours) University of Victoria, 1985M.Sc. University of British Columbia, 1988A Thesis Submitted in Partial Fulfilment of the Requirements forthe Degree of Doctor of PhilosophyinThe Faculty of Graduate Studies(Department of Zoology)We accept this-th sis as conforming to the required standardThe University of British ColumbiaSeptember 1993© J. Mark Shrimpton, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission. (Signature)Department ofThe University of British ColumbiaVancouver, CanadaDate ^al t().1 3 DE-6 (2/88)ABSTRACTSeasonal changes in corticosteroid receptors (CR) in the gills of coho salmon(Oncorhynchus kisutch) were examined to determine what effect smolting and rearingenvironment had on gill tissue sensitivity to cortisol. CR concentration and affinity werefound to change seasonally with the increases in gill Na+K+ATPase activity and thedevelopment of saltwater tolerance. At times of the year when fish showed increasedsaltwater tolerance, the gill CR concentration and affinity decreased. The decline inreceptor numbers and affinity was concurrent with increases in plasma cortisolconcentration. Endocrine control of CR by cortisol and growth hormone (GH) wasexamined. Cortisol downregulated CR in the gills. Acute administration of cortisolresulted in a reduction in CR numbers for 72 hr with no change in affinity. Chroniccortisol treatment resulted in a decrease in CR concentration and affinity The change inaffinity occurred only while plasma cortisol levels remained elevated, but CR populationremained significantly reduced for at least 10 days following cessation of hormonetreatment. Repeated handling stresses resulted in a similar reduction in CR numbers, butwithout an apparent change in affinity. The chronic or repeated elevation in plasmacortisol downregulates the sensitivity of the gills to cortisol by a persistent reduction in CRconcentration, despite the return to non-stress levels of circulating cortisol. In contrast tocortisol, Gil upregulated CR in the gills. Two bovine hormones were used in the study,growth hormone (bGH) and placental lactogen (bPL). These hormones caused a dosedependent increase in concentration of CR and Na+K+ATPase activity in the gills. Theupregulation of CR by bPL and bGH enhanced the gill sensitivity to cortisol and maypartially account for the greater saltwater tolerance exhibited by the Gil treated fish.The effect of rearing environment on cortisol dynamics and Gil changes wereexamined in juvenile hatchery and wild coho salmon over the spring when the fish weresmolting. Plasma cortisol levels showed an increase in concentration during the spring inall groups. The rise in cortisol concentration, however, was significantly greater in the wildsmolts than the hatchery smolts. Timing of maximal Gil levels in the plasma during thespring were similar to the surge in cortisol. The absolute differences between the groupsiiiand sampling times during the spring, however, were much smaller than those observed forcortisol. CR concentration and affinity decreased during the spring. The wild fishconsistently possessed the greatest number of CR. The change in affinity was similar forthe hatchery and wild fish during the spring of 1991. In 1992, the hatchery fish also showeda gradual increase in the dissociation constant (kr,). In contrast, the wild fish did not showan increase in dissociation constant until May. The changes in cortisol concentration in theplasma and the CR dissociation constant occurred synchronously with the increase inNa+K+ATPase activity of the wild fish. The wild fish showed the greatest increase in kmplasma cortisol concentration and Na+1C+ATPase activity. Although the hatchery fish weremuch larger than their wild or colonized counterparts, they consistently showed a reducedsaltwater tolerance as assessed by a much greater perturbation in plasma sodiumconcentration following transfer to saltwater. Within each group there was no relationshipbetween the size of the fish and saltwater tolerance. Following transfer to salt water, thehatchery fish showed a significant increase in circulating plasma cortisol concentration andchanges in plasma GH level that were associated with the osmotic stress experienced by thefish. These changes were not seen in the wild smolts. The hatchery fish possess fewerchloride cells and lower specific activities of the enzymes Na+IVATPase and citratesynthase. The weaker osmoregulatory ability of the hatchery fish was associated with agreater mortality following abrupt transfer to 35 %o sea water.The research presented indicates that in coho salmon, gill CR change seasonally.CR concentration reflects smolting and development of saltwater tolerance. The affinityand concentration of CR are regulated by levels of cortisol and growth hormone in theplasma. Rearing conditions that cause a change in these hormone levels also affects thesensitivity of the gills to cortisol. Rearing environment, therefore, has a large effect oncorticosteroid receptor dynamics and smolting. The changes in CR that are associated withrearing environment contribute to the differences in saltwater tolerances seen between thehatchery and wild coho smolts.ivTABLE OF CONTENTSPageAbstract^ iiTable of Contents^ ivList of Tables viList of Figures^ viiList of Abbreviations xiAcknowledgements^ 3diChapter 1 General Introduction 1Chapter 2 Seasonal changes in gill corticosteroid receptors and smolting in^8juvenile coho salmonIntroduction^ 9Material and Methods^ 11Results 16Discussion 25Chapter 3 Downregulation of corticosteroid receptors in the gills of coho salmon^31due to stress and cortisol treatmentIntroduction^ 32Materials and Methods^ 34Results 37Discussion 50Chapter 4 Increased gill corticosteroid receptor concentration and saltwater^55tolerance in juvenile coho salmon treated with growth hormone andplacental lactogenIntroduction^ 56Materials and Methods^ 58Results 60Discussion 71VChapter 5 Changes in cortisol dynamics in wild and hatchery reared juvenile^77coho salmon during smoltificationIntroduction^ 78Materials and Methods^ 80Results 85Discussion 97Chapter 6 Changes in growth hormone concentration during smolting and^104following saltwater transfer in wild and hatchery-reared juvenile cohosalmonIntroduction^ 105Materials and Methods^ 106Results 107Discussion 111Chapter 7 Saltwater tolerance in wild and hatchery-reared juvenile coho salmon^116smoltsIntroduction^ 117Materials and Methods^ 119Results 123Discussion 136Chapter 8 General Discussion^ 142Literature Cited 147viLIST OF TABLESPageTable 2.1^Plasma levels of cortisol, sodium and haematocrit, and characteristics^49of gill corticosteroid receptors from control and stressed juvenile cohosalmon.Table 7.1 Mean weight and length of the wild, colonized and hatchery Quinsam^126River coho salmon smolts used in the saltwater challengeexperiments.LIST OF FIGURESviiPageFigure 2.1Figure 2.2Figure 2.3Figure 2.4Figure 2.5Figure 2.6Figure 2.7Figure 3.1Figure 3.2Figure 3.3Figure 3.4Figure 3.5The size distribution of the fish sampled during the year.^18The seasonal change in plasma cortisol concentration in three size^19groups of juvenile coho salmon.The seasonal change in gill corticosteroid receptor concentration^20(Bmax) and dissociation constant (IrD) for coho salmon.The relationship between B. and lcD, and the size of the fish^21sampled in November.The relationship between Bina, and lcD, and the size of the fish^22sampled in May.Seasonal changes in gill Na+IVATPase activity of juvenile coho^23salmon.Plasma sodium concentration after 24 hr saltwater challenge for coho^24in May as a function of fish length.Changes in plasma cortisol levels (n = 7), and corticosteroid receptor^40(n = 5) concentration (B.) and affinity (1cD) in the gills following asingle intraperitoneal injection of cortisol-21-hemisuccinate.Plasma sodium concentrations (n = 7) following an intraperitoneal^41injection of cortisol.Verification of the effectiveness of mini osmotic pumps filled with^42cortisol-21-hemisuccinate implanted into the peritoneal cavity atelevating plasma cortisol concentration over 18 days.Scatchard plot of [3H]TA bound to gill homogenate from coho^43salmon implanted with osmotic pumps for 17 days.Changes in plasma cortisol levels, and corticosteroid receptor^44concentration (B..) and affinity (1cD) in the gills of coho implantedwith mini osmotic pumps containing high and low doses of cortisol-21-hemisuccinate or vehicle only.Figure 3.6Figure 3.7Figure 3.8Figure 3.9Figure 4.1Figure 4.2Figure 4.3Figure 4.4Figure 4.5Figure 4.6Figure 4.7Figure 4.8viii^Plasma sodium concentrations of fish chronically exposed to elevated^45plasma cortisol.The relationship between plasma cortisol concentration and^46corticosteroid receptor concentration in the gills of coho salmon.The relationship between corticosteroid receptor concentration (B.)^47and dissociation constant (1cD) for plasma cortisol concentrationsbelow 300 ng/ml.The concentration of gill corticosteroid receptors (B„,..) from juvenile^48coho salmon, 3 days and 6 days after fish had been subjected to adaily handling stress for 10 days.The size of the juvenile coho following six weekly treatments of^63bovine placental lactogen (bPL) and bovine growth hormone (bGH).The control groups consisted of a group injected with bovine serumalbumin (bSA) and a group that was not injected.The plasma cortisol concentration of juvenile coho salmon injected^64once (a) or six times (b) with bPL or bGH.The effect of a single injection of bPL or bGH on corticosteroid^65receptor concentration (B..), affinity (kr,), and Hill coefficient in thegills of coho salmon.The effect of six weekly injections of bPL or bGH on corticosteroid^66receptor concentration (B.), affinity (I(D), and Hill coefficient in thegills of coho salmon.The activity of Na+1VATPase in the gills of coho salmon treated with^67bPL or bGH for one (a) or six (b) weekly treatments.Plasma sodium concentrations one week following a single injection^68of bPL or bGH.Plasma sodium concentrations after six weekly treatments with bPL or^69bGH in fresh water and after exposure to salt water for 24 hr.The relationship between the length of the fish and plasma sodium^70concentration (a) in fresh water and (b) following a 24 hr saltwaterchallenge.ix^Figure 5.1 The weight of the juvenile wild and hatchery-reared coho salmon^89caught during 1991 and 1992.^Figure 5.2 The change in condition factor over the spring of 1991 and 1992 for^90wild and hatchery-reared coho salmon.Figure 5.3 Plasma sodium concentrations during the spring for hatchery and wild^91coho salmon.Figure 5.4 The change in concentration of cortisol in the plasma of juvenile^92coho salmon from the Quinsam River over the spring.Figure 5.5 Half-life of cortisol in the plasma of juvenile coho salmon from the^93Quinsam River during the spring.Figure 5.6 Concentration of corticosteroid receptors in the gills of coho salmon^94during the spring of 1991 and 1992, from the Quinsam River.Figure 5.7 Dissociation constant of the corticosteroid receptors in the gills of^95coho salmon during the spring.Figure 5.8 Changes over the spring in the activity of Na+K+ATPase activity in^96the gills of juvenile hatchery and wild coho from the Quinsam River.Figure 5.9 The relationship between gill Na+K+ATPase activity and cortisol^97concentration in the plasma of juvenile hatchery and wild cohosalmon.Figure 6.1 Plasma GH concentrations of juvenile coho salmon during the spring^109of 1991 and 1992.Figure 6.2 Plasma GH concentrations of hatchery and wild coho smolts before^110and after transfer to salt water for one and seven days.Figure 7.1 Plasma sodium concentration for smolting juvenile coho salmon^127following direct transfer to salt water.Figure 7.2 The relationship between plasma sodium concentration after 24 hr in^128sea water and the weight of the fishFigure 7.3 The change in haematocrit following transfer to salt water for^129hatchery, wild and colonized smolts in 1990.xFigure 7.4Figure 7.5Figure 7.6Figure 7.7Figure 7.8Figure 7.9Circulating plasma cortisol concentrations of hatchery and wild coho^130smolts before and after transfer to salt water.Specific activity of Na+1C+ATPase in the gills of hatchery, wild and^131colonized coho salmon smolts before and after transfer to salt waterfor three consecutive years.Citrate synthase activity in the gills of coho salmon during a saltwater^132challenge conducted in May, 1990.The change in interlamellar chloride cell density for hatchery and^133wild coho smolts following transfer to salt water for 14 days in 1991.The relationship between Na+K+ATPase activity and chloride cell^134concentration in the gills of hatchery and wild coho salmon smoltswhen exposed to salt water for 14 days.Cumulative mortality for hatchery and wild coho salmon smolts^135challenged to 35 %o salt water for seven days in 1992.LIST OF ABBREVIATIONSACS^aqueous counting scintillantACTH^adrenocorticotropic hormoneB.^Scatchard plot X-axis interceptwhich is equivalent to the unbound corticosteroid receptor concentrationbGH^bovine growth hormone (somatotropin)bPL^bovine placental lactogenbSA^bovine serum albuminCF^condition factor (Wan3)*100CR^corticosteroid receptorCRF^corticotropin releasing factorDCC^dextran coated charcoalEDTA^ethylenediamine-tetraacetic acidFW^fresh waterGH^growth hormone (somatotropin)IGF-I^insulin like growth factor Ikp^dissociation constantNFR^non-filterable residueSEI^sucrose:EDTA:imidazoleSW^salt waterT4^thyroxinTA^triamcinolone acetonideTEMS^tris:EDTA:monothioglycerol:sodium molybdatexi3diACKNOWLEDGEMENTSI would like to thank my supervisor, Dave Randall, for his assistance throughout thework conducted on this thesis. I have enjoyed the latitude in direction that he has allowed.I would also like to thank George Iwama for his assistance and support throughout thestudies that were conducted to compare wild and hatchery-reared salmonids. I thank CraigClarke, Robin Liley and Don McPhail for their helpful comments. Financial support formuch of the work conducted toward the completion of this thesis was provided by anNSERC-strategic grant to Dave Randall and George Iwama. A British Columbia ScienceCouncil GREAT Award and Department of Zoology Teaching Assistantships also providedfinancial support.Many individuals have given valuable professional and technical assistance in thecompletion of various aspects of the research. I am indebted to the management and staffof the Quinsam River salmonid enhancement facility for provision of fish, rearing facilitiesand technical assistance in completing the examination of hatchery and wild juvenile cohosalmon presented in Chapters 5, 6 and 7. In particular, I would like to express my sincereappreciation to Kathy Campbell for her assistance in capturing and sampling the fish andher helpful insights on culture of salmonids. As always, Bill McLean provided valuableinsight, suggestions and information that helped in my understanding of physiologicaleffects of intensive culture on juvenile salmonids. Alec Maule of the US Fish and WildlifeService contributed greatly at the beginning of the project by his demonstration of thecorticosteroid receptor assay, employed extensively throughout this work. Chris Moyesprovided valuable biochemical knowledge, particularly in the measurement of citratesynthase activity. I would also like to thank Ward Griffioen and the staff at Lois LakeFishculture for provision of fish and assistance during the studies conducted for Chapter1. I would also like to thank Rusty Sweeting of the Biological Department at Simon FraserUniversity for assistance and the use of the growth hormone radioimunnoassay. Theexperiments presented in Chapter 4 were conducted with the collaboration of Bob Devlinand members of his lab at the West Vancouver Fisheries and Oceans Laboratory.It has been very enjoyable working with the other students in the Randall andIwama labs. It has been a great experience to collaborate in research with Dr. ColinBrauner, Hong Lin, Kira Salonius and Ellen Teng. It has truly been a pleasure to get toknow Nicholas Bernier, in the lab, out in the field and as a friend.Finally, I would like to acknowledge the contribution of my family. My wife Juliehas continually given me support and assistance in many aspects of this project. Herorganization and careful attention to detail aided me immeasurably. My two sons, Michaeland Phillip, gave me a joyous outlet and distraction, and always helped keep my perspectiveon the important matters in life.CHAPTER 1GENERAL INTRODUCTION2Most fish maintain the ionic composition and osmolarity of their body fluids at levelssignificantly different from the external environment. The major ions in the plasma offreshwater and saltwater teleosts are sodium and chloride. The mechanisms ofionregulation, however, are distinctly different between freshwater and saltwater teleosts.Freshwater teleosts must continuously eliminate the excess water that diffuses across theepithelium and take up ions to replace those lost to the dilute medium. Saltwater teleostsmust prevent dehydration caused by loss of water to the hypertonic medium and activelyexcrete the excess ions that are absorbed.In fresh water, teleosts oppose the loss of electrolytes and the infusion of wateracross the epidermis by an integument that has a very low permeability to water and ions.The gills are a thin epithelium to enhance the passage of respiratory gases. As aconsequence of the high permeability to gases, fluxes of water and electrolytes across thegills also occur. To compensate for excess water taken up by the fish, the renal systemmaintains a high glomerular filtration rate and the tubules and the bladder reabsorb filteredions to produce a large volume of dilute urine. Freshwater teleosts compensate for the lossof ions by actively transporting ions from the dilute external water to the blood across thegills against a substantial chemical gradient and enable the fish to maintain plasma ionlevels far more concentrated than the external water. Sodium is taken up from the waterthrough a highly selective Na+ channel. The electrochemical gradient that drives themovement of Na+ into the cell is probably produced by an electrogenic apical proton pump(Lin and Randall 1991). Chloride is taken up across the gill by a carrier protein locatedon the apical membrane in exchange for bicarbonate (C17HCO3- exchange). Although, thetransport pathways for sodium and chloride uptake are separate, there are interactionsbetween the two different mechanisms (Perry and Randall 1981).Marine teleosts drink and absorb sea water by the intestine to replace water lostosmotically to the saline environment. Water conservation is enhanced by reduced urineproduction. Excess sodium and chloride absorbed through the gut from drinking sea waterand from passive influx through the epidermis are excreted by the gills. Chloride is activelyexcreted against an electrical and chemical gradient from blood to sea water. The primary3driving force is a large Na gradient between the blood and the intracellular fluid whichis generated by Na+IVATPase located on the serosal membrane of the chloride cells. TheNa+ gradient drives the entry of cr into the cell via a Na/K/2C1 cotransport protein alsolocated on the serosal membrane. The negatively charged interior of the chloride celldrives the extrusion of across the apical membrane. The active transport of chlorideacross the gill creates an electrochemical gradient between the blood and external waterthat favours the paracellular diffusion of Na+ through the leaky junctions between thechloride cells (Epstein et al. 1980).There are some species of teleosts, salmon for instance, that exploit both freshwaterand the marine environment during their life cycle. Salmon spawn in fresh water, and thejuveniles reside in fresh water for variable periods of time. When the fish obtain a criticalsize, they respond to seasonal environmental cues that stimulate physiological andmorphological changes that transform the juvenile salmonid from a freshwater parr to amarine smolt (Hoar 1988). The changes associated with smolting occur in fresh water andare pre-adaptive for the migration of smolts to the marine environment. During the springthere are increases in Na+K+ATPase activity (Young et al. 1989a), chloride cell density(Langdon and Thorpe 1985) and metabolic enzymes (McCormick et al. 1989a). Thesechanges are controlled by the endocrine system. The most notable hormonal changesduring the spring are the increases in concentration of cortisol, growth hormone (Gil) andthyroid hormones (Hoar 1988).Of the hormonal changes associated with smolting, cortisol is the putative saltwateradapting hormone. The surge in plasma cortisol concentration during the spring is closelycorrelated with the increased saltwater tolerance (Specker and Schreck 1982; Hoar 1988;Young et al. 1989a). Treatment of juvenile salmonids with cortisol has provided evidencefor cortisol having a direct role in enhancing saltwater tolerance. Cortisol has been shownto stimulate Na+K+ATPase activity in vitro (McCormick and Bern 1989), although theresponse of gill tissue to cortisol varies seasonally (McCormick et al. 1991a). The evidencefor cortisol enhancing hypo-osmoregulatory ability in vivo, however, is contradictory.Administration of cortisol has been shown to increase Na+K+ATPase activity and chloride4cell concentration (Richman and Zaugg 1987; Madsen 1990a,b,c), have no effect (Eiband Hossner 1985) or decrease enzyme activity (Redding et al. 1984a) in the gills ofjuvenile salmonids. The role of the putative saltwater adapting hormone cortisol, therefore,is not clear. The changes in circulating cortisol levels (Young et al. 1989a) and metabolicclearance rate of cortisol (Patino et al. 1985) in smolting salmonids support the role ofcortisol as an important factor in control of saltwater adaptation. The reason for a lack ofstimulatory or even an inhibitory effect of cortisol on Na+K+ATPase activity could beassociated with changes in gill sensitivity to cortisol as has been shown by McCormick etal. (1991a) in vitro. The changes in gill responsiveness to cortisol observed by McCormicket al. (1991a) could be associated with changes in receptors for cortisol. Sensitivity of targettissue is dependent on the number of steroid hormone receptors (Danielsen and Stallcup1984). High affinity corticosteroid receptors (CR) have been found in the gills of juvenilesalmonids (Chakraborti et al. 1987; Maule and Schreck 1990). The number and affinityof CR have been found to change following increases in plasma cortisol concentration(Maule and Schreck 1991), and the number of CR decline following cortisol treatment(Weisbart et al. 1987).As cortisol has an effect on CR, any factors within a rearing environment that leadto a rise in plasma cortisol concentration may have a significant impact on cortisolsensitivity in the gills and consequently the parr-smolt transformation. These factors aretermed stresses. Stress is a stimulus experienced by an animal which requires aphysiological response for the animal to maintain homeostasis (Pickering 1981). Stressfulsituations for a fish include sudden changes in environmental parameters (ie. temperatureand salinity, Eddy 1981), water quality degradation (ie. increases in ammonia, Smart 1981,or suspended solids, Redding et al. 1987, and hypoda, Randall 1982), and manyprocedures used in intensive fish culture production (ie. high rearing density, handling,grading and tagging, Schreck 1981). It is clear from the descriptions of stressors abovethat many facets of life for a fish elicit a stress response. In the context of this thesis, stresswill be defined as a stimulus experienced by the animal that results in a significant elevationin cortisol above the fluctuations that occur daily and seasonally in salmonids. The5definition has been restricted to focus specifically on the affect of elevation in cortisol onthe physiology of salmon. Cortisol has long been the stress hormone used by fisheriesbiologists to examine implications of rearing practices on the physiology of culturedsalmonids.When fish are stressed, corticotropin-releasing factor (CRF) is released from thehypothalamus to stimulate release of adrenocorticotropic hormone (ACTH) from thepituitary into adjacent blood vessels. The circulatory system carries ACTH to the headkidney where it stimulates the release of cortisol from the interrenal cells into the postcardinal vein (Donaldson 1981). Cortisol is carried via the circulatory system to the targetcells where it likely enters by diffusion through the lipid membrane. Intracellularly, specificcorticosteroid receptor molecules bind cortisol (Maule and Schreck 1990). In mammals,the receptor-hormone complex undergoes a transformation process that enables binding toDNA. The modulation of gene expression by cortisol is accomplished by attachment of CRto specific genes in the nucleus (Clark 1985). It is likely that similar processes also occurin fish.Stress is an inescapable part of the life of fish, particularly fish raised in culturefacilities. Although the rise in cortisol following stress is to maintain homeostasis for theanimal, there are often other consequences of stress that lead to impaired performance ofthe fish (Wedemeyer and McLeay 1981). These responses take the form of behavioralchanges, decreased growth, increased susceptibility to disease and impaired smolting. Inlight of the importance of cortisol for the development of hypo-osmoregulation as fishmigrate from fresh water to salt water, stressors that cause corticosteroid release maymodify osmoregulatory development. As smoltification in salmonids is associated with arise in plasma cortisol and stress causes a rise in cortisol, it may be expected that stress willenhance seawater survival. This is not the case, however, and stress has been shown toreduce early marine survival (Schreck et al. 1985; Patino et al. 1986). The reason for thisapparent inconsistency may be related to the regulation of CR in the gill by cortisol. CRreceptors show a downregulation following exposure to corticosteroids in mammals (Claireet al. 1981; Sapolsky et al. 1984; Svec 1988) and in mammalian cell cultures (Cidlowskiand Cidlowski 1981; Svec and Rudis 1981; Danielsen and Stallcup 1984). The6desensitization of the target cell may also occur in salmonids that experience prolongedelevation in cortisol due to stress. The result may be a decreased sensitivity to cortisol inthe cells in the gill that differentiate into chloride cells and show an increase inNa+K+ATPase activity.It is likely that the sensitivity of gills to cortisol can be affected by other hormonesthat function in the parr-smolt transformation, specifically GH. GH also improves thehypo-osmoregulatory ability in juvenile salmonids. There is evidence that GH can increasesaltwater tolerance in juvenile salmonids (Komourdjian et al. 1976) by the stimulation ofNa+K+ATPase activity and chloride cell concentration (Richman and Zaugg 1987; Madsen1990b,c). GH, however, has a much greater stimulatory effect on Na+K+ATPase activityand proliferation of chloride cells in the gills when administered in conjunction withcortisol. GH has been shown to have an effect on the sensitivity of the interrenal cells inthe head kidney to ACTH (Young 1988). Elevated cortisol levels in the plasma have alsobeen found to have an effect on circulating GH concentrations. An acute rise in cortisol,results in a decline in plasma Gil and a chronic rise in cortisol results in an increase inplasma Gil concentration (Pickering et al. 1991). The interaction between cortisol andGH in the plasma is also likely to affect CR in the gills. Gil may exert an effect on CRindirectly through changes in plasma cortisol concentration or Gil may act directly on theCR populations in the gills. As the concentration of GH has also been found to increaseduring the spring when juvenile salmonids smolt (Sweeting et al. 1985; Prunet et al. 1989;Young et al. 1989a), it is possible that Gil may also participate in the regulation of CRin the gills of smolts.The objective of the present study was to examine the relationship betweencorticosteroid receptors and the seasonal changes in saltwater tolerance that occur injuvenile coho salmon (Oncorhynchus kisutch). Factors that are known to affect thedevelopment of saltwater tolerance in smolts were assessed to determine their impact onCR in the gills. The factors that were chosen to examine for an effect on CR in the gillswere physical disturbance and cortisol treatment, Gil treatment, and differences in rearingenvironment. Rearing environment involved a comparison of juvenile coho reared in anintensive fishculture facility with wild fish inhabiting a natural river system. The7comparison between hatchery and wild coho also examined physiological changes involvedin smolting. This study determined the relationship between CR, seasonal changes inplasma cortisol and GH concentration, and the increased activity of Na+1C+ATPase insmolts. The results of this thesis provide an understanding of some of the regulatorymechanisms controlling gill CR and how changes in CR may mediate smolting in juvenilecoho salmon.CHAPTER 2Seasonal Changes in Gill Corticosteroid Receptorsand Smolting in Juvenile Coho Salmon(Oncorhynchus kisutch)89INTRODUCTIONAnadromous salmonids are capable of tolerating increases in salinity over a limitedportion of the year. A variety of physiological and morphological changes are associatedwith an increase in saltwater tolerance and smolting. These changes are pre-adaptive, asthe transformation from the freshwater parr to the migratory smolt occurs in freshwater.Smolting is characterized by a loss of parr marks and a more silver appearance (Gorbmanet al. 1982). An important component of smolting is an increase in the number of chloridecells and the activity of the enzyme Na+K+ATPase, which leads to an increased ability toosmoregulate in salt water (Hoar 1988). Smolting occurs during the spring. An increasein hypo-osmoregulatory ability, however, has also been shown to occur in the fall in somegroups of coho salmon (Conte et al. 1966; Otto 1971; Harache et al. 1980). Theincreased hypo-osmoregulatory ability during these two periods is associated with increasesin gill Na+K+ATPase activity (Lasserre et al. 1978; Bjornsson et al. 1989). The elevatedNa+K+ATPase activity in smolts is correlated with increased ocean survival (Zaugg andManken 1991).The parr-smolt transformation is stimulated by seasonal endocrine changes. Oneof the principle endocrine changes associated with smolting is an increase in plasma cortisolconcentration during the spring (Specker and Schreck 1982; Maule and Schreck 1987;Young et al. 1989a; Hoar 1988). The increased cortisol levels in the plasma likelystimulate the physiological and biochemical changes associated with smolting viaintracellular corticosteroid receptors (CR). High affinity CR have been found in the gillof coho salmon (Maule and Schreck 1990). The responsiveness of coho salmon gill tissueto cortisol has been shown to vary with season (McCormick et al. 1991a). It is likely thatthis response is due to variation in gill CR concentration with season and physiological stateof the fish. The changes in gill CR may be associated with the seasonal change in cortisollevels in the plasma as cortisol administration has been found to affect CR numbers andaffinity for cortisol (Maule and Schreck 1991).This study examined the seasonal change in gill corticosteroid receptor concentration10and affinity The changes in CR and plasma cortisol concentration were examined todetermine a relationship between these variables and the increase in Na+K+ATPase activityand development of hypo-osmoregulatory ability. The fish chosen for the study wereKitimat River coho salmon that were reared in lake net pens. This population of fishshowed extreme variation in growth rate and timing of increased saltwater tolerance. Thelarger fish lost their parr marks, developed the silver colouration characteristic of smolts,and could be transferred to salt water in the fall. It appeared, therefore, that the larger fishsmolt within the first year. The smaller fish showed a more typical pattern of development,showing no appearance of smolting until the spring of the second year. This populationprovided an opportunity to examine fish that were the same age, but differedphysiologically. The fish were examined for changes in gill CR concentration and affinity,plasma cortisol levels, and gill Na+K+ATPase activity over the year while resident in freshwater.11MATERIALS AND METHODSExperimental Animals. Coho salmon (Oncorhynchus kisutch) from the Kitimat Riverstock were reared in net pens in Lois Lake, British Columbia, by Westcoast FishcultureLtd.. Juvenile coho were transferred into the lake in March as fry. The net pendimensions were 6m (length) by 6m (width) by 20m (depth).In the lake there is a large seasonal temperature change. The minimumtemperatures were recorded in February; 6.0 °C in 1992 and 4.3 °C in 1993 at the surfaceof the lake. There was little variation in temperature with depth during the winter months.During the summer months, the temperature exceeded 20 °C at the surface. Watertemperature declined with depth and was never greater than 12 °C at 20m. During thewarm summer months, the fish remained deep in the water and would come near thesurface only for brief intervals when they were being fed.Sampling procedure. The fish were sampled on June 22, August 29, November 281992, March 6 and May 1 1993. In June and August, the fish were caught using a dip netafter enticing the fish to the surface with food. This method was effective at capturing arepresentative sample in June. In August, however, the larger fish could not be capturedusing this method and only the small fish could be sampled. In November, March andMay, a seine net was used to catch fish within the net pen. The time required to catch thefish and transfer them into a bucket containing anaesthetic was under one minute using thedip net technique. Pickering and Pottinger (1983) indicate that there was no effect onplasma cortisol level if fish were sampled within three minutes of capture. Capture of fish,however, was considerably slower using the seine net. The fish were first crowded in theseine and then removed from the seine into the anaesthetic bucket using a dip net. Thetime that elapsed from seining to dip netting the fish into the bucket containing anaestheticwas approximately 15 min. Pickering and Pottinger (1983) have shown that cortisol levelscan rise within 15 min of exposure to a stress, however the maximal cortisol levels in theplasma do not occur until 1.5 to 2 hours following initial handling (Barton et al. 1986).After capture, the fish were killed by a lethal dose (200 mg.L-1) of buffered MS-222.12This dose has been shown to inhibit the stress-related rise in plasma cortisol concentrationin salmon (Barton et al. 1986). It was expected, therefore, that this dose of anaestheticwould also inhibit any further rise in plasma cortisol concentration. After the fish wereanaesthetized, the caudal peduncle was severed and at least 60 p.1 of blood collected inheparinized capillary tubes or the blood was collected in a heparinized syringe by caudalpuncture for larger fish. The blood samples and fish were then placed on ice andtransported back to the laboratory on the fish farm. In the lab, the length and weight ofthe carcasses were determined. The plasma was separated by centrifugation and thenstored at -80 °C until determination of cortisol concentration.The fish sampled were classified as small, medium or large based upon size andappearance. In June 1992 there was little difference in size, but some of the fish sampledwere loosing their parr marks. The fish were classified as small if parr marks were presentand large if parr marks were absent. In August 1992, only small fish with parr marks werecaught and these fish were classified as small. In November 1992 and March 1993, threesize groups of fish were apparent. The small fish had visible parr marks. The medium fishwere silver, had no parr marks and were smaller than 45 g (November) or 50 g (March).The large fish were silver and larger than 60 g (November) or 70 g (March) and had noparr marks. In May 1993, all the fish sampled were silver and had lost their parr marks,the classification of groups was by size. The mean and distribution of length for the fishcollected at each sampling interval are shown in Figure 2.1.Tissue Preparation (CR). For determination of corticosteroid receptors, gills wereexcised from the fish, and blotted on tissues to remove excess blood and then placed in asolution of 10 mM Tris-HC1, 1 mM EDTA, 12 mM monothioglycerol, 20 mM sodiummolybdate, and 10% (v/v) glycerol all at pH 7.4 (TEMS). Approximately 0.5 g of gill tissuewere added to 2.5 mL of TEMS and stored in liquid nitrogen until samples were assayedfor corticosteroid receptors. Tissues from up to 5 individuals were pooled for one sample,dependent on the size of the fish. A method similar to that of Maule and Schreck (1990)was used for analysis of corticosteroid receptors. All procedures were carried out at 4 °Cby keeping samples on ice. Tissue was chopped finely, and then homogenized using anIKA-Ultra Turrax TP 18/10S1 homogenizer for three ten-second bursts. Homogenates were13centrifuged in a Sorvall RC5C centrifuge with a SS-34 rotor at 2000g for 10 minutes toremove gross debris. The supernatant was removed and centrifuged at 100 000g for onehour in a Beckman L8-80 Ultracentrifuge with a SW-51 rotor. After this centrifugation,the supernatant was removed mixed with an equal volume of TEMS containing 5% (w/v)Norit A charcoal and 0.5% (w/v) dextran and incubated for 10 minutes to removeendogenous steroids. To separate the charcoal from the liquid, the samples werecentrifuged at 3500g for 20 min. The final supernatant was used to quantify cortisolbinding. Protein content of this fraction was assayed with Bradford reagent (1976), and ifnecessary buffer was added to adjust protein content to 4-6 mg per mL.Steroid Binding Assay. Cortisol binding receptor studies were conducted with[3H]triamcinolone acetonide (TA; 1,4-pregnadien-9a-fluoro-1113,16a,-17a,21-tetrol-3,20-dione-16,17 acetonide) with a specific activity of 57.9 Ci.mmo1-1 (Dupont-NEN). The CRassay had previously been validated by Maule and Schreck (1990) for coho salmon gillhomogenate. Their results showed high specificity of the receptors for cortisol and TA.TA was used throughout the study as Maule and Schreck (1990) showed that in bindingand competition studies on duplicate gill homogenates, TA and cortisol appeared to bindto identical receptor populations, except TA bound to the receptor with a higher affinity.To determine the number of high affinity cortisol receptors, 100 ill of the final supernatantwas incubated in duplicate aliquots with 100 ill of buffer containing [3H]TA with andwithout a 200 fold excess of cold TA. Final concentration of [311]TA ranged from 0.1 to9 nM. The tubes were vortexed and incubated for two hours. After the incubation period,0.5 mL of TEMS containing 2.5% (w/v) Norit A charcoal and 0.25% (w/v) dextran wasadded and vortexed. After 10 min, the charcoal containing unbound ligand was separatedfrom bound ligand by centrifuging at 3000g for 20 min in a Beckman TJ6R refrigeratedcentrifuge. A 0.5 mL volume of the liquid was added to 5 mL of ACS (Amersham)aqueous counting scintillant. Samples were counted on a LKB 1214 rackbeta liquidscintillation counter. Specific binding was determined by subtracting non-specific boundfrom the total bound.Three parameters were calculated to describe the receptors; the concentration ofunbound CR (B.), the dissociation constant (1cD), and the Hill coefficient. Binding of CR14to endogenous cortisol in vivo leads to a hormone-receptor complex that becomes activated,binds to the DNA and will not be detected by the radioreceptor assay used in this study.The CR concentration measured is comprised of the unbound receptor population. Thedissociation constant is an equilibrium value for the ligand and receptor interaction and isa measure of the degree of dissociation of the ligand from the receptor in a solution. Theaffinity of the receptor for the ligand is greatest when kE, is small. The Hill coefficient isa measure of cooperativity between binding sites (if more than one exists). The equilibriumdissociation constant (kr,) and the concentration of corticosteroid receptor sites (B..) werecalculated according to Scatchard (1949). Bm was divided by the homogenate proteinconcentration, and CR concentration was expressed as fmol • mg protein-1. The HillCoefficient was calculated according to Bennett (1978).Tissue Preparation (Na+ ATPase activity). For determination of Na+K+ATPaseactivity, gills from one side of a fish were removed and blotted dry on paper tissues toremove excess blood. Approximately 50 mg of gill tissue from each fish was removed andplaced into 0.5 ml of ice cold SEI (0.3M sucrose, 0.02M EDTA and 0.1M Imidazole) fordetermination of Na+K+ATPase activity. The samples were stored at -80 °C until analysisof Na+K+ATPase activity. Na+K+ATPase activity in the gill samples was determinedaccording to the protocol of Zaugg (1982). Briefly, gill samples were thawed on ice andminced with a pair of scissors. The samples were subjected to two 10 s burst with a Kontesmicro ultrasonic cell disrupter, and then centrifuged for five minutes at 5000 g. Thesupernatant was discarded, the pellet resuspended in 0.5 mL of SEID (SEI and 0.1% w/vsodium deoxycholate), and subjected to a single 10 s burst with the sonicator. Thehomogenate was centrifuged for five min at 5000 g. Ten p.1 aliquot of the supernatant wereadded to tubes containing 0.65 mL of reaction buffer with or without 6 M ouabain. Thereaction buffer contained 23 mM MgC12, 155 mM NaC1, 75 mM KCI and 115 mMimidazole. Samples were incubated at 37 °C for 10 min. After 10 min, the samples werecooled in an ice bath for one min, then 2.0 mL of 0.95% HC104, 3.0 mL of 2-octanol and0.25 mL of 47 mM ammonium molybdate (in 20% concentrated Ha) were added to eachtest tube. The test tubes were covered and vigorously shaken for 30 s, and then 0.5 mL of0.74 M citric acid (adjusted to pH 2.9) was added and the mixture shaken for another 30s.15The test tubes were allowed to stand until separation of the octanol and aqueous phaseswas complete. To quantify the production of inorganic phosphate, the absorbence of theoctanol layer was determined at 312 nm The difference in Pi production between tubeswith or without the presence of ouabain in the reaction buffer was calculated to be theouabain sensitive Na+K+ATPase activity. Protein content in the gill homogenates wasmeasured using the Biuret assay (Alexander and Ingram 1980). The amount of phosphatehydrolysed in the samples was then expressed as limo' Pi • mg protein-1 • hr-1.Saltwater Challenge. In May, 40 fish varying in size were subjected to a 24 hrsaltwater challenge (SWC) (Blackburn and Clarke 1987). The fish were captured from thenet pen and transported to a large (1 m3) transport tank filled with saltwater. Salinity ofthe SWC was 27%o. Oxygen was provided to the tank via a perforated airline. Followingthe saltwater challenge, the fish were anaesthetized and measured for weight and length.Blood was collected by caudal puncture and treated as described above. Plasma sampleswere stored at -80 °C until analysis for sodium concentration.Cortisol Radioimmunoassay. Plasma cortisol levels were determined byradioimmunoassay kit using 125! (Incstar Corp.) as validated by Heath (1993).Plasma Sodium Analysis. Plasma samples were analyzed for sodium using a Perkin-Elmer model 2380 atomic absorption spectrophotometer.Statistical Analysis. The concentration of corticosteroid receptors, kp, and Hillcoefficient, plasma cortisol and sodium levels were compared for significant differencesusing an analysis of variance (ANOVA), followed by a Tukey's test to find significantdifferences between means. Statistical significance was taken at a level of a = 0.05. Allvalues are expressed as means ± 1 s.e..16RESULTSGrowth. The sizes of the fish sampled during the study are shown in Figure 2.1. InJune, all the fish were similar in size, but some were already losing their parr marks andstarting to have the characteristic silver appearance of smolts. The fish that were silver inappearance showed accelerated growth. Consequently by July, there was a large range insizes of fish in the net pen. The largest fish in the pen were approximately 60g in weight,while the smaller fish were less than 10g. At this sampling time, it was impossible tocapture the larger fish using a dip net, and only the small fish could be caught and sampled.There still existed a marked variation in size among the population of coho from the lakenet pen in November. The largest fish maintained the silver appearance, while the smallermembers of the population still exhibited noticeable parr marks. The appearance and sizeof the fish sampled in March and November was similar. By the May sample date,however, all the fish in the population, both large and small, were silver and appeared tohave smolted.Plasma cortisol. Resting levels of plasma cortisol from the Lois Lake coho over theyear are shown in Figure 2.2. Over the year the concentration of cortisol in the plasmaincreased. In the November sample, the plasma cortisol level was significantly greater inthe large coho than the small or medium. The level of cortisol in the large fish was alsosignificantly greater than the two previous sampling times. The March values of plasmacortisol did not differ significantly from the previous sample measurements. In May therewas a significant increase in plasma cortisol concentration in the coho and there was aslight positive relationship between size and plasma cortisol concentration. Thisrelationship was not significant, however, so the data for the three size ranges of fish werepooled for presentation in Figure 2.2. The May surge in cortisol did not differ significantlyfrom the November surge.Corticosteroid Receptors. There were seasonal changes in the concentration andaffinity of gill CR over the year (Figure 2.3). The Hill Coefficient did not vary and wasnever significantly different from one. At some periods of the year, there were also17significant differences in B. and IrD between fish of difference sizes.B. declined during the late summer and fall, increased significantly during the earlyspring, and then declined again over the spring (Figure 2.3a). B. did not differ betweenthe different sized groups of the fish at any time during the year, other than in May. Therewas no correlation between size and B. for the fish sampled in November (Figure 2.4a).In the May sample the gill CR concentration was negatively correlated with the size of thefish (Figure 2.5a). The largest fish showed significantly lower CR concentration than thesmall fish (Figure 2.3a).Over the year, the affinity of the gill CR declined (1cD increased; Figure 2.3b). IrDdid not differ significantly between the different sizes of fish used in the study, except inthe November sample. In November, samples from the medium sized fish differedsignificantly from the small fish, and the large fish differed significantly from both themedium and small fish samples. The correlation between size and IrD is shown in Figure2.4b. Figure 2.5b shows that no relationship existed between size and Icp for the fishsampled in May.Na+ IC ATPase Activity. The activity of Na+K+ATPase in the gill of coho salmon overthe year is shown in Figure 2.6. For the smaller fish, enzyme activity was constant duringthe year until the spring when levels were the highest. The larger fish showed a morevariable pattern. For most of the year, the Na+K+ATPase activity was highest in the largestfish in the population. In November, only the largest fish in the population showedNa+K+ATPase activity that was significantly greater than the small fish. In May, however,there was no difference in the Na+K+ATPase activity between the different sizes of fish.Plasma Na+ Concentration. Following exposure to saltwater in May, all fishexamined survived and maintained plasma sodium concentrations below 170 mEq.C. Themean plasma sodium concentration of the fish challenged to salt water was 156.2 -± 0.82mEq.C. There was no relationship between the 24-hr plasma sodium concentration andthe size of the fish (Figure 2.7).LL1_JLi_000CL0a_15 20 25 3018In rnAUGn=24• • JUN -n=25• SMALLA MEDIUM• LARGE• NOVn=24• A^• MAR -n=32^-201 001 0402004020030201 0030FISH LENGTH111Figure 2.1. The proportion of fish for each size sampled during the study (bar graph).Proportions are expressed as percentages of the total number of fish sampled at each time.The mean si ze of the fish for each sampling period are shown by the symbols at the top ofeach graph.19500Cl)400 -I0 E 30<20(f)Cl_ 10Figure 2.2. The plasma cortisol concentration in three size groups of juvenile coho salmonover the year. When there is no significant difference between the three sizes of fish, thedata are pooled and presented for all the fish. Vertical bars represent J.: 1 se. n for eachsample is listed adjacent to the data point. * indicates the value is significantly differentfrom the value for the small fish at the same sampling period. Size ranges and means forthe groups at each sample interval are shown in Figure 2.1.1 40(f)CC x0I--0 mECI_L.L10LiCC0.6-^1.0czuiI—(r)0^0.8CD— oi---- __YCC0^0.60Figure 2.3. The seasonal change in gill corticosteroid receptor concentration (B.; a) anddissociation constant (kJ); b) for coho salmon. When values do not differ between the threesize groups, the data are represented as just the small fish. n = four unless listed adjacentto the mean value. * and ** indicate the value is significantly different from the value forthe small and medium fish, respectively, at the same sampling period.2011 0805060><00_LiJo^40uJC-21^300LJ^0.9o^0.81.--=cz^0.700.60.52110^100FISH WEIGHTFigure 2.4. The relationship between B. (a) and IcE, (b), and the size of the fish sampledin November. The linear regression line through the data is plotted. Units are fmol • mgprotein-1 for B. and nM for kr,.Cl)fX x 800 om EI—fa_LLI^60(..)LL.14005-f:L.LiI--^0.95Cl)0- o._0r---0.90000.85101 00•R2=0.0467^•^b-• • •••^-• •• •I^1^I10022FISH WEIGHTgFigure 2.5. The relationship between B. (a) and IrD (b), and the size of the fish sampledin May. The linear regression line through the data is plotted. Units are fmol • mgprotein1 for B. and nM for IcD.Figure 2.6. Seasonal changes in gill Na+IVATPase activity of juvenile coho salmon.Symbols as described in Figure 2.2. The number of samples for each value is listedadjacent to the data point.2315^20LENGTH• • • •• •• •17016015024••••• •• R2=0.0081•••• so•••• •-•--••••••Figure 2.7. Plasma sodium concentration after 24-hr saltwater challenge for coho in Mayas a function of fish length.25DISCUSSIONSeasonal changes in Na+K+ATPase activity were observed in coho examined in thisstudy. The temporal increases in Na+K+ATPase activity were consistent with the findingthat coho salmon show seasonal changes in increased hypo-osmoregulatory ability duringthe fall and spring (Conte et al. 1966; Otto 1971; Harache et al. 1980). The increasedsaltwater tolerance is also synchronous with a preference for increased salinity (Otto andMcInerney 1970). At times of the year when the fish show increased saltwater preference,migrations occur into the estuary (Tschaplinski 1987). The increase in saltwater toleranceduring November and May are consistent with increases in Na+KfATPase activity in thegills (Lasserre et al. 1978; Bjornsson et al. 1989; Young et al. 1989a). Although thereare two periods during the year when saltwater tolerance increases, the fall elevation insaltwater tolerance is not as great as the spring elevation. In the fall, gill Na+KfATPaseactivities are not as great (Lasserre et al. 1978; Bjornsson et al. 1989) and theperturbation in plasma sodium is greater when fish are exposed to salt water (Otto 1971;Harache et al. 1980; Tschaplinski 1987). In the present study, the large fish exhibitedNa+K+ATPase activities similar to that seen in the spring in the Lois Lake coho. Thechange in kp and surge in cortisol did not differ significantly in the spring and the fall.Based on the change in Na+K+ATPase activities, CR concentration and affinity, saltwatertolerance and appearance, the physiological changes that occur in the fall are similar tothose that occur in the spring.The larger fish in the population studied showed increases in Na+K+ATPase activityin the fall and then again in the spring, whereas the smaller individuals showed increasesin Na+K+ATPase activity only in the spring. The changes in enzyme activity werecorrelated with increases in plasma cortisol concentration and a decrease in CR affinityThe altered cortisol dynamics and the higher enzyme activities indicate that the large cohoundergo physiological changes that are characteristic of smolting at least two times duringthe year. The physiological changes may also occur in natural populations of coho, as fallmigrations of coho parr into the estuary have been reported (Tschaplinski 1987,26Varnavsky et al. 1992). In both of these studies the fish were small (< 10 g) and stillexhibited parr marks. The fish remain in the estuary, but do not appear to venturecompletely into the marine environment. It does appear, therefore, that changes in hypo-osmoregulatory ability occur not just in the spring, but also in the fall in coho salmon.Environmental changes such as photoperiod and temperature have been implicatedin the control of smolting (Saunders et al. 1989, Clarke and Shelbourn 1986, Hoar 1988).Altering photoperiod has been used to alter the development of coho and produce smoltsin the first year after spawning (Clarke and Shelbourn 1986). No deliberate photoperiodmanipulation was performed on this stock of fish, however, it is possible that there wasinadvertent photoperiod manipulation. The higher temperature of the groundwateradvances hatching time. Due to the earlier date of hatching, the fry are ponded during ashort-day photoperiod. A brief exposure to short-day photoperiod at first feeding leads tothe development of bimodality with a mixture of parr and smolts in the first year (Clarkeand Shelbourn 1986).Size may also play a role in hypo-osmoregulatory development. Only the largest fishshowed significant increases in Na+K+ATPase activity during the fall. Lasserre et al. (1978)examining coho salmon in France also showed a strong size dependence on increases inNa+K±ATPase activity, with larger fish in the population exhibiting significantly greaterenzyme activities. The findings of Lasserre et al. (1978) and this study suggest a critical sizefor smolting in coho. The critical size for coho salmon to smolt has been reported to be12 g (Clarke 1982; Hoar 1988). The fish that appeared to smolt in November were muchlarger (>60g) than the critical size. In the spring, when the rest of the population smolted,the mean size of these fish was also much larger than the reported critical size.The rapid growth rate of the fish reared in Lois Lake is likely a result of the highlake temperatures during the summer. Temperatures close to 21 °C result in very rapidgrowth of coho (Lasserre et al. 1978). It is surprising that the fish exhibited such fastgrowth in Lois Lake. The fish tended to remain deep in the lake where the water wascolder and would move into the warm surface water only when feeding. The large variationin size of the fish could result from genetic differences within the population. Developmentof bimodality in length frequency has been shown in chinook salmon (Heath 1993) and27Atlantic salmon (Thorpe et al. 1980). The reason for the differences in growth aregenetically controlled in these populations (Heath 1993, Thorpe et al. 1982). In Atlanticsalmon the upper mode of the population smolt a year earlier than the smaller fish.Similar genetic factors may control growth in coho and result in the early development ofhypo-osmoregulatory ability of the large fish. The higher gill Na+K+ATPase activity in thefish sampled in June indicates that the physiological changes associated with increasedsaltwater tolerance in the fall begin early in the season. It is not known whether thechanges documented, occur naturally in the Kitimat River population of coho salmon.There is a relationship between the size of the fish and kr,, and B.. The changein kl, in November in the large silver coho was correlated to Na+K+ATPase activity(Figures 2.3 and 2.6) and enhanced saltwater tolerance (W. Griffioen, WestcoastFishculture, personal communication). The changes in kr, could be associated with theincrease in circulating cortisol levels as cortisol can affect the affinity of the receptor(Chapter 3). The affinity changes may also be a function of physiological changes in thesmolts. Seasonal changes in kD have been reported in toads (Lange and Hanke 1988) andglucocorticoid receptor affinity changes in pregnant women (Junker 1983). Changes in ku,with season and physiology of juvenile coho occur with smolting, but the mechanism of thischange is not clear.Seasonal changes in CR concentration were seen. B. was similar for the large andsmall fish at all sample times except for May, where a significant relationship between B.and size of the fish was found (Figure 2.5). Changes in B. have been reported withseason in toads (Lange and Hanke 1988) and with developmental stage during chickembryogenesis (Gendreau et al. 1987). The changes in Brna„ in the present study appearto be correlated to season and development. There are clear difference in B. betweenthe sampling dates. A comparison of the May samples indicates that CR concentration isassociated with size and likely the physiological state of the fish. The differences in B.seen in May could be linked to endogenous cortisol levels. There was a correlationbetween size and plasma cortisol in May, however, the differences in plasma cortisolconcentration do not appear to be large enough to account for the differences in B. seenin these samples. Other factors must also regulate the concentration of CR in the gills of28coho salmon.Plasma cortisol concentrations showed significant variation during the year.Circulating levels of cortisol were significantly elevated in the November and May samples.The surge in cortisol associated with smolting has been shown previously (Specker andSchreck 1982; Young et al. 1989a) and is one of the endocrine changes to increaseNa+1C+ATPase activity and chloride cell density (Madsen 1990a). The fish thatdemonstrated the increase in cortisol concentration also appeared to be smolted as definedby the loss of parr marks and silver colouration. The higher levels of cortisol in the fishsampled in November, March and May when compared to June and August may also bea result of the sampling procedure. The time required to capture and anaesthetize the fishwas less using the dip net than the seine net. As a consequence, the fish seined from thenet pen in the November, March and May samples, may have experienced a crowding stressand the cortisol concentrations may not represent resting levels. As there was a differencein cortisol levels between the small and large coho when sampled in November, thissampling effect may be minor.As steroid hormone sensitivity is dependent on the number of receptors in the targettissue, the seasonal changes in corticosteroid receptor affinity and concentration will affectthe sensitivity of the gills to cortisol (Danielsen and Stallcup 1985). McCormick et al.(1991a) reported the responsiveness of gills to cortisol in enhancing Na+K+ATPase altersseasonally and with the development of smolting. Coho showed the greatest sensitivity inearly spring, but the sensitivity had declined by late spring. Atlantic salmon smolts showedless sensitivity to cortisol than parr that were the same age. The mechanisms for thechanges in sensitivity in coho and Atlantic salmon observed by McCormick et al. (1991a)is likely due to changes in B. and kE, of the gill CR. The increase in kr, and decrease inB. will both lead to a reduced sensitivity. The gill CR changes reported here coincidewith the timing of the gill cortisol sensitivity changes reported by McCormick et al. (1991a).It is probable that the peak in CR concentration in the March sample enhances gillsensitivity to cortisol prior to the spring cortisol surge. An increase in cortisol sensitivitywill provide maximal stimulation of chloride cells and Na+K+ATPase activity. A greatersensitivity to cortisol prior to the fall increase in saltwater tolerance might also be expected,29but was not seen in the fish sampled. The increase in gill sensitivity might be expected tooccur at the end of the summer (August). The fish sampled in August, however, were thesmall parr in the population and did not exhibit signs of smolting in the fall. It is possible,therefore, that the small fish do not experience an increase in sensitivity prior to the fallsmolting window. There is some evidence that the larger coho may show changes in CRdifferent from the small fish. The difference between Na+K+ATPase activity between thefish that exhibited distinct parr marks from the fish that were starting to turn silver in Juneindicates that the fish differ physiologically. The fish used in the downregulation studiespresented in Chapter 3 were the same stock of coho salmon from Lois Lake but theprevious year. For the downregulation studies, the largest fish from the net pen wereselected and transported to the rearing facilities at UBC. The fish examined exhibited faintparr marks and the silver appearance of a smolt. These fish would classified as mediumin size, according to the size classification used in the present study. The control fish usedin this series of studies were sampled in mid-August, 1991. The B. and kr, of this groupof fish were 96.7 ± 1.7 fmol • mg protein' and 0.77 ± 0.03 nM, respectively, bothconsiderably greater than the values for the parr sampled in August of 1992 (B. 54.6 ±3.0 fmol • mg protein' and lc], 0.54 ± 0.03 nM). As the fish sampled in 1991 weretransported back to UBC, it is possible that the CR levels differed from the fish thatremained resident in the lake net pen. If the transport and handling resulted in an increasein plasma cortisol concentration, the effect on the fish would be to reduce the affinityand/or the concentration of the gill CR. The 1991 measurements of CR concentration andaffinity, therefore, probably represent conservative estimates. CR concentration and affinitymay have actually been greater. The consequence of these changes is that the fish showingphysiological changes typical of pre-smolts do show an increased sensitivity to cortisol inthe summer prior to the fall smolting window.Significant changes in CR concentration and affinity that occur seasonally in cohosalmon are associated with development of increased hypo-osmoregulatory ability. Priorto the surge in cortisol concentration associated with smolting, there is an increase in theconcentration of CR in the gills. This increase in CR can explain the increasedresponsiveness of gill tissue to cortisol found by McCormick et al. (1991a) in the early30spring. The decline in CR concentration and affinity when the juvenile coho in this studyshowed the highest levels of Na+K+ATPase activity may be associated with adownregulation of the CR due to the elevated plasma cortisol concentrations at this time.CHAPTER 3Downregulation of Corticosteroid Receptorsin the Gills of Coho Salmon (Oncorhynchus kisutch)due to Stress and Cortisol Treatment3132INTRODUCTIONThe steroid hormone cortisol plays an important role in maintaining electrolytehomeostasis in both freshwater (Chester Jones et al. 1969) and seawater teleosts (Mayeret al. 1967). Although, the mechanisms of ion regulation are quite different in sea waterand fresh water, many fish are capable of migrating between the two environments.Cortisol stimulates many of the physiological changes required for the transition from freshwater to sea water, many of which occur in the gills, the principle site of extrarenal iontransport. Cortisol levels increase as salmon smolt. The surge in cortisol associated withsmolting increases saltwater tolerance by cellular differentiation of chloride cells (Madsen1990a), and stimulation of Na+K±ATPase activity (McCormick et al. 1989a).The action of cortisol in the gills is likely mediated by specific high affinitycorticosteroid receptors (CR). CR have been detected in several species of salmonids; cohosalmon, Oncorhynchus kisutch (Maule and Schreck 1990), rainbow trout, 0. mykiss(Pottinger 1991), and brook trout, Salvelinus fontinalis (Chakraborti et al. 1987). Aftercortisol binds to the receptor, the receptor-hormone complex binds to DNA in the nucleusand initiates changes in gene expression that are translated into a cellular response. Thesensitivity of the tissue is dependent on the number of receptors in the target organ(Danielsen and Stallcup 1984). This sensitivity can be regulated by previous hormonalexposure. The response of the CR population to corticosteroids is tissue specific.Upregulation of CR populations can occur following corticosteroid treatment or removal.In mammals, treatment of lymphocytes with dexamethasone causes an increase in thenumber of glucocorticoid receptors (Eisen et al. 1988). Removal of endogenous steroidsby adrenalectomy also results in an upregulation of the number of rat kidneymineralocorticoid and glucocorticoid receptors (Claire et al. 1981). Downregulation of CRin a number of tissues has been demonstrated in mammals. An increase in glucocorticoidlevels has been shown to cause a decrease in receptor numbers (Cidlowski and Cidlowski1981; Sapolsky et al. 1984). The regulation of CR by cortisol levels could have an effecton smolting in salmonids. McCormick et al. (1991a) found that the responsiveness of gill33tissue to cortisol varied seasonally and with stage of development. The number and affinityof CR also change seasonally and during the parr-smolt transformation (Chapters 2 and 5).It is possible that the variation in CR is dependent on the seasonal changes in plasmacortisol concentration. Cortisol is also a stress hormone (Mazeaud et al. 1977), and stresshas been shown to impair smolting in juvenile salmonids (Strange et al. 1978; Schreck etal. 1985; Patino et al. 1986). It is possible that the elevation in plasma cortisol associatedwith stress alters CR in the gills. The difference in CR between hatchery-reared and wildjuvenile coho salmon support this hypothesis.In rainbow trout, the concentration of CR have been shown to be inversely relatedto the cortisol concentrations (Pottinger 1991). This relationship was determined whilecirculating concentrations of cortisol were still elevated. It is likely, therefore, that thenumber of unbound CR has declined, but the total receptor pool has remained unchanged.Following a single injection of cortisol, there is a rapid depletion in receptor numbers.After the decline in concentration of steroid, receptor concentrations return to apretreatment level within 24 hr (Weisbart et al. 1987). Thus, there is little evidence fora sustained effect of cortisol treatment on CR in salmonids. Long term treatment,however, may lead to a persistent reduction in receptor numbers despite a return to basallevels of the hormone. This study was designed to determine if downregulation of CRpopulations occurs in the gills of coho salmon, following an acute or chronic elevation inplasma cortisol concentration.34MATERIALS AND METHODSExperimental Animals and Protocol.Acute Experiments.^The effects of an acute elevation of plasma cortisol oncorticosteroid receptor concentrations was measured in the gills of yearling post-smolt cohosalmon in July, 1991 (Quinsam River Stock). The fish were silver and did not have parrmarks. Fish were reared in 200L fibreglass tanks, each supplied with dechlorinatedVancouver city tap water. Temperature during the experiment was 7.9 °C. Mean weightwas 32.8 ± 1.3 g, SEM (n=60). Fish were anaesthetized in 25 mg.L-1 MS-222 (tricainemethanesulfonate) buffered with 50 mg.1:1 of sodium bicarbonate. The fish were injectedintraperitoneally with 8J.Lg cortisol-21-hemisuccinate per gram body weight using Cortland'ssaline as a vehicle. After injection the fish were sampled at 0, 0.5, 1, 2, 24, 48, 72, and 96hr. The fish were sacrificed by a lethal dose (200 mg.1:1) of buffered MS-222. This dosehas been shown to inhibit the stress-related rise in plasma cortisol concentration in salmon(Barton et al. 1986). After the fish were anaesthetized, the weight and length weremeasured and the caudal peduncle was severed and the blood collected in heparinizedcapillary tubes. The plasma was separated by centrifugation and then stored at -80 °C untildetermination of cortisol and sodium concentrations. Gills were removed and treated asdescribed below.Chronic Experiments. The effects of chronic elevation in plasma cortisol on receptornumbers was examined in yearling coho salmon in March, 1992 (Kitimat River Stock). Thefish were silver with little or no visible parr marks. To maintain stable elevations in plasmacortisol concentration and avoid repeated handling of the fish, mini-osmotic pumps (Alzetmodel 1007D, California) were implanted into the peritoneal cavity of anaesthetized cohosalmon. The pumps were filled with cortisol-21-hemisuccinate, dissolved in a saline solutioncontaining 30% molecusol HBP (Pharmatec, Alhua Florida), a steroid miscible carrier. Theconcentration of cortisol placed in the pumps was determined according to the flow rateof the pumps (corrected for temperature), the average mass of the animals, and the desiredplasma cortisol levels. The effectiveness of the pumps in maintaining plasma cortisol levels35for an extended period of time was verified. Four fish were implanted intraperitoneallywith pumps filled with a cortisol solution to achieve plasma cortisol levels of 250 ng.m1-1.Four fish not implanted with pumps served as controls. Animals were sampled prior to and3, 7, 12 and 18 days following insertion of the osmotic pumps. Blood samples wereremoved from the caudal artery using a heparinized syringe, after the fish wereanaesthetized in 100 mg.L-1 MS-222.After the effectiveness of the osmotic pumps had been verified, 45 fish wereimplanted with osmotic pumps intraperitoneally. Fifteen pumps contained saline andmolecusol only, the remaining pumps were divided into two groups and filled with thecarrier vehicle and a concentration of cortisol-21-hemisuccinate sufficient to achieve aplasma cortisol level of 100 ng.m1-1 and 200 ng.m1-1. Five fish that were not implanted withcortisol pumps were sampled to determine if implanting the pumps had any long termeffect on the parameters measured. Each group was maintained in a 200 L tank suppliedwith dechlorinated Vancouver tap water. Mean weight of the fish was 56.9 g. The meanwater temperature during the experiment was 7.5 °C and ranged from 6.0 to 8.9 °C. Basedon the temperature dependent pumping rate, the osmotic pumps infused the cortisolsolution into the fish for 18 days. After 17, 21 and 28 days, five fish from each group weresampled for gill corticosteroid receptors and plasma cortisol according to the protocoloutlined above.Stress Experiments. The effect of endogenous cortisol release on corticosteroidreceptors was examined by stressing the fish. Underyearling juvenile coho parr (KitimatRiver Stock) were used to determine the effects of stress on gill cortisol receptorconcentrations in August 1991. The fish were silver, but still displayed faint parr marks.Eighty fish (mean weight 13.7 ± 0.4 g, SEM) were divided into two 200 L fibreglass tanks.Tank 1 was left undisturbed for 10 days. Tank 2 was acutely stressed daily. One of threestressors was used to prevent accommodation of the fish to a specific stressor; netting,crowding or water removal. A netting stress involved aerial suspension for 45 seconds. Forthe confinement stress, fish were caught in a net and held 0.5 hr at a density such that allfish were in physical contact with each other. In the third stressor, the water was drainedfrom the tank until the fish were stranded and on their sides. After 60 s the water was36refilled. All methods have been shown to cause a significant rise in plasma cortisol (Bartonet al. 1987). At the end of the 10 day treatment, the fish were left undisturbed for threedays and then sacrificed as described above.To determine if stress caused changes in corticosteroid receptors that persisted forup to 6 days, the protocol outlined above was repeated and the fish left undisturbed for 6days and then sampled. Coho fry were used in the experiment (mean weight was 14.9 .±2.9 g).Tissue Preparation. Gills were excised from the fish, the cartilage was removed, thetissue was blotted on paper tissues to remove excess blood and then placed in a solutionof 10 mM Tris-HCI, 1 mM EDTA, 12 mM monothioglycerol, 20 mM sodium molybdate,and 10% (v/v) glycerol all at pH 7.4 (TEMS). Approximately 0.5 g of gill tissue were addedto 2.5 mL of TEMS and stored in liquid nitrogen until samples were assayed forcorticosteroid receptors. Tissues from up to 5 individuals were pooled for one sample,dependent on the size of the fish. The protocol for CR analysis is given in Chapter 2.Cortisol Radioimmunoassay. Plasma cortisol levels were determined byradioimmunoassay kit using 'I (Incstar Corp.).Plasma Sodium Analysis. Plasma samples were analyzed for sodium using a Perkin-Elmer model 2380 atomic absorption spectrophotometer.Statistical Analysis. The concentration of corticosteroid receptors, kr,, and Hillcoefficient, haematocrit, plasma cortisol and sodium levels were compared for significantdifferences using a T Test or analysis of variance (ANOVA), followed by a Tukey's test tofind significant differences between means. Statistical significance was taken at a level ofa = 0.05. All values are expressed as means ± 1 s.e..37RESULTSSingle Injection of Cortisol. Resting levels of cortisol in plasma of coho salmon were16.36 ± 10.31 ng.mL-1. Intraperitoneal injection of cortisol-21-hemisuccinate showed asignificant effect on circulating plasma cortisol concentration (P < 0.001, n = 7). Therewas a 1000 fold increase in plasma cortisol by the first sampling period, 30 minutes (Figure3.1). Cortisol levels declined after this time, but remained significantly elevated for 24 hrpost injection (PI). There was no significant difference between levels of cortisol measured48, 72 and 96 hr PI and the pre-injection measurement.Following injection of cortisol there was a significant decline (P < 0.001, n =5) incorticosteroid receptor numbers in the cytosol of the gill by 30 minutes (Figure 3.1). CRlevels remained significantly lower than pre-treatment values for 48 hr. The 72 and 96 hrvalues did not differ significantly from the time 0 measurement. The injection treatmenthad no effect on kr, (P > 0.1) and the Hill Coefficient (P > 0.1).After cortisol injection there was a significant increase in plasma sodiumconcentration by 24 hr PI (P < 0.001, n = 7). The concentration gradually declined to alevel that did not differ significantly from pre-injection by 96 hr PI (Figure 3.2). There wasno significant difference in haematocrit values over the duration of the study.Chronic Cortisol Treatment. Implanting osmotic pumps filled with cortisolintraperitoneally raised plasma cortisol concentration as shown in Figure 3.3. By three daysthe plasma cortisol level had increased to approximately 250 ng m1-1. The cortisol remainedfairly stable, with an average of 257.6 ± 19.9 ng m1-1.After circulating cortisol levels had been elevated for 17 days, five fish from eachdose group were sampled. The binding of TA to the gill homogenate was significantlydifferent between the three groups as shown by Scatchard analysis (Figure 3.4). The shamtreated group did not differ significantly from the fish not implanted with mini-osmoticpumps (T=0). On day 17, the sham treated animals possessed the greatest number ofreceptors as indicated by the intercept on the X-axis, and there was a dose response toelevated plasma cortisol as the high dose group possessed fewer receptors than the low38dose group. The slopes of the curves also differed between the groups, the slope decreasedas the treatment dose of cortisol increased. The change in slope is an expression of achange in the receptor affinity for the hormone, indicating that the increase in cortisollevels decreased receptor affinity. Comparison of the kr, calculated from the slopes of theScatchard plots indicated a significant difference among the three groups.At the experimental water temperature, the osmotic pumps continued to pump for18 days. The fish were sampled before and after circulating plasma cortisol levels hadreturned to a level not significantly different from the sham treated animals (Figure 3.5).Prolonged cortisol treatment had a significant dose dependent effect on the total numberof cortisol receptors in the gills (P < 0.001). As the plasma cortisol concentration in theexperimental fish returned to a level that did not differ significantly from the shamimplanted fish, the reduction in CR at 3 and 10 days following the cessation of osmoticpump activity shows that cortisol downregulates gill CR. The difference in B. between thehigh and low dose groups was significant for the duration of the study. Cortisol treatmentalso had a significant dose dependent effect on the affinity of the corticosteroid receptorsfor TA (P < 0.001). The high dose of cortisol caused a significantly greater increase in kr,than the low dose of cortisol. This change in affinity, however, did not persist for as longas the change in receptor concentration. Three days after the drop in cortisol levels, therewas no longer a significant difference in kr, between the low and high dose groups. After10 days there was no significant difference between the cortisol treatment groups and thesham group. Cortisol treatment had no effect on the Hill coefficient (P = 0.506). Theeffect of cortisol on corticosteroid receptors is to decrease affinity and concentration whilecortisol levels are elevated. The decrease in receptor concentration persists following adecline in plasma cortisol, indicative of a downregulation of corticosteroid receptors.After cessation of cortisol infusion, the plasma sodium concentration did not differsignificantly between the experimental and sham treated fish (Figure 3.6). The haematocritvalues were also not significantly different between the groups.Relation between plasma cortisol and corticosteroid receptor concentration. As theconcentration of cortisol in the plasma increased the number of unbound or free receptorsdeclined (Figure 3.7). This decline appeared to have two stages, an initial rapid decline in39B. with increasing plasma cortisol and a subsequent insensitivity to increases in plasmacortisol. The level of cortisol that resulted in the plateau of receptor numbers is very high,approximately 300 ng.m1-1. Below 300 ng.m1-1, increasing levels of cortisol caused a declinein B.. kr, affected the magnitude of the decline in B. with an increase in cortisol levels.The higher the kr,, the more gradual the change in B. with an increase in plasma cortisolconcentration. At low plasma cortisol levels there is a strong relationship between kr, andB.; fish with higher kl, possess fewer receptors. At higher plasma cortisol levels thedifference between B. for fish with high or low kr, is not as great. This relationship ispresented in Figure 3.8. A multiple linear regression was performed on the data. Therelationship between plasma cortisol (Cp), kr, and B. is expressed in Equation 1.B. = 103.1 - 0.343 Cp - 24.73 IED + 0.106 Cp kr,^(1)R2 for the relationship is 0.77. n = 40, p <0.005.Daily Acute Stress. Following 10 days of acute handling stresses, the fish were leftfor 72 hr to allow circulating plasma cortisol to return to basal levels. As the plasmacortisol concentration was low, it is likely that the majority of the CR concentration wouldbe present in the unliganded form and could be detected using the radioreceptor assayemployed. Daily acute stress resulted in a significant reduction in B. compared to controlfish (P < 0.001, Figure 3.9). Other than the change in haematocrit, there was no otherdifference between parameters found in the experiment; kJ), Hill Coefficient, and plasmasodium concentration were not affected by the stress (Table 1).The daily stress experiment was repeated to determine if the receptor populationwould remain depressed for up to six days following cessation of the stresses. There wasa significant reduction in B. in the stressed fish when compared to the control fish (P <0.025, Figure 3.9). This was the only significant difference in the parameters measured.Daily stresses, therefore, lead to a persistent decline in the number of cortisol receptors inthe gills, without any apparent change in the affinity of the receptor for the hormone.40Figure 3.1. Changes in plasma cortisol levels (n = 7), and corticosteroid receptor (n =5) concentration (B.) and dissociation constant (lcD) in the gills following a singleintraperitoneal injection of cortisol-21-hemisuccinate. * indicates data points aresignificantly different from initial measurements (a = 0.05).155150T--I 1450"I'lE1401351 141I^I^I^II^I^i^I^i**-*--10^1^2^24^48^72^96TIVi_hoursFigure 3.2. Plasma sodium concentrations (n = 7) following an intraperitoneal injectionof cortisol. * indicates data points are significantly different from initial measurements (a= 0.05).350___i0^300(/)1—^250i0 ---E- 200C.)< (7), 150Cco^100<Cl_^500420^5^10^15^20THiEdaysFigure 3.3. Verification of the effectiveness of mini osmotic pumps filled with cortisol-21-hemisuccinate implanted into the peritoneal cavity at elevating plasma cortisolconcentration over 18 days (n = 4). • fish implanted with mini osmotic pumps containingcortisol, 0 fish without mini osmotic pumps.431 0 080I LJJLi_^604002000^20^40^60^80SPECIFICALLY BOUND STEROIDfmol • mg Protein'Figure 3.4. Scatchard plot of [3H]TA bound to gill homogenate from coho salmonimplanted with osmotic pumps for 17 days. Symbols represent means of five samples,vertical and horizontal lines indicate standard error of the mean.44200150(7) 100500TC1 00X CL 800^icr)E E6040201.41.20 M1.0C0.80.60.40 5^10^15^20TIMEdays25^30Figure 3.5. Changes in plasma cortisol levels, and corticosteroid receptor concentration(B..) and dissociation constant (1cD) in the gills of coho implanted with mini osmoticpumps containing high and low doses of cortisol-21-hemisuccinate or vehicle only (n = 5).For the experimental water temperature, the duration of pumping for the osmotic pumpswas calculated to be 18 days. * indicates data points are significantly different from shammeasurements at the same sampling time (a = 0.05).155150145140135450^5^10^15^20^25^30TIMEdaysFigure 3.6. Plasma sodium concentrations of fish chronically exposed to elevated plasmacortisol (n = 5). Cortisol levels in the plasma declined after approximately 18 days.46Figure 3.7. The relationship between plasma cortisol concentration and corticosteroidreceptor concentration in the gills of coho salmon.12047SoV)^100 -C0i—ta_LLI07 80-iil .cCK0 0._ 60-(5- EL.,I-- 73(r) E 400011.-:ci0^20 -0-2.52.0100PLASMA^300250CoRnsoLt-79 . riii,,150 200 0.5\4350^0.0 04Figure 3.8. The relationship between corticosteroid receptor concentration (B.) anddissociation constant OW for plasma cortisol concentrations below 300 ng.m1:1.120100_(1)080cy)60-6E 40Lfa 20co048CONTROL STRESS^CONTROL STRESS3 DAYS 6 DAYSDAYS POST STRESSFigure 3.9. The concentration of gill corticosteroid receptors (Br.) from juvenile cohosalmon, 3 days and 6 days after fish had been subjected to a daily handling stress for 10days (n = 5). * indicates the experimental group differs significantly from the controls(a= 0.05).49Table 1. Plasma levels of cortisol, sodium and haematocrit, and characteristics of gillcorticosteroid receptors from control and stressed juvenile coho salmon.TREATMENT CONTROL STRESStPlasma (n=12)Cortisol (ng.mL-1) 5.5 ± 2.27g 1.9 ± 0.62Sodium (mEq.1:1) 141.2 ± 1.20 142.5 ± 0.91Haematocrit (%) 42.3 ± 1.63 46.7 ± 1.43*Gill Corticosteroid Receptors (n=5)Concentration ( Protein-1) 96.69 ± 1.734 63.49 ± 4.947**Dissociation Constant (nM) 0.77 ± 0.032 0.76 ± 0.035Hill Coefficient 1.06 ± 0.005 1.05 ± 0.014tAnimals were sampled 3 days following the stress period. gAll values are means ± s.e.* and ** indicate values are significantly different from controls at confidence levels of a= 0.05 and a = 0.01, respectively.50DISCUSSIONThe number of hormone receptors is an important parameter in the control ofphysiological mechanisms, as the sensitivity of cells is dependent on the receptorconcentration (Danielsen and Stallcup 1984). This study has shown that exogenous andendogenous elevation of cortisol will cause a downregulation of corticosteroid receptors inthe gills, making the gills less sensitive to cortisol. Consequently, the response of gills tothe cortisol surge associated with smolting is dependent on prior exposure of the tissue tocortisol. This decreased sensitivity of the gills to cortisol may be a mechanism for thedecreased ability of stressed fish to smolt (Wedemeyer and McLeay 1981; Schreck et al.1985; Patino et al. 1986).Acute cortisol treatment results in a rapid decline in unbound corticosteroidreceptors in the gills. The depletion of CR in the first 30 min is likely due to binding ofthe injected cortisol to CR. I do not know whether the hormone-receptor complex isbound to DNA within 30 min. The time course for glucocorticoid receptors to becomeaffiliated with the nuclear fraction, however, is 30 min (Eisen et al. 1988). In chinooksalmon acutely stressed, cortisol levels rise and are maximal approximately three hr afterthe stress, and return to a pretreatment level within 12 hr (Barton et al. 1986). Thechange in CR concentration following acute treatment, is also transient and only lasts fora few days. This would not be expected to downregulate CR in the gills. On the contrarythe depletion in CR in the gills following acute treatment would indicate activation andnuclear binding of the receptor-hormone complex. Thus, in the short term, the release ofcortisol due to stress will likely lead to a stimulation of physiological changes that enhancesaltwater tolerance. This response has been seen in fish reared at high density that initiallyshow an increase in Na+IC+ATPase activity (Shrimpton and Randall, unpublished data).The repeated rise in cortisol following daily stress treatments, however, will have aninhibitory effect on the responsiveness of the gills to cortisol as CR concentrations remaindepressed.Chronic elevations in cortisol also result in a downregulation of corticosteroid51receptors. The continual elevation in cortisol produces two forms of downregulation, adecline in the concentration of receptors and a decrease in the affinity of the receptors(Figure 3.5). The decline in receptor concentration following hormone treatment is acommon phenomenon, and the autoregulation of mineralocorticoid and glucocorticoidhormone receptors has been demonstrated for intact mammals (Claire et al. 1981;Sapolsky et al. 1984; Svec 1988) and mammalian cell cultures (Cidlowski and Cidlowski1981; Sveck and Rudis 1981; Danielsen and Stallcup 1984). A number of mechanismscould cause downregulation of CR in the gills of coho salmon. Downregulation of CR maybe due to the degradation of receptors as demonstrated by Svec and Rudis (1981).McIntyre and Samuels (1985) showed that glucocorticoid treatment using TA caused areduction in the half-life of the activated glucocorticoid receptor, and indicate that thereduction in half-life occurs within the first hour of incubation with TA, and remainsconstant for greater than 24 hr. They propose that the change in receptor half-life accountsfor the reduction in receptor levels, and suggest that TA had little or no effect on receptorsynthesis rate. In contrast, treatment of cell cultures (Rosewicz et al. 1988) and intact rats(Kalinyak et al. 1987) with the glucocorticoid dexamethasone, has resulted in a decreasein mRNA for the glucocorticoid receptor. The decrease in mRNA is maximal by six hr,and persisted of up to 72 hr in the presence of dexamethasone (Rosewicz et al. 1988).There is no indication of how quickly the mRNA concentrations will increase following theremoval of the glucocorticoid. Thus downregulation of CR in salmonid gills may be aresult of both a reduction in mRNA for CR, and a degradation of existing receptors.The change in affinity appears to be a direct result of high hormone levels as it doesnot persist. The change in affinity found is not an artifact of the methodology. Thepresence of unlabelled ligand during Scatchard analysis has little consequence on theestimation of the number of receptors, but lowers the apparent affinity for receptor ligandinteraction. In the experiments examining acute administration of cortisol, the cortisollevels in the plasma reached four times physiological levels and were 10 times greater thanthe cortisol levels for the chronic elevation in cortisol experiments. Therefore, washing ofthe endogenous steroid from the homogenate by dextran charcoal was effective as nochange in kn, was seen for the acute experiments. The washing method used, should have52been effective enough to remove all the endogenous steroid in the chronically treatedgroups.Maule and Schreck (1991) showed that a single meal of food containing cortisol (100mg • kg food-1) resulted in a decrease in affinity and numbers of CR in gills. A similardecrease in affinity and CR concentration was found in the liver of rainbow trout followingdexamethasone administration (Lee et al. 1992). In the present study, acute cortisolelevation did not result in any change in receptor affinity. Even after ten days of stress,there was no change in receptor affinity However, chronic cortisol treatment resulted ina downregulation of CR by a reduction in numbers and affinity, without a change in theHill coefficient. The decline in numbers and affinity would have an additive effect onreduction in tissue sensitivity. In rats, changes in affinity of glucocorticoid receptors havebeen found between tissues (De Nicola et al. 1989) and in response to stress. Hirota etal. (1985) describe the presence of a second type of receptor in the liver when rats werestressed or treated with dexamethasone. The new receptor population had a lower affinity,but possessed multiple binding sites. Matic et al. (1989) found that hyperthermic stress inadrenalectomized rats results in a decrease in kr, and B.. The changes in binding capacitywere independent of endogenous steroids. In the present study the changes in bindingcapacity were due solely to the elevation in cortisol level, as the experimental groups werenot stressed any more than the control fish. The changes in affinity described in themammalian and fish receptors suggests that there are either multiple CR genes or post-transcriptional changes in CR. It is likely that post-transcriptional modulators account forthe change in affinity, as ku, returned to pretreatment levels without a significant increasein receptor population.A relationship exists between CR and plasma cortisol concentration. Thisrelationship, however, is dependent on the kp. Fish that possess lower affinity receptorshave lower absolute numbers of CR, but the effect of elevated plasma cortisol ondecreasing CR is reduced. Several other researchers have shown a relationship betweenplasma cortisol levels and binding capacity in fish. Pottinger (1991) and Weisbart et al.(1987) also found that cortisol caused a marked decline in the number of receptors in theliver of rainbow trout and the gills of brook trout. In both of these publications there is53no indication that affinity of the receptors changes with hormone exposure. In my study,the data for this relationship were determined from fish sampled at different times of theyear. The difference in B., and kr, over the year would indicate that the responsivenessto elevation in plasma cortisol will vary over the year.There is a limit to the depletion of CR by cortisol, in the gills. At cortisol levels inthe plasma above 300 ng.m1-1, there was no further reduction in CR number despiteincreases in plasma cortisol concentration. This "floor effect" has also been reported bySapolsky et al. (1984) and Kalinyak et aL (1987). The "floor effect" may be a result oflimited binding to the DNA in the nucleus by the hormone-receptor complex. If thereceptors are not bound in the nucleus, they may not be discarded in the nuclear pelletfollowing centrifugation. When the supernatant is washed with DCC, cortisol may bestripped from the receptor and CR may be detected by the radioligand assay employed inthe present study. Weisbart et al. (1987), however, did not show a "floor effect" with CRfrom the gills of brook trout, as cortisol administration completely abolished the detectionof CR in the cytosolic fraction. I am not certain why this discrepancy between the resultspresented here and those of Weisbart et al. (1987) exists.The cortisol treatment caused changes in the plasma sodium concentration.Increased plasma sodium concentration 24 hr post single cortisol injection may beassociated with a rise in Na+K+ATPase activity at the gill (and kidney). McCormick andBern (1989) showed that cortisol can increase ouabain binding sites in coho salmon gilltissue in vitro. Lin and Randall (1991) proposed a model that indicated Na uptake acrossthe gills requires Na+K+ATPase in freshwater rainbow trout. The single bolus may haveenhanced Na uptake. Redding et al. (1984a) found that chronic treatment of yearling cohowith cortisol lowered gill Na+K+ATPase activity in fresh water. This would be consistentwith a decline in plasma sodium after chronic cortisol treatment. Another explanation forthe change in plasma sodium found in the present study is the interaction of cortisol withother hormones. There is some evidence that cortisol may have an inhibitory effect onprolactin. Prolactin decreases membrane permeability and aids in retaining sodium. Naesset al. (1980) found that cortisol treatment reduced prolactin production in mammaliantissue. Avella et al. (1990) found a negative relationship between cortisol and prolactin54during seawater challenge of coho smolts. However, chronic stress caused a gradualincrease in circulating prolactin and cortisol (Avella et al. 1991).Cortisol regulation of the concentration and affinity of corticosteroid receptors inthe gills of coho salmon has been shown. Stress is associated with elevated plasma cortisollevels. Chronic elevation in cortisol concentration in the plasma leads to a downregulationof CR, and this could account for differences seen in CR concentration in hatchery andwild juvenile coho (Chapter 5). It is possible that a stress associated downregulation of CRis one mechanism for stress to impair the physiological changes that occur as juvenilesalmonids smolt. The reduction in CR results in a tissue that is less sensitive to the risein cortisol secretion associated with smolting. A reduction in the number of CR in the gillsdoes not result in a significant increase in Na+IVATPase activity (Chapter 5) and chloridecell concentration (Chapter 7) in the gills and consequently results in impaired saltwatertolerance (Chapter 7).CHAPTER 4Increased Gill Corticosteroid Receptor Concentrationand Saltwater Tolerance in Juvenile Coho Salmon(Oncorhynchus kisutch) Treated withGrowth Hormone andPlacental Lactogen5556INTRODUCTIONWhen juvenile salmonids smolt, a series of complex physiological changes occur thatpre-adapt them to life in the marine environment. While still in fresh water, fish respondto environmental stimuli by hormonal changes that cause morphological, biochemical andphysiological changes (Hoar 1988). Many of these changes occur in the gills. Increasesin chloride cell size and numbers, and Na+K+ATPase activity enhance hypo-osmoregulatoryability (Hoar 1988). Cortisol has an important role in stimulating an increase in saltwatertolerance in salmonids (Madsen 1990a). Cortisol concentrations in the plasma surgeduring the spring as the fish smolt (Specker and Schreck 1982; Chapter 5). Cortisol bindsto intracellular corticosteroid receptors (CR) in the gills (Maule and Schreck 1990) andlikely stimulates gene transcription and protein synthesis (Clark 1985). In teleosts, thereis evidence that cortisol treatment increases chloride cell concentration (Perry and Laurent1989; McCormick 1990) and the enzyme responsible for salt secretion, Na+VATPase, invivo (Richman and Zaugg 1987; Madsen 1990a) and in vitro (McCormick and Bern1989).The concentration of growth hormone (GH) has been shown to increase during thespring when fish smolt (Hoar 1988; Prunet et al. 1989; Young et al. 1989a).Administration of Gil has also been found to produce an increase in saltwater tolerancesimilar to that seen in smolts (Komourdjian et al. 1976; Clarke et al. 1977; Miwa andInui 1985). This increase in saltwater tolerance is independent of the growth promotingactions of GH (Bolton et al. 1987a; Madsen 1990c). These findings indicate a possiblerole for Gil during smoltification.Although Gil and cortisol have an effect on saltwater tolerance individually,treatment of salmonids with both hormones together has an additive effect on increasingchloride cell numbers, Na+K+ATPase activity, and hypo-osmoregulatory ability (Madsen1990b,c). The method of interaction of growth hormone and cortisol has not been showndefinitively. Despite the increased sensitivity of the interrenal tissue to ACTH followingGil treatment (Young 1988), there is little evidence that Gil will affect smolting by57increasing plasma cortisol concentration directly. With the exception of Ball and Hawkins(1976), growth hormone administration has not been found to have any effect on plasmacortisol concentration in teleosts (Langhorne 1976) or mammals (Sillence and Etherton1989; Enright et al. 1989). There is evidence, however, that plasma cortisol concentrationhas an effect on GH concentration in the plasma (Pickering et al. 1991).Interaction between GH and cortisol may occur at the receptor level and cause achange in sensitivity of the tissue. The concentration of receptors is altered by hormonelevels in the plasma; for example CR concentration and affinity are altered by cortisoltreatment (Maule and Schreck 1991; Chapter 3). It is possible that other hormones mayalso play a role in the regulation of CR. Regulation of receptors by hormones other thanthe native ligand has been shown. GH, T4, and cortisone will increase GH receptorconcentration in mammals (Gause and Eden 1986). In the present study, the effect ofdifferent bovine somatotropins on cortisol concentration in the plasma, and corticosteroidreceptor concentration and affinity in the gills, were examined. Juvenile coho salmon(Oncorhynchus kisutch) were treated with bovine growth hormone (bGH) or bovineplacental lactogen (bPL). Placental lactogen is produced in the placenta of mammals andaffects growth of the fetus (Handwerger and Brar 1992). bPL and bGH have been shownbind to the salmon growth hormone receptor and to significantly stimulate growth in cohosalmon (Devlin et al. 1993). These two bovine hormones were used in the absence ofsalmon growth hormone. Due to the importance of GH and cortisol in stimulating hypo-osmoregulatory ability, I chose to examine the effect of these hormones on osmoregulationby measuring Na+IVATPase activity following hormone treatment and plasma sodiumconcentration in fresh water and after 24 hr in salt water.58MA IERIALS AND METHODSExperimental Animals and Protocol.Series I, Single Treatment. The effect of a single injection of bovine growth hormoneand placental lactogen on gill CR concentration was determined on juvenile coho salmon(Oncorhynchus kisutch). Fish were anaesthetized in 25 mg.L-1 MS-222 (tricainemethanesulfonate) buffered with 50 mg.L-1 of sodium bicarbonate. The fish were injectedintraperitoneally with one of seven treatments; 5.0, 0.5, 0.05 g bPL per gram body weight,5.0, 0.5 tig bGH per gram body weight, 5.0 g bSA per gram body weight, and saline as acontrol. All treatments used 0.9% saline as a vehicle. Twenty-five fish were injected foreach group. After injection the fish were placed in one of seven randomly assigned 200 Lfibreglass tanks for one week. On October 16, 1992, the fish were sacrificed by a lethaldose of buffered MS-222 (200 mg.L-1). This dose has been shown to inhibit the stress-related rise in plasma cortisol concentration in salmon (Barton et al. 1986). After the fishwere anaesthetized, the weight and length were measured and the caudal peduncle wassevered and the blood collected in heparinized capillary tubes. The plasma was separatedby centrifugation and then stored at -80 °C until determination of cortisol and sodiumconcentrations.For determination of corticosteroid receptors, gills were excised from the fish, andblotted on tissues to remove excess blood and then placed in a solution of 10 mM Tris-Ha,1 mM EDTA, 12 mM monothioglycerol, 20 mM sodium molybdate, and 10% (v/v) glycerolall at pH 7.4 (TEMS). Approximately 0.5 g of gill tissue were added to 2.5 mL of 'TEMSand stored in liquid nitrogen until samples were assayed for corticosteroid receptors.Tissues from up to five individuals were pooled for one sample, dependent on the size ofthe fish. Gill samples were analyzed for CR using a radioreceptor assay as described inChapter 2. Calculation of the concentration and affinity of the corticosteroid receptors inthe gill samples was performed according to the method of Scatchard (1949).For determination of Na÷K÷A'TPase activity, gills from one side of a fish wereremoved and blotted dry on tissues to remove excess blood, and were then placed in a59solution of sucrose, EDTA and imidazole. The samples were stored at -80 °C until analysisof Na+IVATPase activity. Approximately 50 mg of gill tissue from each fish was removedand placed into 0.5 ml of ice cold SEI (0.3M sucrose, 0.02M EDTA and 0.1M Imidazole)for determination of Na+IVATPase activity. Analysis of Na+K+ATPase activity in the gillsamples was determined according to the protocol of Zaugg (1982). Protein content in thegill homogenates was measured using the Biuret assay (Alexander and Ingram 1980).Series 2, Six Weekly Treatments. The effect of repeated intraperitoneal injection ofbGH and bPL were determined on juvenile coho. Forty fish were injected per group. Thesame treatment groups were used as described above, with the exception that the controlgroup was untreated. Each week the fish were measured for weight and length, and thensubjected to one of the seven treatments for six weeks. At the end of the six weektreatment, the fish were left for one week and then sampled (December 18, 1992) asdescribed in the series 1 experiment. In addition, 16 - 18 fish from each group weretransferred to salt water (salinity 25%.0) for 24 hr to assess survival and changes in plasmasodium concentration. Following the saltwater challenge, the fish were killed byconcussion, then measured for weight and length. The caudal peduncle was severed, bloodcollected and treated as above. Plasma samples were stored at -80 °C until analysis forsodium concentration.Cortisol Radioimmunoassay. Plasma cortisol levels were determined byradioimmunoassay kit using 115I (Incstar Corp.).Plasma Sodium Analysis. Plasma samples were analyzed for sodium using a Perkin-Elmer model 2380 atomic absorption spectrophotometer.Statistical Analysis. Condition Factor (CF) was calculated as (weight*100).length'.Fish weight, length, CF, Bm, kp, Hill coefficient, plasma cortisol and sodium levels werecompared for significant differences using analysis of variance (ANOVA), followed by aTukey's test to find significant differences between means. Statistical significance was takenat a level of a = 0.05. All values are expressed as means :IL 1 s.e..60RESULTSGrowth. Treatment of juvenile coho salmon for six weeks with bPL and bGH hada significant effect on fish size (Figure 4.1). The two higher doses of bPL and the highdose of bGH resulted in a significant trend in weight increase (Figure 4.1a). All groupstreated with bPL and bGH showed a significant dose dependent increase in length, whencompared to the control groups (Figure 4.1b). In addition, there was an increase in thecondition factor for the two higher doses of bPL (Figure 4.1c).Plasma Cortisol Concentration. The concentration of cortisol in the plasma did notdiffer significantly from the controls in any of the bGH injected groups, or the groupsinjected with the two higher doses of bPL (Figure 4.2). There was a dose dependent effectof bPL on plasma cortisol. In the single injection experiment, 0.05 bPL and 5.0 bPL groupshad plasma cortisol concentrations significantly greater than the 0.5 bPL group. Injectionwith bSA had no effect on plasma cortisol.Corticosteroid Receptor Concentration and Affinity. Bovine growth hormone andplacental lactogen had a significant effect on CR in the gills one week following a singletreatment (Figure 4.3). Fish injected with the high dose of bGH and the middle dose ofbPL exhibited a significant increase in CR concentration, B. (Figure 4.3a). CRconcentration was not significantly different in the bPL 5.0 fish than the control fish. TheCR affinity was significantly reduced in the low dose of bPL group (an increase in kr),Figure 4.3b). All the other experimental groups had kr) values that were not significantlydifferent from the control group. There was no significant difference between the Hillcoefficients for any of the groups.After six weeks, Bmax values in all the experimental groups except the low dose ofbPL were greater than the controls (Figure 4.4a). The two higher bPL doses and the highbGH dose were significantly greater than in the controls. The experimental groups did notshow kp values that were significantly different from the control and bSA groups (Figure4.4b). None of the treatments had any effect on the Hill coefficient.NeiC+ATPase Activity. Following a single treatment, there was a significant increase61in Na+K+ATPase activity in just the 0.5 bPL group (Figure 4.5a). No other group differedsignificantly from the saline or bSA treated fish.After six weekly treatments, there was a significant increase in Na+K+ATPase activityin all the groups of fish treated with bGH and bPL (Figure 4.5b). There was a dosedependent relationship between the increase in enzyme activity and the increase intreatment dose. The fish treated with bPL showed approximately a 10-fold greater increasein enzyme activity than the fish treated with bGH.Plasma Na÷ Concentration. In fresh water, the plasma sodium concentration was notaffected significantly by a single injection of bGH or bPL, when compared to bSA anduntreated fish (Figure 4.6). The single bPL 5.0 treatment, however, did result in a slightlygreater sodium concentration than the other treatments. Following six weekly treatments,the bPL 5.0 group plasma sodium values were significantly greater than the bSA group(Figure 4.7). In fresh water, therefore, treatment with bPL 5.0 caused a moderate increasein plasma sodium concentration.There was no mortality in any of the groups of fish during the 24 hr saltwaterchallenge. In comparison to the freshwater levels, the concentration of plasma sodiumincreased following saltwater exposure in all groups (Figure 4.7). Saltwater plasma sodiumvalues of the fish treated with bGH and the high dose of bPL, however, did not differsignificantly from the freshwater plasma sodium concentrations. The remaining four groupseach showed a significant increase in plasma sodium concentration following saltwaterexposure. The plasma sodium concentration of the saltwater fish was significantly less inthe bGH groups and the bPL 5.0 group than the control or bSA groups (Figure 4.7). Ofthe experimental groups, only the 0.05 bPL group did not differ significantly from the bSAgroup in plasma sodium concentration.A relationship between fish size and plasma sodium concentration existed for thefish in freshwater (Figure 4.8a) and following 24 hr saltwater exposure (Figure 4.8b). Therelationships determined were:[Na] = 119.17 + 2.71 * LN [FW][Na] = 198.94 - 3.23 * LN [SW]R2 for the freshwater relationship was 0.263 and for the saltwater relationship was 0.153.62Within most of the groups, however, there was no significant relationship between size andplasma Na+ concentration. The exceptions were the bGH 5.0 and bPL 5.0 freshwatergroups. The X coefficients were 4.545 and 4.049 for the bGH 5.0 and bPL 5.0 groups,respectively, and indicated that plasma sodium was positively correlated with fish size. R2values for the two relationships were 0.401 and 0.581 for bGH 5.0 and bPL 5.0, respectively.When transferred to saltwater, the trend between fish size and plasma sodiumconcentration not significant for either.- *- *)^1*-***ib-*-*I 1 ITM I 1 - 7 - ICtil***:;:',i::::***3*xoN4444444*16325H-20CDi I-151010-C--__1 .09LL_ 1.06C) 0.50 0.05^5.00 0.50^bSA control----b P L bG HFigure 4.1. The size of the juvenile coho following six weekly treatments of bovineplacental lactogen (bPL) at 5.0, 0.5, and 0.05 pg.g-1 body weight and bovine growthhormone (bGH) at 5.0 and 0.5 pg.g-1 body weight. The control groups consisted of a groupinjected with 5.0 tig bovine serum albumin (bSA) per gram body weight and a controlgroup that was not injected. Weight and length are in g and cm respectively. CF is thecondition factor. * indicates values that are significantly different from the control group(a=0.05). Vertical bars are 1 SE. n = 40 for each group, except bSA where n = 17.4, 1-T1_-1ab-f 777^/^I--------z800R---<601-zLI^40(..)z0 20C...)^1___,1 E^00v) 0)F- cCh^600CD<Cr)<^20___I11._064405.00 0.50 0.05 5.00 0.50 bSA controlbPL bGHFigure 4.2. The plasma cortisol concentration of juvenile coho salmon injected once (a)or six times (b) with bPL or bGH. n = 12 for each treatment. See figure 4.1 fordescription of treatment groups and other symbols.(f)^1200 0ELiJ0_ OD100Lil0^80LUF—^0.9(r)0o0.800.70.6655.00 0.50 0.05 5.00 0.50 bSA controlbP L bGHFigure 4.3. The effect of a single injection of bPL or bGH on corticosteroid receptorconcentration (B..) and affinity (1cD) in the gills of juvenile coho salmon. Units for B.are fmol mg protein-1, and for IED are nM. n = 5 for each treatment. See figure 4.1 fordescription of treatment groups and other symbols.6610075cm500LiJ 0.6Cl)000.500.45.00 0.50 0.05^5.00 0.50^bSA controlbPL bGHFigure 4.4. The effect of six weekly injections of bPL or bGH on corticosteroid receptorconcentration (B.) and affinity (JO in the gills of juvenile coho salmon. Units for B.are fmol • mg protein-1, and for kr, are nM. n = 5 for each treatment. See figure 4.1 fordescription of treatment groups and other symbols.5.00 0.50 0.05 5.00 0.50 bSA controlbPL bGHFigure 4.5. The activity of Na+KfATPase in the gills of coho salmon treated with bPL orbGH for one (a) or six (b) weekly treatments. n = 10 for each treatment. See figure 4.1for description of treatment groups and other symbols.671551405.00 0.50 0.05 5.00 0.50 bSA controlbPL bGHFigure 4.6. Plasma sodium concentrations for coho in fresh water one week following asingle injection of bPL or bGH. n = 12 for each group. See figure 4.1 for description ofthe groups.68Figure 4.7. Plasma sodium concentrations after six weekly treatments with bPL or bGH in fresh water (n = 12) and afterexposure to salt water for 24 hr (n = 16-18). * indicates significant difference exists between fresh and salt water values forthe same group. + indicates that value is significantly different from the control and bSA groups for the same treatment.= indicates that value is significantly different from the bSA group. See figure 4.1 for description of the groups.ONb PL bGH70O^160I°^150LiJz140LLJU .1800170160(f)150_J1405.00 0.50 0.05 5.00 0.50 bSA control• 010^11^12^13^14LENGTHc mFigure 4.8. The relationship between the length of the fish and plasma sodiumconcentration (a) in fresh water and (b) following a 24 hr saltwater challenge. le for therelationships are 0.263 (FW) and 0.153 (SW). See figure 4.1 for description of treatmentgroups and other symbols.71DISCUSSIONGrowth of juvenile coho salmon increased after injection with bPL and bGH.Placental lactogen showed a much greater potency than bGH for stimulating growth injuvenile coho salmon, confirming the results of Devlin et al. (1993). The greater potencyof bPL for stimulating growth was attributed to the higher affinity of the salmonid liver GHreceptors for bPL than bGH (Devlin et al. 1993). The difference between these twohormones on specific growth rate and feed conversion efficiency is described in detail byDevlin et al. (1993). Both weight and length showed a significant increase, although for thetwo higher doses of bPL, the increase in weight was greater than the length gain asindicated by the significantly greater condition factor (CF, Figure 4.1). The increase in CFis in contrast to that shown by other researchers examining the effect of growth hormoneon specific growth rate of fish (Komourdjian et al. 1976; Clarke et al. 1977). Thedifferences may be attributable to the hormone used, as the effect of bPL has not beenexamined on fish and the physiological actions of bPL may differ from that of bGH. ThebGH groups, however, also showed CF values greater than the controls. The differencescan not be accounted for by feed ration as the experimental animals in this and otherstudies were fed to satiation.Treatment of the fish with bGH and bPL resulted in a change in appearance of thefish. The fish appeared slightly paler in appearance than the control or bSA treated fish,although, parr marks were still visible. This finding is similar to that of Komourdjian et al.(1976), who also noticed a lack of silver coloration, but did notice a slight yellowappearance. Miwa and Inui (1985) found ovine GH increased the occurrence of silveringin amago salmon in an experiment using a similar protocol that lasted for ten weeks. Atsix weeks, however, parr marks were still visible in most fish, and the description of the fishappearance was similar to the present study. The duration of my study, therefore, may nothave been long enough for the silver smolt appearance to develop in the test animals. Theincrease in silver coloration with GH treatment may be a result of a thyrotropic effect byGH (Miwa and Inui 1985; Milne and Leatherland 1978).72Treatment with bPL had a significant effect on plasma cortisol concentration (Figure4.2). The effect, however, was dependent on the dose. Only 0.05 bPL had a significanteffect on plasma cortisol in fish sampled one week following treatment. Growth hormonehas been shown to increase the sensitivity of the interrenal tissue to ACTH (Young 1988),which would suggest that growth hormone should result in an increased plasma cortisolconcentration. Treatment of fish and mammals with growth hormone, however, does notalter plasma cortisol concentrations (Langhorne 1986; Sillence and Etherton 1989;Enright et al. 1989; this study). A notable exception is that mammalian GH injectionsincreased plasma cortisol levels in the hypophysectomized teleost Poecilia latipinna (Balland Hawkins 1976). The increase in plasma cortisol levels following bPL treatment in thepresent study, therefore, indicates that the mode of action of bPL may be quite differentfrom that of bGH in coho salmon.Treatment with bGH or bPL had a pronounced effect on CR concentration in thegills. Only one experimental treatment had a significant effect on kr), the single injectionof the lowest dose of bPL (Figure 4.3b) showing that the primary change in CR was anincrease in concentration following growth hormone treatment. After a single treatment,5.0 bPL, 0.5 bPL and 5.0 bGH increased receptor concentration 19%, 34% and 26%,respectively (Figure 4.3a). After six weekly treatments, 5.0 bPL, 0.5 bPL and 5.0 bGHincreased CR concentration 60%, 38% and 41%, respectively (Figure 4.4a). The 0.05 bPLgroup showed a CR concentration not significantly different from the controls, despite asignificantly greater concentration of cortisol in the plasma. The elevated cortisolconcentration in the plasma would result in binding of cortisol to the intracellular receptorsand a reduction in the unbound receptor population. In Chapter 3, I showed a relationshipbetween the concentration of cortisol in the plasma and CR in the gills. With thisrelationship it is possible to calculate the concentration of CR in the gills if plasma cortisolconcentrations were similar to the control fish. Based on this relationship, CRconcentration in the gills of the fish with bPL would be approximately 111 fmol-mg protein"following a single treatment, and 73  protein' following six weekly treatments.These values are considerably higher than the control values for each treatment.Therefore, it seems likely that the lowest dose of bPL also produced a significant increase73in CR concentration in the gills. An increase in CR concentration in the gill can cause agreater sensitivity of the gill to cortisol. Danielsen and Stallcup (1984) showed that cortisolsensitivity of a tissue is dependent on the number of receptors for cortisol present in thetarget cells. The upregulation of CR by GH may be a mechanism for the synergisticactivity of Gil and cortisol in increasing Na1C+ATPase activity and chloride cell numbers.There is an appreciable difference between the CR concentration in the controlgroups following a single injection compared to the six weekly injections. During the falland winter there is little seasonal change in CR concentration (Chapter 2). The secondexperimental series was sampled during this time period (December 18 1992) approximatelytwo months after the first series was sampled (October 16 1992). Thus seasonal changesare unlikely to account for the differences in CR between the two sampling dates. Thelower number of CR in the gills in the second group were probably caused by repeatedhandling of the fish for hormone injections and growth measurements. Repeated handlingstresses cause a downregulation of CR in the gills (Chapter 3).A single injection of bPL or bGH had little effect on Na+KfATPase activity as onlyone group showed a significant increase. Six weekly injections resulted in a significantincrease in Na+K+ATPase activity for all the groups treated, similar to that shown in coho(Richman and Zaugg 1987), rainbow trout (Madsen 1990b) and sea trout (Salmo trutta,Madsen 1990c). With the experimental protocol used in our study, it is impossible toattribute the rise in enzyme activity solely to growth hormone. The weighing and injectionprocedure required handling and anaesthetizing the fish, both procedures that elicit a stressresponse in the fish (Barton and Iwama 1991). The stress associated rise in cortisol inassociation with the increased concentration of CR in the gills due to growth hormonetreatment, could account for the increase in Na+K+ATPase activity following the six weektreatment. The results with the 0.05 bPL group, however, weakens this hypothesis. Thisgroup showed the highest level of plasma cortisol and by calculation possessed a large poolof corticosteroid receptors, yet showed the smallest increase in Na+1C+A1Pase activity ofthe experimental groups. This finding lends support to the hypothesis of Richman andZaugg (1987) that cortisol enhances all gill ATPases, but growth hormone specificallyincreases Na+IC+ATPase activity in the gills. The 0.05 bPL group, also, showed the largest74rise in plasma Na+ following seawater exposure. Thus, there would appear to be goodevidence for growth hormone directly increasing Na+K+ATPase activity in the gills. I donot know, however, how plasma cortisol levels may have changed during the one weekfollowing each injection.The hypo-osmoregulatory ability, as assessed by the 24-hr saltwater challengediffered between the groups (Figure 4.7) despite the fact that bPL and bGH treatmentsappear to have a similar effect on growth, stimulation of Na+K+ATPase activity andincreasing the concentration of CR in the gills. There are differences between theconcentration of plasma sodium after saltwater transfer in bPL and bGH groups but adirect relationship between plasma sodium concentration and Na+K+ATPase activity doesnot exist. That is the significantly greater Na+K+ATPase activity of the bPL 5.0 group(Figure 4.5) does not result in a significantly lower plasma sodium concentration in saltwater (Figure 4.7) when compared to the bGH treated groups. These differences may beaccounted for by specific actions of bPL and bGH. Growth hormone and placentallactogen are part of a family of hormones that includes prolactin, and are related byfunction, immunochemistry and structure (Slater et al. 1986). Although both bPL andbGH caused improved Na+ excretory capabilities, the similarity of bPL to prolactin mayaccount for the differences observed in plasma sodium concentration between bGH andbPL treated fish. That is, bPL may bind to both GH and prolactin receptors in the fish,similar to a model presented for human GH (Nicoll et al. 1986). The excretion of Na+ andCl across the gills is dependent on the active transport of C1 transcellularly and the passivemovement of Na+ paracellularly across the gill epithelium (Epstein et al. 1980). Prolactinprevents the passive loss of Na+ from the gills (Hirano 1980), and decreases chloridesecretion (Foskett et al. 1982). If bPL binds to the prolactin receptor and restricts theefflux of Na+ and cr across the gills, saltwater tolerance will be reduced. Clarke et al.(1973), however, found that human placental lactogen (hPL) did not function in the sameway as prolactin in tilapia. The dose of hPL used by Clarke et al. (1973) was much greaterthan the highest dose of bPL used in the present study. The relatively high saltwatertolerance of the bPL 5.0 group is likely a function of the significantly greaterNa+K±ATPase activity in the gills.75Unlike bPL, at the concentrations of bGH used in the present study, there is a doseresponse for growth and CR, but there was no dose response seen for the increasedNa+K+ATPase activity and plasma sodium values of the saltwater challenge fish. Thisfinding is in agreement with that found for the ovine growth hormone by Clarke et al.(1977), where the levels of somatotropin required to stimulate growth appeared tomaximally enhance saltwater tolerance. The affinity of GH receptors for GH can differbetween tissues. Consequently, the difference between bPL and bGH could be a functionof tissue specific differences in receptor affinity for the two hormones. It would appear,however, that the efficacy of bGH to enhance saltwater tolerance is much greater than forstimulating growth in juvenile coho.There exists a strong synergism between cortisol and growth hormone in enhancingNa+K+ATPase activity (Madsen 1990b,c). The effect of combined GH and cortisoltreatment, however, differ dependent on the life stage of the fish. The enhancement ofNa+K+ATPase activity is much greater in parr (Madsen 1990c) than in presmolts (Madsenand Korsgaard 1991) using the same treatment protocol. In vitro, McCormick et al.(1989b) found that Gil had no effect on Na+K+ATPase activity alone or in combinationwith cortisol. These differences in responsiveness of the tissues to cortisol and GHtreatment may be associated with receptor concentration changes, as receptors have beenshown to change with different physiological states of animals (Nicoll et al. 1986; Chapter5). The results shown in the present study may apply only to juvenile coho parr. Theseasonal changes in physiology associated with smolting may alter upregulation ofcorticosteroid receptors by GH.The bovine hormones, growth hormone and placental lactogen, have a significanteffect on cortisol dynamics and increase hypo-osmoregulatory ability in juvenile cohosalmon. The upregulation of CR by GH also gives some insight into the• how GH andcortisol interact to increase hypo-osmoregulatory ability. During the spring, smoltingjuvenile coho show a reduction in the concentration and affinity of CR in the gills (Chapter5). These changes occur at a time when plasma cortisol levels surge (Specker and Schreck1982). It has been reported that the rise in cortisol occurs synchronously with a surge ingrowth hormone levels (Young et al. 1989a). As cortisol causes a downregulation of CR76in the gills (Chapter 3), one of the possible roles of GH during the spring may be to limitthe decline in CR that occurs during the spring as the fish smolt (Chapter 5).CHAPTER 5Changes in Cortisol Dynamics in Wild and HatcheryReared Juvenile Coho Salmon (Oncorhynchus kisutch)During Smoltification7778INTRODUCTIONThe development of saltwater tolerance in Pacific salmonid smolts is affected byrearing environment. Increased rearing density during the freshwater residence of hatcherycoho has also been shown to be inversely related to adult returns (Fagerlund et al. 1983).Other factors associated with rearing environment in the hatchery have also been shownto affect adult survival. Numbers of returning adult differs between sections of an earthenrearing channel. Fish from the top section of an earthen channel show a greater smolt toadult survival than those from the bottom section of the same channel (W.E. McLean,Quinsam River Salmonid Hatchery, personal communication). Fish reared in high densityand reduced water flow show impaired smolting; that is characterized by reduced saltwatertolerance (Schreck et al. 1985, Patino et al. 1986). Strange et al. (1978) showed thatincreases in rearing density resulted in significantly reduced Na+K+ATPase activities.There is a direct relationship between Na+VATPase activity and smolt to adult survivalrate (Wahle and Zaugg 1982; Zaugg 1989). The rearing environment during early lifehistory of salmonids can affect smolt development. Consequently, the numbers of returningadult salmon is partially dependent on the rearing environment of juveniles. There is alsoevidence that the smolt to adult survival is higher for wild than hatchery-reared juvenilecoho (J. Van Tine, Quinsam River Salmonid Hatchery, personal communication). Thesaltwater tolerance of coho salmon is greater for fish reared in the wild than those rearedin a hatchery (Chapter 7). Differences found between wild and hatchery smolts wereattributed to differences in the rearing environment.During the spring juvenile salmon change from animals adapted to the freshwaterenvironment, to ones capable of osmoregulating in the marine environment. The processof smoltification involves numerous physiological and structural modifications that arestimulated by hormonal changes within the animal (Hoar 1988). Cortisol is one of thehormones involved in smolting, and has been shown to increase saltwater tolerance whenadministered to juvenile salmonids by causing an increase in chloride cell proliferation andNa+K+ATPase activity (Madsen 1990a; Richman and Zaugg 1987). The hatchery rearing79practices that have been shown to impair development of hypo-osmoregulatory ability arestressors and elevate plasma cortisol levels (Barton and Iwama 1991). Due to the increasein plasma cortisol concentration, stresses associated with rearing practices would beexpected to increase the saltwater tolerance of hatchery-reared smolts. This, however, isnot the case.Stressors associated with the rearing environment affect cortisol dynamics in juvenilesalmonids. For example, stress has been shown to cause an increase in plasma cortisolconcentration (Barton and Iwama 1991), as well as to decrease the half-life of cortisol inthe plasma (Redding et al. 1984), and decrease the number of corticosteroid receptors inthe gills (Chapter 3). Factors that change the cortisol dynamics may alter the sensitivityof the gills to cortisol and have a direct effect on smolting.As there is much evidence that rearing environment will affect smolting and cortisoldynamics, I examined the difference in cortisol dynamics in both hatchery and wild cohosalmon (Oncorhynchus kisutch) during the spring when the fish were smolting. Thecomparison between wild and hatchery-reared coho was included as a large difference inthe saltwater tolerance of juvenile coho salmon was found between these two groups in acompanion study (Chapter 7). Due to the differences in adult survival associated withlocation in a rearing channel, two groups of hatchery coho from the top and bottomsections of an earthen rearing channel were examined. As the water flows from the top tothe bottom sections, dissolved oxygen concentration declines, and ammonia concentrationand suspended particulate matter increase (K. Campbell, Quinsam River SalmonidHatchery, personal communication). These changes have all been documented to have anegative impact on growth and development of juvenile salmonids (Wedemeyer et al.1980).80MATERIALS AND METHODSStudy site. The Quinsam River is located northwest of the town of Campbell River,on Vancouver Island. The river is approximately 40 km long, and drains an extensivenetwork of tributaries and lakes, but only the lower 29 km of the river are accessible tocoho salmon. The Quinsam River is the main tributary of the Campbell River and theirconfluence is 3.5 km above the Campbell River estuary (Blackmun et al. 1985). In 1974,a Salmonid Enhancement Program Facility was constructed on the Quinsam River. Thehatchery is located approximately one km above the confluence with the Campbell River.The hatchery production of coho smolts is greater than 120 000 each year. Production ofwild smolts in the Quinsam River is estimated to be greater than 35 000 per year. Inaddition, a colonization program of hatchery fry transplanted to the upper Quinsam Riverwatershed produces greater than 20 000 smolts per year (see Chapter 7 for furtherdescription of this group). Consequently, the majority of coho smolts produced in theQuinsam River system are of hatchery origin (D. Ewart, Quinsam River SalmonidHatchery, personal communication).Fish. Three groups of juvenile coho salmon from the Quinsam River were examinedover the spring of 1991 and 1992. The groups consisted of wild coho, and two hatcherygroups. The hatchery groups were taken from the top and bottom section of an earthenrearing channel, and designated hatchery 1 and hatchery 2, respectively. The earthenchannel was supplied with ground water. The temperature of the water supplied to theearthen channel exhibited little seasonal variation. The mean hatchery water temperaturewas approximately 10 °C with a seasonal fluctuation of less than 2 °C. In contrast, the riverwater temperature showed a large seasonal variation from a low of less than 2 °C in thewinter, to a maximum of greater than 15 °C at the time of smolt migration in the spring.In addition to the differences in water temperature between the hatchery and wildfish, suspended particulate matter as measured by non-filterable residue (NFR) differed.The NFR in the river was primarily inorganic in nature (W.E. McLean, Quinsam RiverSalmonid Hatchery, personal communication). Particulate matter within the hatchery is81composed of food and faecal material, and contains a higher percentage of organic materialthan the inert particles of river silt. This high organic component results in ammoniaproduction (K. Campbell, Quinsam River Salmonid Hatchery, personal communication).The water entering the top of the earthen channel was first passed through rearingcontainers on the Quinsam hatchery upper site and then passed through a clarifier toremove gross particulate matter. The outflow from the clarifier fed into the top section ofthe earthen channel. NFR values at the inflow to the earthen channel showed littlevariation over the spring and were approximately 0.2 mg.L-1. NFR was higher at theoutflow from the bottom of the earthen channel. In early spring (March), the NFR wasapproximately 0.5 mg-L-1. By May, the NFR increased in both years and was greater than1 mg.L-1 in early May and typical almost 2 mg.L-1 at the end of May when the smolts werereleased (K. Campbell, Quinsam River Salmon Hatchery, personal communication).Dissolved oxygen levels were typically lower in the hatchery than the river. Riverdissolved oxygen was typically near saturation at all times of the year (> 10 mg.L-1). Waterat the top of the earthen channel was supplemented with oxygen to raise dissolved oxygento saturation levels. At the bottom of the earthen channel, dissolved oxygen levels wereless and often below 8 mg.L-1 (D. Ewart, Quinsam River Salmonid Hatchery, personalcommunication).In 1991, the wild fish were caught in baited minnow traps. The traps were left inthe river for up to 6 hr, at which time the traps were rapidly removed from the water andthe fish transferred to a bucket containing 200 mg.L1 tricaine methanesulphanate (MS-222),buffered with 400 mg.L-1 sodium bicarbonate. This dosage of MS-222 has been shown tobe sufficient to prevent the release of cortisol in fish following capture (Barton et al. 1986).The residence time in the minnow traps proved to be stressful for the fish, therefore, thewild fish were captured by electrofishing in 1992. A sufficient voltage was used whileelectrofishing to stun the animals immediately, and by holding the fish in the electrical fieldit was possible to kill the animals very quickly. To ensure that the animals did not revivefollowing electro-anaesthesia, the fish were placed into a bucket containing a lethal doseof MS-222 prior to sampling (procedure described below). In 1991 and 1992, when the wildsmolts began to migrate, they were caught at the downstream enumeration fence on the82Quinsam River. Dip nets were held over the trap entrance to capture fish prior to entryinto the enumeration trap. Each minute the net was raised and examined for fish. If a fishhad been caught, it was immediately transferred to a bucket containing a lethal dose of MS-222. After sacrificing the fish, the weight and length were measured and the caudalpeduncle was severed and blood samples were collected in heparinized capillary tubes. Thetubes were kept on ice until they could be centrifuged, and the plasma separated andcollected. Plasma samples were frozen at -80 °C until determination of cortisol and sodiumconcentrations. Gills were removed and treated as described below.The hatchery fish were captured by seining the earthen channel. The seine wasdrawn toward the shore and fish collected from the seine were immediately transferred toa bucket containing MS-222 (1991) or a bucket with water and then sacrificed by electro-anaesthesia (1992). In both years the seining procedure required approximately five minto capture the fish. Pickering and Pottinger (1983) showed that cortisol did not risesignificantly within three minutes after capture. As the time for capture used in the presentstudy was similar to that of Pickering and Pottinger (1983), it is likely that the cortisollevels measured represent resting levels for the fish.Tissue preparation. Corticosteroid Receptors. Gills were excised from the fish andblotted on tissues to remove excess blood and then placed in a solution of 10 mM Tris-Ha,1 mM EDTA, 12 mM monothioglycerol, 20 mM sodium molybdate and 10 % (v/v) glycerolall at pH 7.4 (TEMS). Approximately 0.5 g of gill tissue was added to 2.5 mL of TEMSand stored in liquid nitrogen until samples were assayed for corticosteroid receptors (CR).Tissues from up to 8 individuals were pooled for one sample. Four samples were assayedfor CR for the hatchery groups per sampling interval. Capture of a large number of wildfish was often very difficult. Consequently, enough tissue for four CR samples could notalways be obtained and fewer than four wild samples were assayed on several occasions.The method of Maule and Schreck (1990) as modified in Chapter 2 was used to assay forthe receptors. The concentration (B...) and dissociation constant (kJ)) were calculatedaccording to Scatchard (1949).In 1991, the use of minnow traps to catch the wild fish resulted in a significant stressresponse. Due to the high circulating plasma cortisol concentrations, it is likely that a83proportion of the receptor population in the gills of the wild fish was bound to theendogenous cortisol and removed from detection by the radioreceptor assay used in thisstudy. In Chapter 3, I showed a relationship for plasma cortisol and gill CR concentrationand affinity Using this relationship, I calculated the total concentration of gill CR that thefish would have possessed for a plasma cortisol concentration of 5 ng.m1-1 for the first threesampling intervals for the wild fish in 1991. I assumed that a cortisol concentration of 5ng.m1-1 was the resting value for wild fish at that time of the year as this level is similar tothe values measured in 1992 for the wild fish.Na+KPATPase Activity. Gill arches from one side of the animal (approximately 50mg) were removed and blotted dry on paper tissues. Gill tissue was placed in 0.5 ml of icecold SEI (0.3 M sucrose, 0.02 M EDTA and 0.1 M imidazole). Samples were frozen andstored at -80 °C until analysis for enzyme activity following the method of Zaugg (1982).Ten samples were assayed for each sampling time, except for the wild fish where samplesize was limited to the number of fish that could be caught. Enzyme activities wereexpressed per mg protein in the gill homogenate. Protein was assayed using the Biuretmethod described in Alexander and Ingram (1980).Plasma cortisol half-life. Half-life of cortisol and cortisol metabolites in the plasmaof juvenile wild and hatchery coho was measured using a method similar to that of Reddinget al. (1984b). Wild juvenile coho were caught by minnow trap or at the downstreamenumeration fence, and transported to an aluminum trough at the Quinsam RiverHatchery. The hatchery groups were seined from their respective sections of the earthenchannel and also transported to an aluminum trough. The fish were left in the trough for24 hours before estimation of cortisol half-life. The change in rearing environment for thefish, and the stress associated with capture should not have affected the half-life of cortisolin the plasma as Redding et al. (1984b) showed that acute stress did not affect theclearance of cortisol from the plasma.To determine cortisol half-life in the plasma, fish were removed from the aluminumtrough and anaesthetized using buffered MS-222 (50 mg.1-1). Up to 32 fish from each groupwere injected with 2 tiCi.g4 of 41-cortisol (Dupont-NEN) intracardially. After 1.5, 3, 4.5,6 hrs, 5 - 8 fish were killed with a lethal dose of MS-222, the caudal peduncle severed and84duplicate blood samples were collected in heparinized capillary tubes. The tubes werecentrifuged at 11 500 g for five min and 20 ILI of plasma was placed into a 20 mlscintillation vial. Five ml of aqueous scintillation cocktail (ACS; Amersham) was added toeach vial and the radioactive decay of the samples was counted for five min in a LKBRackbeta scintillation counter. Plasma cortisol half-life was calculated according toNormand and Fortier (1970). Fish that had not been successfully injected intracardiallyhad significantly lower decay values and were excluded from calculation of plasma cortisolhalf-life.Analysis. Plasma sodium concentration was measured using a Perkin-Elmer AtomicAbsorption Spectrophotometer, model 2380. Twelve samples were analyzed from eachgroup. Plasma cortisol was measured using a radioimmunoassay kit using 'I (Incstar,Corp.). Ten samples were analyzed from each group.Statistical analysis. Condition factor (CF) was calculated as (weight*100).1ength-3.Fish weight, condition factor, plasma cortisol concentration, CR concentration and affinity,Na+K+ATPase activity, and plasma sodium concentration for the hatchery and wild fishwere compared for significant differences using Analysis of Variance, followed by a Tukey'stest to find significant differences between means. The half-life of cortisol in the plasmawas compared for significant difference between the groups and sampling intervals byAnalysis of Covariance. Statistical significance was taken at a level of a = 0.05. All valuesare expressed as means ± standard error.85RESULTSWeight and condition factor. The fish sampled from the hatchery were significantlylarger than the wild fish at all sampling intervals during the course of this study. Therelative sizes of the fish are shown in Figure 5.1.The condition factor (CF) was not significantly different between the wild, hatchery1 or hatchery 2 groups of fish examined (Figure 5.2). Although, there was a significantreduction in CF over the spring, the pattern of decline was not the same for wild andhatchery fish. Both groups of hatchery fish showed a gradual reduction in CF over thespring. The wild fish, in contrast showed a marked, but not significant increase in CFbetween March and April in 1991 and 1992. This increase in CF in the wild was followedby a significant reduction in CF between April and May, for both years of this study.Coloration. The temporal colour change differed between the wild and hatchery-reared fish. For the spring of 1991 measurements, parr marks were clearly visible on allfish examined in March. On the April 12 sampling date 24% of the fish from the hatchery2 group were silver and appeared to be smolted. On the April 26 sampling date all thehatchery fish examined were silver. It was not until the May 10 sampling period that anyof the wild fish showed the colouration typical of smolts. A similar temporal change incolouration was seen for the spring of 1992. Some of the hatchery-reared fish had begunto turn silver by March 12; 25% of hatchery 1 and 15% of hatchery 2. All of the hatcheryfish sampled on April 17 were silver and appeared to have smolted. The wild fish sampledin April all exhibited pronounced parr marks, and it was not until May that the wild fishshowed any signs of silvering.Plasma sodium concentration. The plasma sodium concentration showed a gradualdecline over the spring (Figure 5.3). This trend was consistent in all groups. Analysis ofVariance indicated that the decline was significant, but a Tukey's test showed that therewas no significant difference between the groups.Plasma cortisol concentration. In 1991, plasma cortisol concentrations in the hatcherygroups showed a gradual increase to a maximum at the end of April (hatchery 1) and the86beginning of May (hatchery 2); there was a decline in plasma cortisol concentration afterthe maximum (Figure 5.4). The fish sampled from the hatchery 2 group showed asignificant increase in plasma cortisol concentration, whereas the fish sampled from thehatchery 1 group did not show a significant increase in plasma cortisol over earlier springmeasurements. The cortisol concentrations in the wild fish were always much higher thanthe hatchery fish. For the first three sampling periods, the wild fish were caught in minnowtraps, which resulted in a considerable stress associated rise in cortisol. The plasma cortisolconcentrations were 149 ± 29, 130 ± 25 and 62 ± 16 for the March 16, 29 and April 12,1991 sampling dates, respectively. During the last three sampling periods, the wild fishwere captured at the downstream enumeration fence. The fish were sacrificed immediatelyafter capture, preventing any stress associated rise in cortisol. During this time the cortisolconcentration in the wild fish was significantly greater than hatchery fish. In 1992, a similartrend was observed as in 1991. The increase in plasma cortisol was significantly greater inlate April for both groups of hatchery smolts. The peak in plasma cortisol occurred in Mayfor the wild fish. As the peaks in plasma cortisol concentration were out of phase in thehatchery and wild populations of smolts, the wild fish had significantly lower cortisol levelsin April than the hatchery fish. This pattern was reversed in May. The maximum cortisolconcentration achieved during the spring was significantly greater in the wild smolts (May)than in both the hatchery groups (April).Plasma cortisol half-life. During the spring, the half-life of cortisol and cortisolmetabolites in the plasma showed a decline in all groups of fish examined (Figure 5.5). ByMay, the half-life of cortisol was significantly less than the March values. For eachsampling date, however, there was no significant difference in plasma cortisol half-lifebetween the groups.Corticosteroid receptors, numbers and affinity. The concentration of corticosteroidreceptors in the gills declined over the spring as the fish smolted (Figure 5.6). The wildfish consistently had the greatest concentration of corticosteroid receptors in the gills. Thehatchery 1 fish had slightly higher concentrations than hatchery 2 fish. The dissociationconstant (1(D) increased over the spring (Figure 5.7). In 1991, there was a gradual increasein kr, in both wild and hatchery reared fish. In 1992, the hatchery fish showed a similar87gradual increase in IcD, however, the wild fish did not show an increase in k D from Februaryto April and exhibited a large increase in kr, from April to May. The temporal change inkr, meant that the affinity of CR for cortisol in the wild fish was significantly different fromthe two hatchery groups. The Hill coefficients did not differ significantly from 1.0 for allthe samples analyzed.NeICATPase Activity. The specific activity of Na+K+ATPase in the gills showed anincrease during the spring in the juvenile coho salmon examined (Figure 5.8). The wild fishshowed a significant increase in activity during May. Although, the enzyme activity wasgreater in the gills of hatchery fish in May than the preceding months, the levels did notdiffer significantly. Consequently, during May the Na+K+AT'Pase activity was significantlygreater in the wild fish compared to the hatchery fish. The differences seen during thespring of 1991 were also seen in the spring of 1992.The activity of Na+1C+ATPase was plotted against the circulating plasma cortisolconcentrations measured during each sampling date during the spring of 1991 and 1992 forall the data except that from the wild fish caught in the minnow traps. Separaterelationships were found for each group examined during the study (Figure 5.9). The wildfish exhibited the greatest increase in NalleAT'Pase activity for a given increase in plasmacortisol concentration. The hatchery fish did not show an increase in Na+K+ATPaseactivity with an increase in plasma cortisol concentration. The slopes of the lines asdetermined by linear regression were 0.101, 0.018, 0.001 Na+WATPase activity / cortisolconcentration for wild, hatchery 1 and hatchery 2 groups, respectively. The regression linesfor the two hatchery groups did not differ significantly from zero, whereas, the wild fish didshow a significant (a =0.10) relationship between Na+K+ATPase activity and plasma cortisolconcentration.30252015F--^1 0= v)0 E 5i] EL^,3025201510588February^March^April^May^JuneFigure 5.1. The weight of the juvenile coho salmon caught during 1991 and 1992. Wildfish were caught in the Quinsam River, and hatchery 1 and 2 fish were seined from the topand bottom sections of an earthen rearing channel, respectively. Hatchery 1 and 2 fishwere significantly larger than wild fish at all sampling intervals. n = 25 unless indicatedadjacent to the data points.1 .00.90.8February March April May June1 .2Figure 5.2. The change in condition factor over the spring of 1991 and 1992 for wild andhatchery reared coho salmon. * indicate data that are significantly different (a = 0.05)from wild values at the same sampling interval. Groups description and sample size aregiven in Figure 5.1.89February March April May June15014515014590Figure 5.3. Plasma sodium concentrations during the spring for hatchery and wild cohosalmon. n = 12 for each data point. Symbols and groups are as described in Figure 5.1.February March April May June1251 0 07550250Figure 5.4. The change in concentration of cortisol in the plasma of juvenile coho salmonfrom the Quinsam River over the spring. n = 10 for each data point. Symbols and groupsare as described in Figures 5.1 and 5.2.91March^April^May^June92Figure 5.5. Half-life of cortisol and cortisol metabolites in the plasma of juvenile cohosalmon from the Quinsam River during the spring. Groups are as indicated in Figure 5.1.93120(f)^100C0F—^80a_LJ 6U T 0LJI c.-i2 o-I--^400L_D0^ ^200 CT)ELLI^120F— 0cri E1000CD80E—CK0^60U40February^March^April^May^JuneFigure 5.6. Concentration of corticosteroid receptors in the gills of coho salmon during thespring of 1991 and 1992, from the Quinsam River. The first three data points for the wildfish in 1991 were calculated to correct for the high circulating cortisol levels in the plasma.n = 4 unless otherwise indicate adjacent to the data points. Symbols and groups are asdescribed in Figures 5.1 and 5.2.94 1.5Z:<F—(J)^1 .01.0CD^ ^0.5Z0 EF-1-< 1.0-.60V)^0.800^0.60.4February^March^April^May^JuneFigure 5.7. Dissociation constant of the corticosteroid receptors in the gills of coho salmonduring the spring. n numbers are as indicated in Figure 5.6. Symbols and groups are asdescribed in Figures 5.1 and 5.2.March^April May June121 125a:20-3E-s.^15105FebruaryFigure 5.8. Changes over the spring in the activity of Na+1VATPase activity in the gills ofjuvenile hatchery and wild coho from the Quinsam River. n = 10 unless indicated adjacentto the data points. Symbols and groups are as described in Figures 5.1 and 5.2.95• WILDA HATCHERY 1V HATCHERY 29630_c 25CD 200H-15+„ E100^20^40^60^80^100^120PLASMA CORTISOLng • ml'Figure 5.9. The relationship between gill Na+K+ATPase activity and cortisol concentrationin the plasma of juvenile hatchery and wild coho salmon. Linear regression lines are alsoshown on the plot. R2 for the regression lines were 0.4438, 0.0003 and 0.0003 for the wild,hatchery 1 and hatchery 2 groups, respectively.97DISCUSSIONDuring the spring, both hatchery and wild juvenile coho showed many morphologicaland biochemical changes that indicated the fish were smolting. There was a gradualreduction in the condition factor (CF), indicating that the fish were becoming leaner(Figure 5.2). The change in CF has long been associated with the parr-smolttransformation (McCormick and Saunders 1987). The temporal change in CF, however,is not the same for both the hatchery and wild smolts. The hatchery fish showed a gradualreduction in CF throughout the spring. In contrast, the pattern was more variable in thewild fish, but showed a dramatic drop between April and May. The difference in smoltdevelopment between the groups of fish was also seen in the timing of colour change. Thehatchery groups changed colour and appeared to be smolted at an earlier date than thewild fish. These morphological observations suggest that the temporal physiologicalchanges associated with smolting are not the same for hatchery and wild coho salmon.The concentration of plasma sodium in fresh water showed a reduction over thespring, another parameter that has been found to represent a physiological changeassociated with smolting (McCormick and Saunders 1987). There was no difference,however, between the wild and hatchery coho plasma sodium concentrations for anysampling interval.Numerous changes in cortisol dynamics occur as the fish smolt during the spring.There is an increase in plasma cortisol concentrations (Figure 5.4), a decrease in the half-life of cortisol in the plasma (Figure 5.5), and a decline in the number (Figure 5.6) andaffinity (Figure 5.7) of the receptors for cortisol in the gills. Although the pattern ofchanges in cortisol dynamics are similar for hatchery and wild fish, the magnitude of thechanges is often greater in the wild fish when compared to the hatchery fish. Thesuppressed changes in cortisol dynamics in the hatchery-reared fish are associated with nosignificant rise in gill Na+K+ATPase activity. As Na+K+ATPase activity in the gills hasbeen correlated with the development of saltwater tolerance in smolts (Zaugg 1989;Chapter 7), the hatchery-reared coho smolts would not be expected to adapt to the98increased salinity of the marine environment as well as the wild smolts. In a companionstudy examining the saltwater tolerance of hatchery and wild coho smolts from the samestock, the hatchery smolts showed a weaker saltwater tolerance than their wild counterparts(Chapter 7). The reasons for the reduced hypo-osmoregulatory ability in the hatchery fishmay be a result of genetic differences between the population or environmentally inducedmodifications in development. There is not likely to be a significant difference in thehatchery and wild populations genetically. Field observations conducted by the hatcherypersonnel indicate that adults of hatchery origin migrate along with colonization and wildadults into the Quinsam River above the enumeration fence (R. Kraft, Quinsam RiverSahnonid Hatchery, personal communication). The wild smolts produced in the QuinsamRiver could be the progeny of hatchery or wild adults. Consequently, the differences seenbetween the juvenile wild and hatchery coho in the present study is attributed to rearingenvironment. Rearing salmonids in intensive fish culture facilities has resulted in the fishbeing exposed to a number of stresses which they do not experience at all in the naturalenvironment or do not experience to the same degree (Donaldson 1981). Stress has alarge effect on plasma levels of cortisol (Barton and Iwama 1991), half-life of cortisol inthe plasma (Redding et al. 1984b) and the concentration of corticosteroid receptors in thegills (Maule and Schreck 1991; Chapter 3).The seasonal increase in cortisol during the spring is a characteristic physiologicalchange associated with smolting (Specker and Schreck 1982, Barton et al. 1985, Hoar1988). Based on differences between hatchery and wild fish, rearing environment has apronounced affect on the rise in cortisol over the spring as the fish smolt. The surge incortisol during the spring is significantly greater in the wild than the hatchery fish. Thesampling times, however, may have exaggerated this difference, as the wild fish weresampled during the night and the hatchery fish were sampled during the day. Part of thedifferences in cortisol concentration could be attributed to an increase in plasma cortisolduring the night (Thorpe et al. 1987, Pickering and Pottinger 1983, Rance et al. 1982).These researchers have shown a two fold increase at night, peaking at 2400 hr. Barton etal. (1986), however, did not show any diurnal change in plasma cortisol concentration. Thediscrepancy between these findings could be attributed to feeding regimes. Feeding has an99important role on the phasing of the corticosteroid rhythm (Bry 1982). Fish given a singlemeal near the start of the light phase show a corticosteroid peak during the hours ofdarkness, whereas feeding at other times or feeding randomly throughout the day eithermodified the phasing of the rhythm or suppressed it altogether (Delahunty et al. 1978).As the hatchery fish are fed over the course of the day, the cortisol levels in the plasmamay not fluctuate greatly. The measured levels of cortisol in the hatchery and wild smoltsfound in this study are likely representative of the resting cortisol levels for the fish.The surge in cortisol during the spring is a result of enhanced interrenal activityduring smoltification (Nishioka et al. 1982). The response of the interrenal to stress isalso dependent on rearing history (Barton et al. 1987). Salonius (1991) showed that fishreared naturally in the river had significantly greater plasma cortisol concentrations thantheir hatchery counterparts when subjected to a 60-s handling stress. The mechanism forthe difference in cortisol production from the interrenal during stress may be due to areduced sensitivity of the interrenal to ACTH. Regulation of cortisol secretion is mediatedby negative feedback to the hypothalamus and pituitary (Fryer and Peter 1977) and cortisolsecretion from the interrenal cells inhibits further interrenal response (Donaldson 1981).Plasma cortisol levels reveal only partial information on hormone dynamics. Levelsof cortisol are similar in saltwater and freshwater eels (Henderson et al. 1974) and cohosalmon (Redding et al. 1984b). Saltwater fish, however, have higher metabolic clearanceand production rates than freshwater animals. The decline in half-life of cortisol in theplasma indicates an increase in cortisol utilization.Clearance of cortisol from the plasma was estimated by the reduction in total-3Hafter injection of 3H-cortisol and would therefore include clearance of cortisol metabolites.Brown et al. (1986) emphasize that clearance rates calculated using total radioactivity canunderestimate the clearance rate of authentic cortisol by a factor of 2 to 4, as cortisol israpidly converted to cortisone in the plasma of the fish (Patino et al. 1985). Still, the half-life of radioactivity in the plasma can be used as a reasonable approximation of theclearance rate of cortisol from the plasma (Redding et al. (1984b). It is possible that therate of conversion of cortisol to cortisone may vary between the hatchery and wild fish. To100what extent this would affect my results is not known.Stress, transfer to sea water (Redding et al. 1984b) and smolting (Patino et al. 1985)result in a decrease in the half-life of cortisol in the plasma of coho salmon. I also presentevidence that the half-life of cortisol in the plasma declines as the fish smolt. The half-lifeof cortisol at each sampling interval did not differ significantly between the groups. Asrearing environment will affect cortisol dynamics, and the environments for the hatcheryand wild fish are different, it is surprising that there is little difference between the groups,especially in that the other parameters measured differ significantly between the groups.Redding et al. (1984b) found that acute stress and acute treatment with cortisol did notsignificantly affect the clearance of cortisol from the plasma, although chronicadministration and stress did result in an increase in cortisol clearance. It is possible,therefore, that changes in the metabolism of cortisol are not as sensitive to stress as theother parameters measured.Hormone sensitivity of a tissue is dependent on the number of hormone receptorsin the target tissue (Danielsen and Stallcup 1985). As the wild juvenile coho alwayspossessed a greater number of corticosteroid receptors than the hatchery fish, the gills ofwild coho were more sensitive to increases in cortisol concentration. Cortisol in turn willmodify the concentration and affinity of the corticosteroid receptors in the gills of cohosalmon (Maule and Schreck 1991, Chapter 3). As the decline in receptor number andaffinity during the spring is similar to that seen due to cortisol treatment, the seasonalchange appears to be regulated by the cortisol surge associated with smolting. The changein CR, however, began to occur before the plasma cortisol concentrations began to increasein the spring. Factors other than cortisol may also affect CR concentration and affinity.Growth hormone (GH) treatment will upregulate the concentration of gill CR (Chapter 4).The possibility of GH affecting CR concentration differentially between the two groups isunlikely as the endogenous levels of Gil in the plasma were seldom significantly differentbetween the hatchery and wild coho during this study (Chapter 6). The absolute differencein CR concentration between the hatchery and wild coho, however, is likely due toregulation of CR by cortisol. In Chapter 3, it was shown that stress will cause adownregulation of CR in the gills of juvenile coho salmon. The lower concentration of CR101in the gills of hatchery fish may be a result of stresses associated with hatchery life,downregulating the gill CR. Residence in a hatchery exposes the fish to a rearingenvironment that is unlike that experienced by wild fish (Donaldson 1981). Rearingconditions are not uniform within a fishculture facility. Although, rearing density is similarbetween the top and bottom of the earthen rearing channel examined in this study,dissolved oxygen, NFR and ammonia levels differed between the top and bottom of thechannel (K. Campbell and D. Ewart, Quinsam River Sahnonid Hatchery, personalcommunication). Despite these differences there was no significant difference in cortisoldynamics, Na+1C+ATPase activity and saltwater tolerance (Chapter 7) between the hatchery1 and 2 fish. It is possible that the parameters within the rearing environment that differedmeasurably do not have a large effect on development of hypo-osmoregulatory ability. Itis also possible, however, that the water quality at the top of the earthen channel had anadverse effect on smolting. As the water entering the top of the channel had previouslysupplied ponds in the upper Quinsam hatchery site, water contained appreciable levels ofparticulate matter and ammonia. Increases in NFR and ammonia, the higher rearingdensities, physical disturbances and other rearing practices may account for the differencesin cortisol dynamics seen between the hatchery and wild coho.Due to the greater concentration of CR, the sensitivity of the gills to cortisol isgreater in the wild than the hatchery smolts. This corresponds with a much greaterresponsiveness to cortisol in the wild fish. This is shown by a significantly greater increasein Na+K+ATPase activity during the spring and a strong positive relationship betweenNa+VATPase activity and plasma cortisol concentration. This relationship is shown inFigure 5.9. The rise in Na+K+ATPase activity associated with cortisol increase is greatestin the wild fish and lowest in the hatchery fish from the bottom section of the earthenchannel. The relationship between Na+K+ATPase activity and cortisol concentration is inagreement with the findings for the corticosteroid receptor data.Wild fish show an increase in AT'Pase activity, plasma cortisol and decline in CRaffinity (increase in KO that all follow a similar pattern. As NeK+ATPase activity andcortisol have been used extensively to quantify hypo-osmoregulatory ability and kr) showsa similar seasonal change in pattern, then IED may be a good parameter to assess smolting.102The significantly higher kr, seen in the 1992 data for the wild fish and their excellentperformance in the 35 %o saltwater challenge test (Chapter 7) is evidence for kr, being auseful parameter for evaluating smolting in coho salmon.A correlation exists between cortisol dynamics (this Chapter) and saltwater tolerance(Chapter 7) in smolting coho salmon. The fish that exhibit the greatest surge in cortisolduring the spring and possessed the greatest number of cortisol receptors in the gills, alsoshowed the greatest increase in Na+K+ATPase activity and hypo-osmoregulatory abilitywhen transferred to salt water (Chapter 7). The role of cortisol in stimulation of smoltingappears to be suppressed or inhibited in the fish reared in the hatchery. The differencebetween the CR concentration in wild and hatchery 1 fish is similar to that between thehatchery 1 and hatchery 2 fish. The wild fish, however, show a significantly greatersaltwater tolerance than both the hatchery groups of fish, which did not differ significantly.Therefore, it is likely that the difference in CR concentration in conjunction with thegreater surge of cortisol in the wild smolts may account for the difference in saltwatertolerance of the hatchery and wild smolts.The morphological, biochemical and physiological changes that occur in the wildsmolting juvenile coho appear to occur synchronously. Between April and May, there isa marked decline in CF and change in colouration to the silvery appearance of a smolt.These morphological changes occur at the same time as the activity of Na+K+ATPaseactivity showed a significant increase and the circulating levels of cortisol in the plasma areat the highest level. The change in kr, of the gill CR between April and May also indicatesthat the wild fish show a rapid physiological change when they smolt. The physiologicalchanges that occur in the wild coho are closely timed with the onset of migration into themarine environment. The wild smolts are able to hypo-osmoregulate well in salt water, andlikely have the best chance for survival. In contrast, many of the morphological andphysiological events that occur in the hatchery fish are not synchronous. There tends tobe a more gradual change in physiology of the hatchery coho. The decrease in CF and theincrease in kr) occur more gradually over the spring, the change in plasma cortisol andcolouration are much earlier, and there is an insignificant increase in gill Na+IVATPaseactivity.103The synchronous changes that occur in the wild fish caught during this study couldbe partially attributed to sampling. For the May sampling period, the wild fish were caughtin the downstream enumeration fence during migration. Consequently, only migrants werecaught and apparent synchronous timing of physiological and morphological changes in thewild fish could be an artifact of sampling. The migration of wild smolts down the QuinsamRiver, however, is short in duration. Over 75% of the wild smolts migrate over a 12 dayperiod (Blackmun et al. 1985). The fact that most of the wild fish migrate within a narrowtime period and that there are few two year old smolts in the river (<0.1%, M. Trenholme,Quinsam River Salmonid Hatchery, personal communication) is evidence that the protocolemployed in the present study resulted in the capture of a representative sample of the wildsmolt population. When given access to the river, the hatchery smolts leave the rearingchannel within five days. Up to 90% of the fish leave within 48 hr. It is estimated that thefish reach the estuary within 48 hr after entering the river (R. Kraft, Quinsam RiverSalmonid Hatchery, personal communication). The hatchery smolts were sampled just priorto release to obtain a representative sample for the population. The differences observedbetween timing of smolting in the hatchery and wild coho in the present study is likely tobe an accurate representation of differences between the two populations.The hatchery fish show a reduced saltwater tolerance when compared to their wildcounterparts (Chapter 7, Figure 7.9). The higher number of CR in the gills and the greatersurge of cortisol during the spring enhance the development of Na+K+ATPase activity inthe gills of the wild coho smolts. It is likely that factors associated with rearingenvironment in the hatchery suppress physiological changes and disturb the synchrony ofevents required for successful smolting. These findings have important implications forhatchery rearing practices in the production of salmon smolts.CHAPTER 6Changes in Growth Hormone Concentration DuringSmolting and Following Saltwater Transfer in Wild andHatchery-Reared Juvenile Coho Salmon(Oncorhynchus kisutch)104105INTRODUCTIONDuring the spring when salmonids smolt, the concentration of growth hormone(GH) in the plasma has been shown to increase (Sweeting et al. 1985; Prunet et al. 1989;Young et al. 1989a). The surge in growth hormone is one of the endocrine responses thatis associated with the transformation of the freshwater parr to the migratory smolt (Hoar1988). Following transfer to salt water, a further increase in GH concentration has alsobeen reported (Sweeting et al. 1985; Sweeting and McKeown 1987; Sakamoto et al.1990, 1991). The changes in GH levels seen during smolting and after transfer to salt waterplay an important role in stimulating hypo-osmoregulatory ability in salmonids. There isconsiderable evidence that GH has a direct effect on stimulating an increase in saltwatertolerance. Exogenous administration of GH has been shown to increase hypo-osmoregulatory ability (Komourdjian et al 1976) and the perturbation in plasma sodiumconcentration following saltwater transfer (Bolton et al. 1987a; Collie et al. 1989;Chapter 4). The increased saltwater tolerance can be attributed to the increasedNa+K+ATPase activity (Richman and Zaugg 1987; Madsen 1990b; Chapter 4) and anincrease in chloride cell concentration (Madsen 1990b,c).The development of saltwater tolerance is, however, affected by rearing environment.Juvenile coho salmon reared in earthen channels in a hatchery have been found to showa much lower tolerance to salt water than their wild counterparts (Chapter 7). This groupof coho also showed marked differences in cortisol dynamics during the spring (Chapter5). The significant difference in cortisol dynamics may contribute to the reduced saltwatertolerance in the hatchery-reared coho. It is also possible that plasma GH levels may differbetween hatchery and wild coho. There is considerable evidence supporting an interactionbetween cortisol and GH (Pickering et al. 1991; Young 1988). Therefore, the differencesin cortisol dynamics found in this group of fish may have an impact on circulating Gillevels in the plasma and affect the development of saltwater tolerance in juvenile coho.Growth hormone concentrations in plasma samples from juvenile hatchery and wild cohosalmon (Oncorhynchus kisutch) were measured during the spring and after transfer to saltwater in May, in two consecutive years.106MATERIALS AND METHODSFish. The three groups of fish examined in the present study have been describedin detail in Chapter 5. Briefly, the groups from the Quinsam River consisted of wild cohoand two hatchery groups. The hatchery fish were sampled from the top and bottomsections of an earthen channel that was divided into three sections, and designated hatchery1 and hatchery 2, respectively. These three groups of fish were sampled over the springand following transfer to salt water for seven days of two consecutive years, 1991 and 1992.In 1992, during the months of February to April, the wild fish were caught byelectrofishing. Once the wild fish began to migrate, they were caught at the downstreamenumeration fence on the Quinsam River. After capture, the fish were rapidlyanaesthetized following immersion in a bucket containing 200 mg.I) tricainemethanesulphanate (MS-222), buffered with 400 mg.L-1 sodium bicarbonate. The samesampling method used for the wild fish was also used for the two hatchery groups.After the fish were anaesthetized, the weight and length were measured and thecaudal peduncle was severed and the blood collected in heparinized capillary tubes. Theblood was centrifuged and plasma stored at -80 °C until determination of growth hormoneconcentrations.Growth Hormone Analysis. Plasma GH was measured using a radioimmunoassaydeveloped for chum salmon (0. keta) GH. Analysis for GH was performed according tothe method of Wagner and McKeown (1986). Plasma samples were thawed and triplicate20-1i.1 aliquots were used for the determination of circulating Gil concentration in the fish.All procedures were carried out at room temperature. The addition of 4% polyethyleneglycol to the goat-anti rabbit gamma globulin improved pellet stability in the finalcentrifugation and prevented pellet loss following decanting the supernatant. All volumeswere half of those given in the protocol by Wagner and McKeown (1986).Statistical Analysis. Growth Hormone levels were compared for significantdifferences using an analysis of variance (ANOVA), followed by a Tukey's test to findsignificant differences between means. Statistical significance was taken at a level of a =0.05. All values are expressed as means -± 1 s.e..107RESULTSSeasonal Changes in Plasma Growth Hormone Concentration. The changes in plasmagrowth hormone concentration during the spring of 1991 and 1992 for hatchery and wildcoho are shown in Figure 6.1. In 1991, the hatchery 1 group showed a significant declinefrom March to May. In contrast, the hatchery 2 fish showed an increase in GHconcentration between March and April. By the end of May, the Gil levels had declinedsignificantly from the maximum concentration in April. In 1992, the pattern of changes inplasma GH concentration were similar to that seen in the hatchery 2 fish from 1991. Therewas a gradual increase in Gil levels until a maximum in April. Subsequently, there was adecline in Gil levels in both the hatchery groups. Although the pattern of changes in Gilwere similar between the fish from the hatchery 1 and hatchery 2 groups in 1992; thehatchery 2 fish had consistently higher levels than the hatchery 1 fish. The absolutedifferences, however, were only significant in the February sampling period.In the 1992 measurements, the changes in plasma Gil levels did not differsignificantly over time for the wild fish. GH levels showed the opposite trend to that seenin the hatchery fish; a minimum in April for wild fish as opposed to a maximum for thehatchery fish. Due to the high circulating plasma cortisol concentrations associated withthe capture method used in 1991 (see chapter 5), growth hormone levels of the wild fishwere not measured for March and April. At the beginning of May, the wild fish showedsignificantly greater plasma growth hormone concentrations than the hatchery 2 fish. Bylate May the growth hormone levels had declined to levels similar to the hatchery fish.Changes in Plasma Growth Hormone Concentration After Saltwater Transfer. Thechanges in plasma growth hormone concentration following transfer to salt water weredifferent between 1991 and 1992 (Figure 6.2). In 1991, following saltwater transfer, allthree groups showed an increase in plasma Gil concentration. The increase, however, wasonly significant in the hatchery 1 fish. The surge in Gil following transfer to salt water haddeclined in all groups by seven days. After seven days saltwater exposure, the Gilconcentration was significantly lower than the 24-hr measurement in the hatchery 2 fish.The decline was not as great in the other two groups. In fact, the hatchery 1 fish exhibited108significantly higher plasma GH concentration than the hatcheiy 2 fish.Following transfer to salt water in 1992, growth hormone levels declined significantlyin the wild smolts, and remained low for the rest of the saltwater challenge (Figure 6.2).In the hatchery 1 group, the circulating Gil concentrations did not differ significantly onday one or seven of the saltwater challenge when compared to the freshwater measurement.In the hatchery 2 fish, there was an increase in GH concentration. The increase in GH wasonly slight after one day in salt water. By seven days, however, the concentration of GHin the plasma was significantly greater than the initial levels. At seven days, the GH levelsin the hatchery 2 fish were also significantly greater than the wild and hatchery 1 animals.10910February MayMarch April June3 5Li]^3 002 5CL0^20I25Figure 6.1. Plasma GH concentrations of juvenile coho salmon during the spring of 1991and 1992. Wild fish were caught in the Quinsam River, and hatchery 1 and 2 fish wereseined from the top and bottom sections, respectively, of an earthen rearing channel. n =7 unless indicated adjacent to the data points. Vertical lines represent 1 se. "a" orindicate the values are significantly different (a=0.05) from the April or March level,respectively, for the same group. "c" or "d" indicate the values are significantly different(a=0.05) from the wild or hatchery 1 fish, respectively, for the same sampling period.11030LL1^250200^1510F-- E•0 Z) 350 302520150^1^ 7DAYSFigure 6.2. Plasma GH concentrations of hatchery and wild coho smolts before transferto salt water and after one and seven days. Salinity of the saltwater challenge was 29%oin 1991 and 35%o in 1992. n = 7 unless indicated adjacent to the data points. Verticallines represent 1 se. "a" or "b" indicate the values are significantly different (a=0.05) fromthe freshwater (T=0) or the 24-hr (T=1) value, respectively, for the same group. "c" orindicate the values are significantly different (a=0.05) from the wild or hatchery 1 fish,respectively, for the same sampling period.111DISCUSSIONThe changes in circulating plasma GH concentration during the spring differedbetween the two years in the Quinsam River coho salmon examined in this study. GHlevels in the wild fish also showed a different pattern from the hatchery-reared fish Forthe hatchery 2 fish in 1991 and both hatchery groups in 1992, the highest plasma Gilconcentration occurred in mid April. The 1991 hatchery 1 fish showed the highest GHvalues in March, the first sampling period. All subsequent measurements were lower thanthe March value, and the value at the end of May was significantly lower. As samples werenot taken prior to March in 1991, it is possible that the level of Gil in the plasma duringMarch and April are the peaks in these groups of fish. A second possibility is that thereis no seasonal peak in plasma Gil concentration in this stock of coho. Although a distinctsurge in plasma Gil concentration has been demonstrated to occur during the spring incoho salmon (Sweeting et al. 1985; Young et al. 1989a) and Atlantic salmon (Prunet etal. 1989), no correlation between saltwater adaptability and seasonal changes in Gil levelswere found in sockeye or amago salmon (Yada et al. 1991). In the published studies, thechanges in circulating Gil levels during the spring show considerable variation. There wasnot a consistent rise in Gil levels during the spring, but several surges were seen. Thereis also variation in the timing of maximal plasma GH levels; peaks have been reported inMay (Prunet et al. 1989), July (Young et al. 1989a) and August (Sweeting et al. 1985).Growth hormone levels during the spring show little change in the wild coho in 1991and 1992 (Figure 6.1). The data set for 1991 is not complete, however, and themeasurements include only two sampling times in May. Pickering et al. (1991) showed thatan acute stress resulted in a significant decline in plasma Gil concentration. Consequently,samples taken during March and April were not analyzed for Gil as the capture stressassociated with confinement in the minnow traps (see Chapter 5) may have affected thesevalues. In 1992, the Gil levels in February and March are not significantly different fromthe April and May values. The February and March values consisted of only four samplesfor each period. In both months, two samples were high and two were low. It is not clearwhether the high or the low values are characteristic of Gil levels in the wild smolts at this112time of the year. Many factors can affect circulating GH concentrations in the plasma.Starvation has been shown to increase Gil levels (Wagner and McKeown 1986; Sumpteret a/. 1991). It is possible that food supply was limited for the wild fish at this time of theyear as the condition factors for some wild fish were low in February and March (Chapter5, Figure 5.2). A low condition factor indicates a thin fish. There was a significant increasein the condition factor in April for the wild fish in both 1991 and 1992, possibly indicatingan increase in productivity of the river during the spring as the river temperature anddaylength increased. The increase in condition factor coincided with the reduced GH levelsof the wild fish, supporting this hypothesis.In May of both years, there was little difference in GH concentration between thehatchery and wild salmon. Migrating smolts show higher GH levels when compared tonon-migrant parr in both coho and Atlantic salmon (Varnavsky et al. 1992; McCormickand Bjornsson 1992). This finding is in contrast to my results as the wild smolts werecaught during migration, and yet did not exhibit significantly greater plasma Gil levels thanthe non-migrant hatchery fish (Figure 6.1).During the spring there appears to be a correlation in timing between the surge inplasma GH concentration and cortisol levels in the plasma, particularly for themeasurements taken during the spring of 1992 (Chapter 5, Figure 5.4 for cortisol values).The concomitant increase in Gil and cortisol has also been shown by Young et al. (1989a).The mechanisms of interaction between cortisol and Gil have been discussed elsewhere(Chapter 4). A synchronous rise in Gil and cortisol is likely to increase hypo-osmoregulatory ability maximally (Madsen 1990b,c). The increases in Gil that were seensynchronous with the cortisol surge in the spring and following transfer to salt water mayplay a role in opposing downregulation of CR due to the endogenous cortisol releaseassociated with smolting or stress. The higher plasma Gil concentration in the hatchery2 compared to the hatchery 1 fish and yet lower CR concentration does not support thishypothesis as a prominent role for Gil. It would appear that cortisol (or some additionalfactor) has a much greater impact than Gil in the mediation of CR in the gills of cohosalmon.Growth hormone has a direct effect on increasing saltwater tolerance (Komourdjian113et al. 1976; Chapter 4) by stimulating Na+VATPase activity (Richman and Zaugg 1987,Madsen 1990b,c) and chloride cell numbers (Madsen 1990c). The action of GH onstimulating an increase in osmoregulatory ability in salt water is likely via binding of GHto membrane receptors that are present in the liver, gill and kidney (Fryer and Bern 1979;Sakamoto et al. 1991). The transient increase in GH levels following saltwater transfer isassociated with a decrease in Gil binding to the liver due to receptor occupancy, but nochange in binding to the gill or kidney (Sakamoto and Hirano 1991). The increase inhypo-osmoregulatory ability due to GH may be associated with liver production of IGF-I.McCormick et al. (1991b) showed IGF-I to increase hypo-osmoregulatory abilitysignificantly in rainbow trout.The transfer to salt water resulted in changes in plasma Gil concentration thatdiffered between groups and years. All groups showed an increase in Gil levels afterseawater transfer in the 1991 saltwater challenge. By seven days, Gil levels in the plasmahad declined from the high at 24 hr in all three groups. In contrast, following 24 hrexposure to salt water, Gil levels in the plasma declined in the wild and hatchery 1 fish,and marginally increased in the hatchery 2 fish during the 1992 measurements. An increasein plasma Gil concentration following saltwater exposure has been shown previously(Sweeting et al. 1985; Sweeting and McKeown 1987; Sakamoto et al. 1990, 1991;Rydevik et al. 1990). A lack of an increase in plasma Gil concentration following transferto salt water (Young et al. 1989a; Rydevik et al. 1990) or even a decline in concentration(Yamauchi et al. 1991), however, has also been shown. In the studies that havedemonstrated an increase in Gil concentration, there was a significant perturbation inplasma sodium concentration associated with saltwater exposure. The plasma sodium levelswere similar to the criteria of 170 inEcrL-1 for smolts (Blackburn and Clarke 1987). Thestudies showing little change or a reduction in Gil concentration appear to be associatedwith contrasting findings; fish that show a strong hypo-osmoregulatory ability followingsaltwater transfer and exhibit little change in plasma sodium concentrations, or fish thatshow no saltwater tolerance and exhibit massive increases in plasma sodium concentration.Rydevik et al. (1990) proposed that the increased Gil levels after saltwater exposure may114be to increase ionregulatory ability in fish that are not adapted for salt water. Thishypothesis agrees with the saltwater challenge results in the present study for the 1991measurements. The wild smolts showed the greatest hypo-osmoregulatory ability, asindicated by the smallest elevation in plasma sodium following saltwater exposure (Chapter7, Figure 7.1), and showed the smallest elevation in plasma Gil levels in 1991 (Figure 6.2).In contrast, the hatchery smolts which do not show hypo-osmoregulatory ability as great asthe wild smolts (Chapter 7, Figure 7.1), showed increases in GH levels.The 1992 24 hr saltwater challenge values for the hatchery smolts and the findingsof Yada et al. (1991) are not consistent with this hypothesis. The significant perturbationin plasma sodium concentration after 24 hr saltwater exposure in the hatchery groups(Chapter 7, Figure 7.1) should have elicited a significant surge in plasma GH concentrationaccording to the proposal by Rydevik et al. (1990). The apparent discrepancy in thesefindings could be related to the massive elevation in plasma cortisol exhibited by thehatchery smolts following transfer to salt water (Chapter 7, Figure 7.4). Pickering et al.(1991) found that an acute increase in cortisol depressed GH levels in the plasma. Thecortisol rise 24 hr following transfer to salt water could account for the lack of a significantchange in plasma Gil levels in the 1992 hatchery 1 and 2 smolts. It is likely that twoopposing stimuli are affecting plasma GH concentration. Saltwater exposure is enhancingGil levels and the acute rise in cortisol is depressing Gil levels. The consequence is littlechange at 24 hr in the saltwater challenge.One week following transfer to salt water, GH levels in the plasma were notsignificantly different from the freshwater values in all the groups examined, except thehatchery 2 smolts in 1992 (Figure 6.2). The surge in Gil concentration following transferto salt water is transient and has not been documented to persist for longer than a few days(Sweeting and McKeown 1987; Sakamoto et al. 1990, 1991) and even to decline belowfreshwater levels (Young et al. 1989a) in smolts adapting to salt water. The declinebetween one and seven days in GH levels in 1991 and in the wild fish in 1992 agree withthe published findings. The 1992 measurements on the hatchery 2 smolts showed asignificant increase and the hatchery 1 smolts showed a slight increase, however, in Gillevels by seven days. The salinity of the saltwater challenge was higher in 1992 than 1991115(35%e and 29%0, respectively), and this could account for the differences seen in the GHlevels following saltwater transfer. The persistent increase may be associated with anosmoregulatory disfunction, as significant mortality occurred in this group of fish followingsaltwater transfer (Chapter 7, Figure 7.9). Elevated GH levels have been associated withimpaired growth in salt water in coho stunts. After several months in salt water, fish thatdo not grow, and become thin and emaciated are referred to as stunts. Stunts have beencharacterized as having depressed levels of cortisol and thyroid hormones, but exhibitsignificantly greater plasma GH levels than smolts (Bolton et al. 1987b; Bjornsson et al.1988; Young et al. 1989b). Despite the endocrine dysfunction in stunts, they have plasmasodium levels similar to smolts in sea water. The high salinity of the saltwater challengein 1992 may have forced the hatchery 2 fish to stunt. The elevated GH in the plasma maybe an attempt to maintain ionic homeostasis.The changes in plasma GH concentration during the spring and following transferto salt water differ between hatchery and wild smolts. The GH levels measured followingtransfer to salt water confirm the findings that the wild fish show a strong pre-adaptiveability to osmoregulate effectively in salt water, and that the parr-smolt transformation inthe groups of hatchery coho examined in the present study does not appear complete.CHAPTER 7Saltwater Tolerance in Wild and Hatchery-Reared JuvenileCoho Salmon (Oncorhynchus kisutch) Smolts116117INTRODUCTIONCoho salmon (Oncorhynchus kisutch) represent an economically important speciesfor the commercial and sport fisheries in British Columbia. Large numbers ofenhancement facilities have been established to increase the numbers of smolts releasedinto streams, with the objective of supplementing the wild fishery. Therefore, it isimportant to produce a fish with as great a chance of survival as possible. The rearingenvironment within a hatchery has been shown to have a substantial impact ondevelopment and smolting in salmonids (Schreck et al. 1985; Patino et al. 1986). Thesmolt to adult survival is greatest for fish that show a high degree of smolt developmentprior to release from the hatchery (Zaugg and Mahnken 1991). The adult returns ofsahnonids produced within the British Columbia hatchery systems, however, have declinedover the past decade (Cross et al. 1991). It is not clear what factor is limiting the successof the smolts that are produced within the hatchery system. Poor ocean survival of smoltsproduced by enhancement facilities has been suggested as a possible explanation for thedecline in these salmon stocks (Walters 1988; Wedemeyer et al 1980).The migration of smolts into the ocean is timed to coincide with a number ofmorphological and physiological changes that prepare the fish for transition into thehyperosmotic medium. In fresh water, the fish must actively take up ions, and prevent theinflux of water to maintain body ion homeostasis. In sea water, the fish must activelyexcrete ions and drink water to compensate for the water lost to the environment. Theprocess of smoltification pre-adapts fish to salt water and minimizes the perturbations inplasma ions following transition into salt water.Typically the transition from fresh water to the marine environment is a time oflarge mortality in juvenile salmonids (Healey 1982). Fisher and Pearcy (1988) concludedthat the success of a year class was determined early in the summer, soon after mostjuvenile coho had entered the ocean. It would appear that the developmental stage of thefish that are released from the hatchery is critical for the survival of these animals to adult(Zaugg and Mahnken 1991).118I conducted experiments to compare differences in saltwater tolerance betweenhatchery-reared and wild coho salmon smolts. Quinsam River coho salmon were used forthis investigation. The Quinsam River provided a unique opportunity to study the effectsof rearing environment on selected aspects of the physiology of coho salmon and directlycompare wild and hatchery populations of fish, from a common stock. In addition, acolonization program for hatchery fry release into the upper Quinsam watershed made itpossible to compare fish of hatchery origin reared in a wild environment, with fish thatwere reared exclusively within a hatchery. Experiments were conducted during May of1990, 1991 and 1992.119MATERIALS AND METHODSFish. Juvenile coho salmon smolts from the Quinsam River were used to examinethe affect of rearing environment on saltwater tolerance over three years, 1990 - 1992. In1990, four groups were chosen; wild, colonized, hatchery 1 and hatchery 2. Wild smoltswere fish that had developed from eggs spawned naturally in the Quinsam River.Colonized fish were a group spawned and raised at the Quinsam River Salmon Hatcheryfor ten months and then released as fiy, in September, into the upper Quinsam watershed.The fish resided in the upper Quinsam until May of the following year, at which time theymigrated downstream. This group had not been marked and was distinguished from thewild smolts on the basis of size, as earlier marking studies showed a bimodal sizedistribution existed for the fish migrating down the Quinsam River. The colonized fishcomprised the larger population (D. Ewart, Quinsam River Salmonid Hatchery, personalcommunication). Hatchery 1 and hatchery 2 were smolts that had been reared in the top(water inflow) and bottom section (water outflow), respectively, of an earthen channeldivided into 3 sections, at the Quinsam hatchery. The earthen channel was supplied withground water. The water was first passed through rearing containers on the Quinsamupper site and then passed through a clarifier to remove gross particulate matter. Theoutflow from the clarifier fed into the top section of the earthen channel. In 1991 and1992, the colonized fish were not included in the saltwater challenge (SWC) tests. Themean lengths and weights of the fish from all groups used over the three years in this studyare presented in Table 1.Experimental procedures and sampling. In May, 1990, wild and colonized smolts werecaught during their downstream migration and held overnight in the trap at thedownstream enumeration fence across the Quinsam River. The fish were then transferredto aluminum troughs. The two hatchery groups were seined from the top and bottom ofthe earthen channel. All four groups were held in aluminum troughs for less than a weekand starved for 48 hr prior to experimentation. To examine hypo-osmoregulatory ability,the fish were transferred into 100 1 buckets containing salt water on May 25, 1990. Ten to12012 fish were placed in each bucket. The density of fish in the bucket was maintained belowthe 3.0 g.L-1, as recommended by Blackburn and Clarke (1987) for a saltwater challenge.Each bucket that held fish was part of a recirculating saltwater system. The recirculationsystem consisted of three 100 L buckets. Two buckets held the fish and drained into acommon reservoir between them. The reservoir was aerated and the water was thenpumped back into the buckets containing the fish. Several recirculation systems wereplaced into an aluminum trough and surrounded with flowing fresh water to maintainconstant temperature. Despite the cooling system, there were slight increases in watertemperature during the day (< 3 °C), but these temperature changes were similar for allgroups. Dissolved oxygen levels did not drop below 8.5 ppm during the day.Approximately one-third of the water was replaced daily to maintain water quality.In 1990, the fish were sampled before exposure to salt water (0), and after 1 and 7days in salt water. Salinity of the water in the challenge was 29 %o, and the temperatureaveraged 12.6 °C. At the designated intervals, the fish were killed by a cephalic blow, thecaudal peduncle severed and blood collected in heparinized capillary tubes. The tubes werecentrifuged for 5 minutes at 11 500 g, haematocrit read, and plasma was saved.Approximately 50 mg of gill tissue from each fish was removed and placed into 0.5 ml ofice cold SEI (0.3M sucrose, 0.02M EDTA and 0.1M Imidazole) for determination ofNa+VATPase activity. A further 50 mg of gill tissue was removed and placed in ice coldHET (0.02M Hepes, 0.001M EDTA and 0.1% Triton X-100) for the determination ofcitrate synthase activity. All samples were stored at -80 °C prior to analysis.In 1991 and 1992 the same recirculating saltwater system was used to assess saltwatertolerance. The protocol for sampling, however, was modified to include measurement ofplasma cortisol concentration. To prevent the stress related rise in cortisol followingnetting, all fish were rapidly transferred into a bucket of tricaine methanesulphonate (200mg.L-1, MS-222) buffered with sodium bicarbonate (400 mg.I:1) after capture. This dosehas been shown to inhibit the stress-related rise in plasma cortisol concentration in salmon(Barton et al. 1986). On May 19 and 26, 1991, and May 17, 1992, wild fish were caughtduring their nocturnal migration as they passed through the downstream enumeration fenceon the Quinsam River. The following day, their hatchery counterparts were randomly121sampled from the top and bottom section of the earthen channel. After the fish had beenanaesthetized, the caudal peduncle was severed, and the fish were bled and sampled asdescribed above. Additionally, in 1991, the second gill arch was removed and placed inbuffered formalin for determination of chloride cell concentration on the primary lamellae.In 1991, the fish were sampled after 0,1,2,3,7 and 14 days in salt water. In 1992, the fishwere sampled after 0,1 and 7 days in salt water. Salinity for the saltwater challenge was 29%o in 1991, and 35 %o in 1992. The temperature ranged from 13.0 to 15.3 °C during theday in 1991, but was slightly cooler in 1992; 12.0 to 14.6 °C.Analysis. Plasma sodium samples (1990) were measured on a Shimadzu ionchromatograph, model HIC-21 with a 5 mM nitric acid mobile phase. The 1991 and 1992plasma sodium samples were measured on a Perkin-Elmer Atomic AbsorptionSpectrophotometer, model 2380. Na+IC-ATPase activity was measured following themethod of Zaugg (1982). Citrate synthase activity was measured following the method ofSuarez et al. (1986). Enzyme activities were expressed per mg protein of the gillhomogenate. Protein concentration was determined by the biuret method according to theprotocol outlined in Alexander and Ingram (1980). Plasma cortisol concentration wasdetermined by radioimmunoassay using 125I-cortisol (Incstar, corporation). In 1990, samplesfrom six fish (n=6) were assayed for each parameter measured. In 1991 and 1992, thenumber of samples measured for each group was increased to ten (n=10). Mortalitiesduring the 1992 SWC reduced the number of samples that could be taken after seven daysin salt water for the hatchery fish.Histology. For consistency, the second gill arch from the left side was removed andplaced in 10% buffered formalin immediately after killing and sampling the fish. Aftersamples had been fixed in formalin, they were dehydrated through graded series of absolutealcohol to xylene and then imbedded in paraffin blocks. The blocks were cut in 6 - 10 gmsections, mounted on slides and stained with Harris's haematoxylin and eosin stains.Each microscope slide was examined for interlamellar chloride cell density (thenumber of chloride cells located on the primary lamellae). Seven fields for each gill archwere randomly selected and examined to determine the number of chloride cells, andcalculate a mean for each fish. A grand mean was then calculated for each group (n=6).122Statistical analysis. Parameters measured were compared for significant differencesusing Analysis of Variance, followed by a Tukey's test to determine significant differencesbetween means. Linear regression analysis was performed on the plasma sodiumconcentrations against fish weight for fish held in water for 24 hr. Statistical significancewas determined to be at a level of a = 0.05. All values are expressed as means -± onestandard error.123RESULTSPlasma Sodium Concentration. In all years, the plasma sodium concentrations forthe fish in fresh water did not differ significantly between groups (Figure 7.1). After 24 hrin salt water, all groups of fish showed an increase in plasma sodium concentration. Theelevation in plasma sodium concentration was significantly greater in the hatchery rearedgroups (hatchery 1 and 2) than the wild or colonized fish. The plasma sodiumconcentrations in the wild and colonized fish following 24 hr exposure to salt water did notdiffer significantly (Figure 7.1a). The concentration of sodium in the plasma declined byseven days to a level not significantly different from the freshwater values in the wild andcolonized fish, but was still significantly elevated in the two hatchery groups. The trendsobserved in 1990 were also seen in the 1991 and 1992 SWC tests. Plasma Na+concentration showed a significantly greater perturbation, and the perturbation persistedat a higher level for longer in the hatchery fish than for the wild fish. The May 25, 1991,plasma sodium concentrations after 24 hr exposure to salt water for the wild fish, did notshow a significant change from freshwater levels. Despite the higher salinity of the SWCin 1992, the wild fish still did not show a significant elevation in plasma sodiumconcentration following transfer to sea water.There was no relationship between size of the fish within each group and theelevation in plasma sodium concentration following transfer to salt water for 24 hr (Figure7.2). The slopes of the regression lines did not differ from zero for any of the groups. Adirect relationship between plasma sodium concentration (24 hr SWC) and weight existedwhen all the groups of fish are examined together, as the smaller wild and colonized fishexhibited the smallest perturbation in plasma sodium concentration following seawatertransfer. The relationship for all the groups pooled together was:[Na] = 161.3 + 1.17*Wt^R2=0.1558, n=120Similar results were obtained when plasma sodium was regressed against length (data notshown).Haematocrit. Changes in haematocrit following transfer to sea water were greatest124in the hatchery fish, exhibiting a decline in haematocrit that was significantly lower thanthat in the colonized and wild fish, and the freshwater haematocrit values (Figure 7.3).After seven days in sea water, the haematocrit had declined significantly in the colonizedand wild fish as well, and there was no significant difference among any of the groups.Plasma Cortisol. The circulating cortisol concentration in the plasma of fish in freshwater was the greatest in the wild fish (Figure 7.4). Following 24 hr of exposure to saltwater, there was a significant increase in circulating cortisol levels in hatchery fish (Figure7.4). The elevation in plasma cortisol appeared to be dependent on the severity of theosmotic challenge, as the hatchery fish that were challenged to 35 %o in 1992 (Figure 7.4c)exhibited a much greater elevation in plasma cortisol than the fish challenged to 29 %o(Figure 7.4a,b) the previous year. After 48 hr in salt water, the cortisol titers had declinedto freshwater levels. The rapid increase in circulating cortisol concentration in the plasmawas not seen in the wild fish. The wild fish initially had a significantly greater plasmacortisol concentration than the hatchery fish, but this level gradually declined during thechallenge.Enzyme Activity. Specific activities of Na+VATPase and citrate synthase in the gill,also differed between the wild and hatchery-reared fish. Levels of Na+K+ATPase activitywere consistently lower in the hatchery groups (Figure 7.5). In 1990, the difference wasonly significant between the colonized and hatchery 2 fish, before and after 24 hr in seawater. In 1991, the Na+K+ATPase activity showed a significant increase after 14 days insalt water in all groups. Due to the increase in enzyme activity in the wild fish gills, therewas still a significant difference in Na+K+ATPase activity between the wild and hatcheryfish. The difference in levels of citrate synthase activity in the gills were similar to that ofNa+ICATPase (Figure 7.6). The hatchery reared fish gills had significantly lower enzymeactivity than the wild and colonized fish.Chloride Cells. Chloride cell density on the primary lamellae was significantly greaterin the wild fish than both the hatchery groups, prior to exposure to salt water (Figure 7.7).The wild fish did not show an increase in chloride cell density following transfer to saltwater for 14 days. After 7 days in salt water, the number of chloride cells had significantlyincreased in both the hatchery groups, however, the increase was significantly greater in125hatchery 1 than hatchery 2 fish. By 14 days, there was no difference between the chloridecell concentration in the 3 groups.Figure 7.8 shows the relationship between Na÷1C+ATPase activity and chloride celldensity in the gills for the 14 day SWC conducted in 1991. The two hatchery groupsshowed both an increase in chloride cell density and Na+K+ATPase activity, whereas thewild fish show only an increase in Na+K+ATPase activity.Mortality. There was little mortality of smolts during the SWC experiments carriedout in 1990 and 1991. In 1990, two hatchery 2 smolts died after 24 hr in salt water and onewild smolt died after five days in salt water. There was no mortality due to saltwaterexposure in the smolts sampled during 1991. At the higher salinity (35 %o) of the SWCconducted in 1992, there was considerable mortality of the hatchery smolts. No wild fishdied throughout the study period, but approximately 25% and 30% of hatchery 1 and 2 fishdied over a one week period, respectively. The cumulative mortality for the wild andhatchery fish for the 1992 SWC is shown in Figure 7.9.126Table 7.1. Mean weight and length of the wild, colonized and hatchery Quinsam Rivercoho salmon smolts used in the saltwater challenge experiments.1990 1991 1992WildWeighttLengtht8.03 ± 0.27g9.65 .± 0.1111.77 ± 0.3810.85 ± 0.149.75 ± 0.5710.26 ± 0.19n 32 100 42Weight 16.17 ± 0.94Colonized Length 12 28 ± 0.22ii 33Weight 27.34 ± 0.90 21.36 ± 0.53 24.71 ± 0.98Hatchery 1 Length 14.19 ± 0.15 13.18 ± 0.10 13.63 ± 0.16n 34 106 44Weight 25.87 ± 1.44 22.92 ± 0.73 26.79 ± 1.10Hatchery 2 Length 13.95 ± 0.22 13.38 ± 0.12 14.06 ± 0.16n 31 97 40Units are g wet weightt and cm fork lengtW. gAll values are means ± s.e.160140220200180aba/aIIIIIIIMay 25, 1990II^i^iii^1-^ May 26,T baAai1991^—T aVi^1^IIIII-la b-IIIIIIII-_Tabzz■D-------._________9May 17,1111111 --11992_••AvWILDCOLONIZEDHATCHERY 1HATCHERY 2jab1111111May 19,^19912202001801401600^1^2^3^7^14^0^1^2^3^7^14DAYSFigure 7.1. Plasma sodium concentration for smolting juvenile coho salmon following direct transfer to saltwater. Initial measurement (time = 0) were values for fish held in fresh water. "a" indicates value issignificantly different from time 0 value of the respective group, and "b" indicates value is significantly differentfrom wild fish value for same time interval (a = 0.05).1 1. .A•-• 7• WILD• COLONIZEDA HATCHERY 1Y HATCHERY 2A• AAA^AA NVV A AVA V--A^ AL-v3T • V^• AA••A^V.I^I---•1 • 140 . A • •• V- • I • 0^• A11-Thir------ •I^V r• •^Irli^A0.A.A-250225E3Toc) _,200< ifE0^175<....1DL15012810 20 30 40WEIGHTgFigure 7.2. The relationship between plasma sodium concentration after 24 hr in sea waterand the weight of the fish. Symbols for each group are as shown in figure 7 1 Linearregression lines are shown for each group. R2 for the regression lines are 0.0035 (wild),0.0508 (colonized), 0.0011 (hatchery 1) and 0.0049 (hatchery 2).50I--()^45CD0H— 40<< 35I30129• WILD• COLONIZEDA HATCHERY 1^ HATCHERY 20^1^2^3^4^5^6^7DAYSFigure 7.3. The change in haematocrit following transfer to salt water for hatchery, wildand colonized smolts in 1990. Letters indicate significant differences between groups asdescribed in figure 7.1.130May 26, 1991^-1501 00500(f)17=^1001:10^75r^E50< c25Cir)^0OL3002502001501005000^1^2^3^7^14DAYSFigure 7.4. Circulating plasma cortisol concentrations of hatchery and wild coho smoltsbefore and after transfer to salt water. See figure 7.1 for description of symbols.0 1 2 3 7 14131• WILD• COLONIZEDA HATCHERY 1v HATCHERY 2151 07^2010--^II 19901 i 11 1 1-^^!--DAYSFigure 7.5. Specific activity of Na+K+ATPase in the gills of hatchery, wild and colonizedcoho salmon smolts before and after transfer to salt water for three consecutive years. Seefigure 7.1 for description of symbols.5• WILD• COLONIZEDA HATCHERY 1^ HATCHERY 20 1 2 3 4 5 6 7DAYS132Figure 7.6. Citrate synthase activity in the gills of coho salmon during a saltwater challengeconducted in May, 1990. Symbols description as in figure 7.1.31330^7^14D YsiiFigure 7.7. The change in interlamellar chloride cell density for hatchery and wild cohosmolts following transfer to salt water for 14 days in 1991. "c" indicates value is significantlydifferent from hatchery 1 value for same time interval (a = 0.05). All other symbols asdescribed in figure 7.1.1341 8T1...._c 16.E 14-60 12L_.0_10E 8CL:- 6-6E 422.0^2.5^3.0^3.5^4.0CHLORIDE CELLSInterlamellar DensityFigure 7.8. The relationship between Na+K+ATPase activity and chloride cellconcentration in the gills of hatchery and wild coho salmon smolts when exposed to saltwater for 14 days. Data is from a SWC conducted in 1991.135-i—E 3 0>--- (c1:))E-- L--- Q)__I a.,_< cio 2 0F._ >OL :460 D 1 07 EDCD00^1^2^.3^4^5^6^7DAYSFigure 7.9. Cumulative mortality for hatchery and wild coho salmon smolts challenged to35 Too salt water for seven days in 1992.136DISCUSSIONThe hatchery fish showed consistently a lower hypo-osmoregulatory ability comparedto the wild or colonized fish. Both the hatchery-reared groups of fish exhibited a muchgreater perturbation in plasma sodium concentration following seawater transfer than thewild smolts (Figure 7.1). The large elevation in plasma sodium is a reflection of the factthat fish do not possess adequate physiological ability to hypo-osmoregulate in sea water(Blackburn and Clarke 1987). I found no evidence for a size relationship in saltwatertolerance within the groups of fish used in this study (Figure 7.2). Comparison among thehatchery, wild and colonized smolts all together showed a correlation between saltwatertolerance and fish size. The smaller wild and colonized fish appeared to be better adaptedfor the transfer to salt water, indicated by the small increase in plasma Na+ concentrationfollowing seawater exposure. This is in contrast with the finding that larger salmonidsmolts have greater gill Na+K+ATPase activity (Ewing et al. 1979), and appear to possessa greater hypo-osmoregulatory ability than smaller fish (Hoar 1988). Due to the largersize of the hatchery smolts, it would be expected that they should possess a greater hypo-osmoregulatory ability when compared to their wild counterparts. The size differentialbetween the wild and hatchery smolts, therefore, can not account for the significantdifference in hypo-osmoregulatory.Capture of the fish, and holding them in the trap at the downstream enumerationfence exposed the wild smolts to stressors that were vastly different from those normallyencountered in the river environment. The holding conditions that the smolts experiencedduring the saltwater challenge were also more similar to those encountered by the hatcheryfish. In addition, the hatchery smolts were not subjected to a treatment similar to thatexperienced by the wild smolts in the enumeration trap. Consequently, the wild smoltsexperienced stressors that the hatchery smolts did not encounter, which would be expectedto be disadvantageous for the wild smolts. On this basis, the differences in hypo-osmoregulatory ability found between the hatchery and wild smolts may be a conservativeestimate.137The rapid change in haematocrit following transfer to salt water also indicates thatthe hatchery smolts experience a large osmotic perturbation (Figure 7.3). Blackburn andClarke (1987) also report a decrease in haematocrit after 24 hr exposure to salt water, andattributed the reduction to dehydration of the red blood cells. Similar reductions inhaematocrit were also reported by Brauner et al. (1992). They speculate that an increasein plasma volume can account for the decline in haematocrit.The greater saltwater tolerance of the wild smolts is associated with higher activitiesof the enzymes Na+K÷ATPase (Figure 7.5) and citrate synthase (Figure 7.6) in the gills.Na+K+ATPase is the primary enzyme for excretion of NaC1 in saltwater teleosts (Epsteinet al. 1980). Citrate synthase activity also increases during the spring, possibly to meetenergetic demands associated with cellular differentiation that occur in smolts (McCormicket al. 1989b). The wild fish had a higher Na+K+ATPase and citrate synthase activity, anda greater concentration of chloride cells on the primary lamellae than both groups ofhatchery-reared smolts. The greater activity of the enzymes and concentration of the cellsresponsible for excretion of Na+ and Cl- across the gills in a saltwater teleost would reducethe osmotic perturbation in the plasma of fish during migration into the ocean.The rapid transfer of fish to salt water is an extreme test to measure their ability totolerate a hyperosmotic challenge. Normally fish would migrate into the estuary and reside,possibly enabling the fish additional time to adapt to the increase in salinity. Levings et al(1986) found that wild juvenile chinook smolts reside in the estuary for 40 - 60 days,whereas, hatchery smolts used the estuary for approximately half that time. Theimportance of residence in an estuary to the survival of smolts has been shown byMacdonald et al. (1988). Large hatchery coho, however, were found not to utilize theestuary extensively, but were found to inhabit deeper more saline waters further from shorethan their smaller wild counterparts (Argue et al. 1986; Macdonald et al. 1987).Macdonald et al. (1987) speculated that the larger animals were more likely to bephysiologically prepared for movement to the marine environment than smaller salmon.In contrast to this, I found that the larger hatchery smolts showed a weaker hypo-osmoregulatory ability than the smaller wild smolts. The migration of the hatchery smolts138directly into the more saline, outer estuary would not be beneficial for survival. Thelocation of the hatchery smolts in the outer estuary may be due to a preference for highersalinity water. The preference for more saline waters, however, may develop even thoughthe smolts are not well adapted for the marine environment (Wedemeyer et al. 1980).The time for fish to adapt completely to salt water has been shown to beapproximately two weeks (Leray et al. 1981). After transfer to sea water, there is an initialcrisis period where plasma Na+ and a levels become elevated. On the basis of ion levels,it would appear that the animals are fully acclimated to sea water by approximately six days.There is evidence, however, that complete physiological adaptation to salt water may takesubstantially longer. Zaugg and McLain (1970) found that coho salmon smolts releasedinto salt water take approximately 30 - 35 days for Na+K+ATPase activity to reach amaximum. Based on the saltwater tolerance and the areas of the estuary the hatchery fishinhabit, it is possible that entry into the marine environment is a time of high mortality.If there is not adequate time or space for the fish to remain in the estuary, the smolts thatenter the marine environment directly will likely experience an elevation in plasma sodiumconcentration. The implications of a perturbation in the plasma sodium concentration isa reduction in physiological performance. This could be manifested as a decrease inswimming performance. Brauner et al. (1992) showed that the swimming ability of fish iscorrelated to plasma sodium concentration. As plasma sodium concentration increased theswimming ability of the fish decreased. Consequently, following saltwater exposure thehatchery fish suffer a much greater decrease in swimming performance (Brauner et al.1993) and may be more susceptible to predation.I have presented evidence that the Na+K+ATPase activity may not have reached amaximum by two weeks. In contrast, chloride cell concentrations do appear to be maximalwithin two weeks after transfer to salt water. This is clear for wild coho, where there is noincrease in chloride cell concentration following transfer to sea water, but Na+K+ATPaseincreased significantly. Hatchery smolts showed an increase in chloride cell density andNa+K+ATPase activity following transfer to sea water. Therefore, complete adaptation tosea water may take more than two weeks even though the fish appear to be adapted to saltwater based on stabilization of plasma sodium concentrations. The greater proliferation139of chloride cells in hatchery 1 than hatchery 2 fish, could be related to corticosteroidreceptor concentration in the gills. As the fish in the hatchery 1 group possess a greaternumber of corticosteroid receptors in the gills than hatchery 2, they would be expected toshow a greater sensitivity to cortisol (Chapter 5). The surge in cortisol following seawatertransfer may provide the stimulus necessary for an increase in chloride cell proliferation.This may account for the more rapid proliferation of chloride cells following transfer to saltwater in the hatchery 1 smolts.There may be several reasons for the differences in saltwater tolerance between thefish that are reared in a wild environment and those that are reared in a hatcheryenvironment. There is evidence that hatchery stocks have a significantly lower amount ofgenetic variability than wild populations (Waples 1991). It is possible that reduced geneticvariability in a hatchery population may lead to reduced fitness (Danzmann et al. 1989),and may compromise the smolts saltwater tolerance. However, the colonized and hatcheryfish used in this study were the progeny of the same broodstock. As the colonized fishshow as great a saltwater tolerance as the wild fish, it is not likely that reduced variabilitywithin the gene pool can account for the differences seen in the saltwater tolerance of wildand hatchery reared coho salmon.An alternate hypothesis to account for the difference in smolting in the hatchery andwild fish is that rearing environment affects smoltification. Seasonal changes in structureand function occur prior to seawater exposure and are responsible for increased salinitytolerance during smoltification (McCormick and Saunders 1987). These seasonal changesare a response to environmental stimuli. Photoperiod plays a large role in the timing anddevelopment of smolting in coho and Atlantic salmon (Clarke et al. 1981; McCormick etal. 1989a). The seasonal change in photoperiod will be similar for both hatchery and wildfish. There are, however, several stimuli experienced by the wild fish during the spring thatare not present or are suppressed in a hatchery environment. At the Quinsam Riverhatchery, ground water is used to supply the ponds. In comparison to the river, there isalmost no change in water temperature over the year. The change in ambient watertemperature during the spring may be a stimulus for smoltification (Hoar 1988). The lackof an increase in water temperature may adversely affect smolting in the hatchery fish.140There does not appear, however, to be a relationship between smolt survival in the marineenvironment and rearing juvenile salmonids using ground water or surface water in theSalmonid Enhancement Program hatcheries of British Columbia (Cross et al. 1991).Diet is different between hatchery and wild fish. Hatchery fish are fed a highprotein diet composed primarily of herring meal. Nutritional studies on hatchery fish havedemonstrated that diet affects growth, development and marine survival of sahnonids.Addition of polyunsaturated fatty acids to the diet improved the rate of smolt productionin masu salmon, 0. masou (Ogata and Murai 1989). Higgs et al. (1992) found that fattyacid composition affected the smolt to adult survival in chinook salmon. The fatty acidcomposition in the feed altered the lipid composition in the fish tissues. The change inmembrane composition may have an involvement in gill function and ion transport (VanPraag et al. 1987). A comparison of fish reared in a semi-natural environment to hatcheryfish, however, did not indicate any difference in marine survival (Mundie et al. 1990). Thediet of fish reared in a semi-natural environment consisted of natural prey items thataccounted for a significant proportion of the dietary items. The survival (as estimated byadult returns) of the semi-natural smolts was similar or slightly lower than the survival ofthe hatchery fish. Therefore, a wild diet has not been shown to confer a significant survivaladvantage for coho smolts.The rearing conditions within the hatchery also affect smoltification. Increasedrearing density causes a reduction in Na+K+ATPase activity in the gills (Strange et al. 1978;Patino et al. 1986), and reduced saltwater tolerance (Schreck et al. 1985). Increasedrearing density during the freshwater residence of hatchery coho has been shown to beinversely related to adult returns (Fagerlund et al. 1983). Rearing density within ahatchery is much greater than that experienced by wild fish in the river. Increasing rearingdensity to maximize production is stressful to fish. The source of the stress may be socialinteraction (Peters et al. 1980; Ejike and Schreck 1980) or a reduction in water quality.An increase in suspended solids as non filterable residue (NFR) in the water will cause anincrease in circulating levels of cortisol in the plasma (Redding et al. 1987). The materialcomprising the NFR in the river is composed almost entirely of inorganic material and is141different from the hatchery water which is composed of predominantly organic material.In addition, the high organic content of the water as a result of accumulation of food andfaecal matter promotes bacterial production (K. Campbell, Quinsam River SalmonidHatchery, personal communication). The high organic content of the NFR in the hatcherymay be a source of physical and chemical damage to the gills (Banks et al. 1977). Thus,fish at the entrance to the rearing channel are exposed to much cleaner water than fish inthe middle or bottom sections. This results in physiological and anatomical differencesbetween the fish at different sites in the rearing channel. In the present study, not muchdifference was found between the fish from the top and bottom of the earthen channel.As the inflow water to the channel had previously supplied other parts of the hatchery, thelack of a significant difference between hatchery 1 and 2 fish may indicate the clarifiers arenot effective enough at removing particulate matter and both groups are exposed to lowquality water.The hatchery rearing environment appears to have profound effects on the smoltingphysiology of sahnonids. As cortisol levels in the plasma increase following many rearingpractices common to hatcheries (Donaldson 1981; Barton and Iwama 1991), thedevelopment of saltwater tolerance and smolting will be affected by rearing environment.Cortisol has a direct effect on the concentration of corticosteroid receptors in the gills ofcoho salmon. Physical disturbance resulting in the release of endogenous cortisol andexogenous treatment with cortisol cause a downregulation of the corticosteroid receptors(CR) in the gills (Chapter 3). The concentration of corticosteroid receptors in the gills wassignificantly different between hatchery and wild fish, and it was proposed that the hatcheryfish were less sensitive to the surge in cortisol during the spring because of the lowernumber of corticosteroid receptors present in the gills (Chapter 5). It is possible, therefore,that rearing practices within the hatchery that are stressful, lead to a downregulation of thecorticosteroid receptors in the gills and decrease the sensitivity of the hatchery fish to thecortisol surge in the spring when the hatchery fish smolt. The lack of response to anenvironmental trigger for smolting can account for the reduced saltwater tolerance observedin the hatchery smolts which in turn could account for the differences in survival betweenthe wild and hatchery fish.CHAPTER 8GENERAL DISCUSSION143Seasonal changes in corticosteroid receptor concentration and affinity occurred injuvenile coho salmon. A reduction in gill CR concentration and affinity was correlated withphysiological changes that enhanced hypo-osmoregulatory ability. The changes in gill CRobserved could be a function of endocrine changes associated with the parr-smolttransformation, notably plasma cortisol and growth hormone (Hoar 1988). Cortisol wasfound to downregulate gill CR by decreasing receptor concentration and affinity (Chapter3), and GH was found to upregulate gill CR by increasing receptor concentration (Chapter4). The dynamics of plasma cortisol (Chapter 5 and 7) and growth hormone (Chapter 6)concentration that occurred seasonally and following transfer to salt water regulate gill CR.Stress was also found to affect CR concentration in the gills (Chapter 3). Themechanism by which stress affected CR is likely mediated by cortisol. The rise incirculating plasma cortisol concentration following the acute stressors inflicted on the fishhad the same effect as exogenously administered cortisol, a reduction in CR concentration(Chapter 3). It is also possible that the stress associated rise in plasma cortisolconcentrations affected Gil levels. Stress has been shown to affect growth rate (Bartonand Iwama 1991) and plasma growth hormone levels (Pickering et al. 1991). The effectof stress on GH levels was biphasic however, with chronic stress resulting in an increase inGil and acute stress depressing Gil concentrations. Following multiple acute stresses asperformed in Chapter 3 (and commonly occur in a hatchery environment), cortisol levelsrise. This acute rise in cortisol would be expected to lead to a reduction in plasma GHlevels. The consequence is that CR numbers might be depressed due to the rise in cortisolconcentration and a reduction in Gil concentration. These two factors could account forthe reduced numbers of CR detected in the gills of hatchery-reared coho salmon, comparedwith their wild counterparts.Corticosteroid receptors were found to be modified following cortisol treatment bychanges in affinity and concentration. The observed changes in affinity of the receptors arenot likely due to the transcription of different CR genes. Scatchard analysis consistentlyresulted in linear relationships. The Hill coefficients had a value of approximately one inall the samples examined. These findings indicate that the CR is a single class receptor144with no cooperativity and cortisol binding to the receptor follows classic mass action laws.If a second form of CR with a different affinity for cortisol was synthesized during any ofthe experiments examined, it would be likely that the two forms could occur simultaneouslyin the gills. The presence of two receptor molecules with different affinities would beshown by a curvilinear Scatchard plot. No curvilinear Scatchard plots were seen followingchronic cortisol treatment (Chapter 3) or seasonally (Chapter 5).Affinity of the CR in the gills could be regulated by intracellular modulators. Thereceptor is thought to be associated with heat shock proteins, and steroid binding isproposed to cause the dissociation of the CR from the heat shock proteins, therebypermitting receptor binding to DNA (Sanchez et al. 1985). An endogenous low molecularweight activation inhibitor has also been identified (Bodine and Litwack 1990). Thesemolecules, however, function to stabilize steroid hormone receptors and are not likely tohave an effect on receptor affinity for the ligand.Some other mechanism(s) must account for the changes in affinity observedseasonally or following cortisol administration. A conformational change occurs in steroidhormone receptors following ligand binding that results in a decrease in affinity (Nemotoet al. 1990). Lee et al. (1992) suggest that conformational changes of the receptor mayaccount for the decreased affinity seen during downregulation. This hypothesis couldexplain the changes in kJ, observed in Chapters 3 and 5. The lower affinity did not persistfor as long as the reduction in b. following the decline in plasma cortisol concentration(Chapter 3) and a gradual increase in kp occurred during the spring when endogenouscortisol levels in the plasma began to rise (Chapter 5). It is, therefore, possible that cortisolmay have a direct effect on CR affinity.In mammals, downregulation of glucocorticoid receptors (GR) is due to theinhibition of transcription of the GR gene following treatment with glucocorticoids(Kalinyak et al. 1987; Rosewicz et al. 1988). Continual protein synthesis is required forthe maintenance of hepatic CR in rainbow trout (Lee et al. 1992). Consequently, thedecline in concentration of CR in the gills of coho salmon seen this study may be a resultof reduced CR gene transcription by cortisol. The upregulation of CR by GH may alsofunction by regulating the production of CR. GH may exert a direct effect on stimulating145transcription of the CR gene or reduce the inhibitory effect of cortisol. It is not possibleto attribute upregulation of CR by Gil to one of these mechanisms in particular. The lackof a reduction in CR despite a significant increase in plasma cortisol concentrationfollowing treatment with 0.05 g bPL • g body weight4 for one and six weeks (Chapter 4)might indicate that GH opposes the inhibition of CR gene transcription by cortisol. It isalso not possible to rule out the role of additional factors in the regulation of CRconcentration. Despite the greater GH levels observed in the hatchery 2 than hatchery 1fish examined in Chapter 6, CR concentration was significantly lower in this group.Consequently, differences in cortisol or some other regulatory factor between the hatcherygroups must have contributed to this finding.The concentration of CR was found to differ with rearing environment and season.The seasonal changes in gill CR concentration are likely associated with endocrineregulation. Cortisol dynamics associated with stress could account for the differences inB. observed between the groups of fish examined that were reared in differentenvironments. Although cortisol levels in the hatchery fish did not differ from theconcentrations measured in the wild fish during much of the spring, the CR population inthe gills of hatchery fish (Chapter 5) appear to be downregulated in comparison to the wildfish. The differences can not be attributed solely to rearing in a fishculture facility, as cohoreared in net pens in Lois Lake (Chapter 2) had gill CR concentrations similar to thoseseen in the wild fish from the Quinsam River. The rearing conditions in a lake net pen aredifferent from those of an earthen channel. The rearing density is lower in a net pen, thereis a greater opportunity to capture natural food, there is a significant seasonal change inwater temperature, and the water quality is different (less particulate matter). The fishreared in the net pens also exhibited seasonal increases in Na+K±ATPase activity andstrong hypo-osmoregulatory ability (Chapter 2). The seasonal changes in these twoparameters were similar to those seen in the wild fish and much greater than the hatcheryfish (Chapter 5). The fish reared in the earthen channels might experience factors that arespecific to the rearing channels and are not encountered by the wild or Lois Lake coho.There are seasonal increases in mortality in the earthen channel that peak when thebiomass of fish held in the upper site of the hatchery is greatest (D. Ewart, Quinsam River146Salmonid Hatchery, personal communication). When the biomass in the hatchery upper siteis great, there is an increase in the amount of waste products dissolved and suspended inthe water that passes through the earthen channel where the fish examined in Chapters 5,6and 7 were raised. Many of the observed differences that were seen between the hatcheryand wild coho may be a function of water quality. The differences in water quality maythen be the factor that leads to the functional difference in saltwater tolerance exhibitedbetween the hatchery and wild coho smolts (Chapter 7). The findings of this thesis haveimportant implications for optimizing fishculture practices used to enhance natural stocksof salmonids. The mechanism for impairment of hypo-osmoregulatory development ofjuvenile coho reared in the Quinsam River earthen channel is not known. It seems clear,however, that procedures must be employed to minimize factors that stress the fish.Preventing stress in the hatchery environment may prevent downregulation of gillcorticosteroid receptors, and maintain cellular responsiveness of the gills to the hormonalsurges characteristic of smolting such that hatchery coho can survive release into thenatural environment and the transfer to salt water.147LITERATURE CITEDAlexander, J.B. and G.A. Ingram. 1980. A comparison of five of the methods commonlyused to measure protein concentrations in fish sera. J. Fish Biol. 16:115-122.Argue, A.W., B. Hillaby and C.D. Shepard. 1986. Distribution, timing, change in size,and stomach contents of juvenile chinook and coho salmon caught in Cowichanestuary and bay, 1973, 1975, 1976. Can. Tech. Rep. Fish. Aquat. Sci. No. 1431.Avella, M., C.B. Schreck and P. Prunet. 1991. Plasma prolactin and cortisolconcentrations of stressed coho salmon, Oncorhynchus kisutch, in fresh water or saltwater. Gen. Comp. Endocrin. 81: 21-27.AveIla, M., G. Young, P. Prunet and C.B. Schreck. 1990. Plasma prolactin and cortisolconcentrations during salinity challenges of coho salmon (Oncorhynchus kisutch) atsmolt and post-smolt stages. Aquaculture 91:359-372.Ball, J.N., and E.F. Hawkins. 1976. Adrenocortical (interrenal) responses tohypophysectomy and adenohypohyseal hormones in the teleost Poecilia latipinna.Gen. Comp. Endocrin. 28:59-70.Banks, J.L., W.G. Taylor and S.L. Leek. 1977. Carrying capacity recommendations forOlympia area national fish hatcheries. Unpublished Manuscript.Barton, B.A. and G.K. Iwama. 1991. Physiological changes in fish from stress inaquaculture with emphasis on the response and effects of corticosteroids. AnnualRev. of Fish Diseases. 1:3-26.Barton, BA., C.B. Schreck and L.D. Barton. 1987. Effects of chronic cortisoladministration and daily acute stress on growth, physiological conditions and stressresponses in juvenile rainbow trout. Dis. Aquat. Org. 2:173-185.Barton, B.A., C.B. Schreck and L.A. Sigismondi. 1986. Multiple acute disturbancesevoke cumulative physiological stress responses in juvenile chinook salmon. Trans.Am. Fish. Soc. 115:245-251.Barton, B.A., C.B. Schreck, R.D. Ewing, A.R. Hemmingsen and R. Patin°. 1985.Changes in plasma cortisol during stress and smoltification in coho salmon,Oncorhynchus kisutch. Gen. Comp. Endocrin. 59:468-471.Bennett, J.P. 1978. Methods in Binding Studies. In: Neurotransmitter receptor binding.H.I. Yamamura, S.J. Enna and M.J. Kuhar, eds. Raven Press, New York. pp. 57-90.148Blackburn, J. and W.C. Clarke. 1987. Revised procedure for the 24 hour seawaterchallenge test to measure seawater adaptability of juvenile salmonids. Can. Tech.Report Fish. Aquat. Sci. No. 1515.Blackmun, G.J., B.V. Lukyn, W.E. McLean and D. Ewart. 1985. Quinsam WatershedStudy: 1983. Canadian Manuscript Report of Fisheries and Aquatic Sciences No.1832. 65pp.Bjornsson, B.Th., G. Young, R.J. Lin, L.J. Deftos and H.A. Bern. 1989. Smoltificationand seawater adaptation in coho salmon (Oncorhynchus kisutch): plasma calciumregulation, osmoregulation, and cakitonin. Gen. Comp. Endocr. 74:346-354.Bjornsson, B.Th., T. Ogasawara, T. Hirano, J.P. Bolton and H.A. Bern. 1988. Elevatedgrowth hormone levels in stunted Atlantic salmon, Salmo salar. Aquaculture73:275-281.Bodine, P.V. and G. Litwack. 1990. The glucocorticoid receptor and its endogenousregulators. Receptor 1:83-119.Bolton, J.P., N.L. Collie, H. Kawauchi and T. Hirano. 1987a. Osmoregulatory actionsof growth hormone in rainbow trout (Salmo gairdneri). J. Endocr. 112:63-68.Bolton, J.P., G. Young, R.S. Nishioka, T. Hirano and H.A. Bern. 1987b. Plasmagrowth hormone levels in normal and stunted yearling coho salmon, Oncorhynchuskisutch. J. Exp. Zool. 242:379-382.Bradford, M.M. 1976. A rapid and sensitive method for the quantiation of microgramquantities of protein using the principle of protein dye binding. Anal. Biochem.72:248-252.Brauner, C.J., G.K. Iwama and D.J. Randall. 1993. The effect of short duration seawaterexposure on the swimming performance of hatchery and wild reared coho salmonsmolts (Oncorhynchus kisutch). Unpublished manuscript.Brauner, C.J., J.M. Shrimpton and D.J. Randall. 1992. Effect of short-duration seawaterexposure on plasma ion concentrations and swimming performance of coho salmon(Oncorhynchus kisutch) parr. Can. J. Fish. Aquat. Sci. 49:2399-2405.Brown, S.B., J.G. Eales and T.J. Hara. 1986. A protocol for estimation of cortisolplasma clearance in acid-exposed rainbow trout (Salmo gairdneri). Gen. Comp.Endocrinol. 62:493-502.149Bry, C. 1982. Daily variations in plasma cortisol levels of individual female rainbow troutSalmo gairdneri: evidence for a post-feeding peak in well-adapted fish. Gen. Comp.Endocrin. 48:462-468.Chakraborti, P.K., M. Weisbart and A. Chakraborti. 1987. The presence ofcorticosteroid receptor activity in the gills of the Brook Trout, Salvelinus fontinaaGen. Comp. Endocrin. 66:323-332.Chester Jones, I., D.K.O. Chan, I.W. Henderson and J.N. Ball. 1969. Theadrenocortical steroids, adrenocorticotropin and the Corpuscles of Stannius. In:Fish Physiology. Vol 2. W.S. Hoar and D.J. Randall, eds. Academic Press, NewYork. pp.322-376.Cidlowski, J.A. and N.B. Cidlowski. 1981. Regulation of glucocorticoid receptors byglucocorticoids in cultured HeLa S3 cells. Endocrinol. 109:1975-1982.Claire, M., M-E. Oblin, J-L. Steimer, H. Nakane, J. Misumi, A. Michaud and P. Corvol.1981. Effect of adrenalectomy and aldosterone on the modulation ofmineralocorticoid receptors in rat kidney. J. Biol. Chem. 256:142-147.Clark, C.R. 1985. Intracellular localisation of steroid receptors. In: Interaction of SteroidHormone Receptors with DNA. M. Sluyser, ed. Ellis Horwood Ltd., Chichester(England). pp.7-56.Clarke, W.C., H.A. Bern, C.H. Li and D.C. Cohen. 1973. Somatotropic and sodium-retaining effects of human growth hormone and placental lactogen in lowervertebrates. Endocrinology 93:960-964.Clarke, W.C., S.W. Farmer and K.M. Hartwell. 1977. Effect of teleost pituitary growthhormone on growth of Tilapia mossambica and on growth and seawater adaptationof sockeye salmon (Oncorhynchus nerka). Gen. Comp. Endocrin. 33:174-178.Clarke, W.C., J.E. Shelbourn and J.R. Brett. 1981. Effect of artificial photoperiod cycles,temperature, and salinity on growth and smolting in underyearling coho(Oncorhynchus kisutch), chinook (0. tshawytscha), and sockeye (0. nerka) salmon.Aquaculture 22:105-116.Clarke, W.C. 1982. Evaluation of the seawater challenge test an an index of marinesurvival. Aquaculture 28:177-183.Collie, N.L., J.P. Bolton, H. Kawauchi and T. Hirano. 1989. Survival of salmonids inseawater and the time-frame of growth hormone action. Fish Phys. Bioc. 7:315-321.150Conte, F.P., H.H. Wagner, J. Fessler and C. Gnose. 1966. Development of osmotic andionic regulation in juvenile coho salmon Oncorhynchus kisutch. Comp. Biochem.Physiol. 18:1-15.Cross, C.L., L. Lapi and E.A. Perry. 1991. Production of chinook and coho salmon fromBritish Columbia hatcheries, 1971 through 1989. Can. Tech. Report of Fish. Aquat.Sci. No. 1816.Danielsen, M. and M.R. Stallcup. 1984. Down-regulation of glucocorticoid receptors inmouse lymphoma cell variants. Mol. Cell. Biol. 4:449-453.Danzmann, R.G., M.M. Ferguson and F.W. Allendorf. 1989. Genetic variability andcomponents of fitness in hatchery strains of rainbow trout. J. Fish Biol.35(Supplement A):313-319.Delahunty, G., J. Okese, M. Prack, M.J. Vodicnik, C.B. Schreck and V. de Vlaming.1978. Diurnal variation in the physiology of the goldfish, Carassius auratus. J.Interdiscipl. Cycle Res. 9:73-88.De Nicola, A.F., D.F. Moses, S. Gonzalez and E. Orti. 1989. Adrenocorticoid actionin the spinal cord: Some unique molecular properties of glucocorticoid receptors.Cell. Mal. Neurobiol. 9:179-192.Devlin, R.H., J.C. Byatt, E. McLean, T.Y. Yesaki and E.M. Donaldson. 1993. Bovineplacental lactogen is a potent stimulator of growth in coho salmon. UnpublishedManuscript.Donaldson, E.M. 1981. Pituitary-interrenal axis as an indicator of stress in fish. In Stressand Fish (A.D. Pickering, ed.), pp 11-47. Academic Press, London and New York.Eddy, F.B. 1981. Effects of stress on osmotic and ionic regulation in fish. In: Stress andFish. A.D. Pickering, ed. Academic Press, London. pp. 78-102.Eib, D.W. and K.L. Hossner. 1985. The effects of cortisol and actinomycin D injectionson chloride cells and branchial NaW-ATPase in rainbow trout (Salmo gairdneri).Gen. Comp. Endocrin. 59:449-452.Eisen, L.P., M.S. Elsasser and J.M. Harmon. 1988. Positive regulation of theglucocorticoid receptor in human T-cells sensitive to the cytolytic effects ofglucocorticoids. J. Biol. Chem. 263:12044-12048.Ejike, C. and C.B. Schreck. 1980. Stress and social hierarchy rank in coho salmon.Trans. Am. Fish. Soc. 109:423-426.151Enright, W.J., L.T. Chapin, W.M. Moseley, S.A. Zinn, M.D. Kamdar, L.F. Krabill andH.A. Tucker. 1989. Effects of infusions of various doses of bovine growthhormone-releasing factor on blood hormones and metabolites in lactating Holsteincows. J. Endocr. 122:671-679.Epstein, F.H., P. Silva and G. Kormanik. 1980. Role of Na-K-ATPase in chloride cellfunction. Am. J. Physiol. 238:R246-R250.Ewing, R.D., S.L. Johnson, H.J. Pribble and J.A. Lichatowich. 1979. Temperature andphotoperiod effects on gill (NeK4)-ATPase activity in chinook salmon(Oncorhynchus tshawytscha). J. Fish. Res. Bd. Canada. 36:1347-1353.Fagerlund, U.H.M., J.R. McBride, B.S. Dosanjh, E.T. Stone, F.K. Sandercock. 1983.Influence of culture density on juvenile coho salmon production and ocean survival.Can. Tech. Rep. Fish. Aquat. Sci. No. 1229. 27pp.Fisher, J.P. and W.G. Pearcy. 1988. Growth of juvenile coho salmon (Oncorhynchuskisutch) off Oregon and Washington, USA, in years of differing coastal upwelling.Can. J. Fish. Aquat. Sci. 45:1036-1044.Foskett, J.K., T.E. Machen and H.A. Bern. 1982. Chloride secretion and conductanceof teleost opercular membrane: effects of prolactin. Am. J. Physiol. 242:R380-R389.Fryer, J.N. and H.A. Bern. 1979. Growth hormone binding to tissues of normal andstunted juvenile coho salmon, Oncorhynchus kisutch. J. Fish Biol. 15:527-533.Fryer, J.N. and R.E. Peter. 1977. Hypothalamic control of ACTH secretion in goldfish.III. Hypothalamic cortisol implant studies. Gen. Comp. Endocr. 33:215-225.Gause, I. and S. Eden. 1986. Induction of growth hormone (GH) receptors in adipocytesof hypophysectomized rates by GH. Endocrinology 118:119-124.Gendreau, P., J-G. Lehoux, S. Belisle and D. Bellabarba. 1987. Glucocorticoidreceptors in chick embryos: properties and ontogeny of the nuclear receptor. Gen.Comp. Endocr. 67:58-66.Gorbman, A., W.W. Dickhoff, J.L. Mighell, E.F. Prentice and F.W. Waknitz. 1982.Morphological indices of developmental progress in the parr-smolt coho salmon,Oncorhynchus kisutch. Aquaculture 28:1-19.Handwerger, S. and A. Brar. 1992. Placental lactogen, placental growth hormone, anddecidual prolactin. Seminars Reprod. Endocrin. 10:106-115.152Harache, Y., G. Boeuf and P. Lasserre. 1980. Osmotic adaptation of Oncorhynchuskisutch Walbaum. III. Survival and growth of juvenile coho salmon transferred tosea water at varous times of the year. Aquaculture 19:253-273.Healey, M.C. 1982. Juvenile Pacific salmon in estuaries: the life support system. InEstuarine Comparisons. (V.S. Kennedy, ed) Academic Press, Toronto. pp.315-341.Heath, D.D. 1993. Genetic, environmental, and physiological factors involved in theprecocious sexual maturation of chinook salmon (Oncorhynchus tshawytscha). PhDThesis, Department of Animal Science, University of British Columbia.Henderson, I.W., M.N. Sa'di and G. Hargreaves. 1974. Studies on the production andmetabolic clearance rates of cortisol in the european eel, Anguilla anguilla (L.). J.Steroid Bioc. 5:701-707.Higgs, D.A., B.S. Dosanjh, M.D. Plotnikoff, J.R. Markert, D. Lawseth, J.R. McBrideand J.T. Buckley. 1992. Influence of dietary protein to lipid ratio and lipidcomposition on the performance and marine survival of hatchery reared chinooksalmon (Oncorhynchus tshawytscha). Bull. Aquacul. Assoc. Canada 92:46-48.Hirano, T. 1980. Prolactin and osmoregulation. In: Endocrinology. I.A. Cummings, ed.Australian Acad. Sci. pp.186-189.Hirota, K., T. Hirota, Y. Sanno and T. Tanaka. 1985. Physicochemical characterizationof a new glucocorticoid receptor. Biochim. Biophys. Acta 843:171-179.Hoar, W.S. 1988. The physiology of smolting salmonids. In: Fish Physiology Vol. 11B.W.S. Hoar and D.J. Randall, eds. Academic Press, New York. pp.275-344.Junker, K. 1983. Glucocorticoid receptors of human mononuclear leukocytes in vitro. J.Clinical. Endocrinol. Metabol. 57:506-512.Kalinyak, J.E., RI. Dorin, A.R. Hoffman and A.J. Perlman. 1987. Tissue-specificregulation of glucocorticoid receptor mRNA by dexamethasone. J. Biol. Chem.262:10441-10444.Komourdjian, M.P., R.L. Saunders and J.C. Fenwick. 1976. The effect of porcinesomatotropin on growth, and survival in seawater of Atlantic salmon (Salmo salar)parr. Can. J. Zool. 54:531-535.Langdon, J.S. and J.E. Thorpe. 1985. The ontogeny of smoltification: developmentalpatterns of gill Na+/K+-ATPase, SDH, and chloride cells in juvenile Atlantic salmon,Salmo salar L. Aquaculture 45:83-95.153Lange, C.B. and W. Hanke 1988. Corticosteroid receptors in liver cytosol of the clawedtoad, Xenopus laevis: daily and seasonal variations. Gen. Comp. Endocr. 71:141-152.Langhorne, P. 1986. Somatostatin stimulates ACTH release in brown trout (Salmo truttaL.). Gen. Comp. Endocrin. 61:71-75.Lasserre, P., G. Boeuf and Y. Harache. 1978. Osmotic adaptation of Oncorhynchuskisutch Walbaum. I. Seasonal variation of gill Na+-K+ ATPase activity in cohosalmon, 0+-age and yearling, reared in fresh water. Aquaculture 14:365-382.Lee, P.C., M. Goodrich, M. Struve, H.I. Yoon and D. Weber. 1992. Liver and brainglucocorticoid receptor in rainbow trout, Oncorhynchus mykiss: Downregulation bydexamethasone. Gen. Comp. Endorin. 87:222-231.Leray, C., D.A. Colin and A. Florentz. 1981. Time course of osmotic adaptation and gillenergetics of rainbow trout (Salmo gairdneri R.) following abrupt changes in externalsalinity. J. Comp. Physiol. B. 144:175-181.Levings, C.D., C.D. McAllister and B.D. Chang. 1986. Differential use of the CampbellRiver estuary, British Columbia, by wild and hatchery-reared juvenile chinooksalmon (Onchorhynchus tshawytscha). Can. J. Fish. Aquat. Sci. 43:1386-1397.Lin, H. and D. Randall. 1991. Evidence for the presence of an electrogenic proton pumpon the trout gill epithelium. J. Exp. Biol. 161:119-134.Macdonald, J.S., I.K. Birtwell and G.M. Kruzynski. 1987. Food and habitat utilizationby juvenile sahnonids in the Campbell River estuary. Can. J. Fish. Aquat. Sci.44:1233-1246.Macdonald, J.S., C.D. Levings, C.D. McAllister, U.H.M. Fagerlund and J.R. McBride.1988. A field experiment to test the importance of estuaries for chinook salmon(Oncorhynchus tshawytscha) survival: short-term results. Can. J. Fish. Aquat. Sci.45:1366-1377.Madsen, S.S. 1990a. Cortisol treatment improves the development of hypoosmoregulatorymechanisms in the euiyhaline rainbow trout, Salmo gairdneri. Fish Phys. Bioc. 8:45-52.Madsen, S.S. 1990b. Enhanced hypoosmoregulatroy response to growth hormone aftercortisol treatment in immature rainbow trout, Salmo gairdneri. Fish Phys. Bioc.8:271-279.154Madsen, S.S. 1990c. The role of cortisol and growth hormone in seawater adaptation anddevelopment of hypoosmoregulatroy mechanisms in sea trout parr (Salmo truttatrutta). Gen. Comp. Endocrin. 79:1-11.Madsen, S.S. and B. Korsgaard. 1991. Opposite effects of 1713-Estradiol and combinedgrowth hormone-cortisol treatment on hypo-osmoregulatory performance in seatrout presmolts, Salmo trutta. Gen. Comp. Endocrin. 83:276-282.Matic, G., D. Trajkovic, M. Susa, S. Damjanovic and J. Petrovic. 1989. In vitroevidence for modification of rat liver glucocorticoid receptor binding properties andtransformation by hyperthermia. J. Steroid Biochem. 32:263-270.Mayer, N., J. Maetz, D.K.O. Chan, M. Forster and I. Chester Jones. 1967. Cortisol,a sodium excreting factor in the eel (Anguilla anguilla L.) adapted to sea water.Nature. 214:1118-1120.Maule, A.G. and C.B. Schreck. 1990. Glucocorticoid receptors in leukocytes and gill ofjuvenile coho salmon (Oncorhychus kisutch). Gen. Comp. Endocrin. 77:448-455.Maule, A.G. and C.B. Schreck. 1991. Stress and cortisol treatment changed affinity andnumber of glucocorticoid receptors in leukocytes and gill of coho salmon. Gen.Comp. Endocrin. 84:83-93.Maule, A.G., C.B. Schreck and S.L. Kaattari. 1987. Changes in the immune system ofcoho salmon (Oncorhynchus kisutch) during the parr-to-smolt transformation andafter implantation with cortisol. Can. J. Fish. Aquat. Sci. 44:161-166.Mazeaud, M.M., F. Mazeaud and E.M. Donaldson. 1977. Primary and secondary effectof stress in fish: some new data with a general review. Trans. Am. Fish. Soc.106:201-212.McCormick, S.D. and B.Th. Bjornsson. 1992. Physiological and hormonal differencesamong naturally-reared Atlantic salmon migrants and non-migrants, and hatcherysmolts. In: Proceedings from the 4th Interantional Smolt Workshop, St. Andrews,New Brunswick, Canada. Oct 19-23. (Abstract).McCormick, S.D., W.W. Dickhoff, J. Duston, R.S. Nishioka and H.A. Bern. 1991a.Developmental differences in the responsiveness of gill Na+,K+,-ATPase to cortisolin sahnonids. Gen. Comp. Endocr. 84:308-317.McCormick, S.D., T. Sakamoto, S. Hasegawa and T. Hirano. 1991b. Osmoregulatoryactions of insulin-like growth factor-I in rainbow trout (Oncorhynchus mykiss). J.Endocrinol. 130:87-92.155McCormick, S.D. 1990. Cortisol directly stimulates differentiation of chloride cells intilapia opercular membrane. Am. J. Physiol. 259:R857-R863.McCormick, S.D. and H.A. Bern. 1989. In vitro stimulation of Na÷-K+-ATPase activityand ouabain binding by cortisol in coho salmon gill. Am. J. Physiol. 256:R707-715.McCormick, S.D., R.L. Saunders and A.D. MacIntyre. 1989a. Mitochondrial enzyme andNa+,KtATPase activity, and ion regulation during parr-smolt transformation ofAtlantic salmon (Salmo salar). Fish. Phys. Biochem. 6:231-241.McCormick, S.D., R.S. Nishioka and H.A. Bern. 1989b. In vitro hormonal stimulationof Na+,K+-ATPase activity and ouabain binding in gill tissue of coho salmon.Aquaculture 82:379 (Abstract).McCormick, S.D. and R.L. Saunders. 1987. Preparatory physiological adaptations formarine life of salmonids: osmoregulation, growth, and metabolism. Am. Fish. Soc.Symp. 1:211-229.McIntyre, W.R. and H.H. Samuels. 1985. Triamcinolone Acetonide regulatesglucocorticoid-receptor levels by decreasing half-life of the activated nuclear-receptorform. J. Biol. Chem. 260:418-427.Milne, R.S. and J.F. Leatherland. 1978. Effect of ovine TSH, thiourea, ovine prolactinand bovine growth hormone on plasma thyroxine and tri-iodo-thyronine levels inrainbow trout, Salmo gairdneri. J. Comp. Physiol. 124:105-110.Miwa, S. and Y. Inui. 1985. Effects of L-thyroxine and ovine growth hormone onsmoltification of amago salmon (Oncorhynchus rhodorus). Gen. Comp. Endo.58:436-442.Mundie, J.H., D.E. Mounce and K.S. Simpson. 1990. Semi-natural rearing of cohosalmon, Oncorhynchus kisutch (Walbaum), smolts, with an assessment of survival tothe catch and escapement. Aqua. Fish. Manag. 21:327-345.Naess, 0., E. Huag and K. Gautvik. 1980. Effects of glucocorticosteroids on prolactinand growth hormone production and characterization of the intracellular hormonereceptors in rat pituitary tumour cells. Acta Endocrin. 95:319-327.Nemoto, T., Y. Ohara-Nemoto, M. Denis and J.A. Gustafsson. 1990. The transformedglucocorticoid receptor has a lower steroid-binding affinity than the nontransformedreceptor. Biochemistry 29:1880-1886.156Nicoll, C.S., J.F. Tarpey, G.L. Mayer and S.M Russell. 1986. Similarities anddifferences among prolactins and growth hormones and their receptors. Amer.Zool. 26:965-983.Nishioka, R.S., H.A. Bern, K.V. Lai, Y. Nagahama and E.G. Grau. 1982. Changes inthe endocrine organs of coho salmon during normal and abnormal smoltification.An electron microscope study. Aquaculture 28:21-38.Normand, M. and C. Fortier. 1970. Numerical versus analytical integration of hormonaldisappearance data. Can. J. Physiol. Pharmac. 48:274-281.Ogata, H and T. Murai. 1989. Effects of dietary fatty acid composition on growth andsmolting of underyearling masu salmon, Oncorhynchus masou. Aquaculture. 82:181-189.Otto, R.G. and J.E. McInerney. 1970. Development of salinity preference in pre-smoltcoho salmon, Oncorhynchus kisutch. J. Fish. Res. Bd. Canada 27:793-800.Otto, R.G. 1971. Effects of salinity on the survival and growth of pre-smolt coho salmon(Oncorhynchus kisutch). J. Fish. Res. Bd. Canada 28:343-349.Patino, R., C.B. Schreck and J.M. Redding. 1985. Clearance of plasma corticosteroidsduring smoltification of coho salmon, Oncorhynchus kisutch. Comp. Biochem.Physiol. A. 82:531-535.Patino, R., C.B. Schreck, J.L. Banks and W.S. Zaugg. 1986. Effects of rearingconditions on the developmental physiology of smolting coho salmon. Trans. Am.Fish. Soc. 115:828-837.Perry, S.F. and P. Laurent. 1989. Adaptational responses of rainbow trout to loweredexternal NaC1 concentration: contribution of the branchial chloride cell. J. exp. Biol.147:147-168.Perry, S.F. and D.J. Randall. 1981. Effects of amilmide and SITS on branchial ion fluxesin rainbow trout, Salmo gairdneri. J. Exp. Zool. 215:225-228.Peters, G., H. Delventhal and H. Klinger. 1980. Physiological and morphological effectsof social stress in the eel, (Anguilla anguilla L.). Arch. FischWiss. 30:157-180.Pickering, A.D., T.G. Pottinger, J.P. Sumpter, J.F. Carragher and P.Y. Le Bail. 1991.Effects of acute and chronic stress on the levels of circulating growth hormone inthe rainbow trout, Oncorhynchus mykiss. Gen. Comp. Endocrin. 83:86-93.157Pickering, A.D. and T.G. Pottinger. 1983. Seasonal and diel changes in plasma cortisollevels of the brown trout, Salmo trutta L. Gen. Comp. Endocrinol. 49:232-239.Pottinger, T.G. 1991. The effect of stress and exogenous cortisol on receptor-like bindingof cortisol in the liver of rainbow trout, Oncorhynchus mykiss. Gen. Comp.Endocrin. 78:194-203.Prunet, P., G. Boeuf, J.P. Bolton and G. Young. 1989. Smoltification and seawateradaptation in Atlantic salmon (Salmo saki.): plasma prolactin, growth hormone, andthyroid hormones. Gen. Comp. Endocrin. 74:355-364.Rance, T.A., B.I. Baker and G. Webley. 1982. Variations in plasma cortisolconcentrations over a 24-hour period in the rainbow trout Salmo gairdneri. Gen.Comp. Endocrin. 48:269-274.Randall, D. 1982. The control of respiration and circulation in fish during exercise andhypoxia. J. Exp. Biol. 100:275-288.Redding, J.M., R. Patino and C.B. Schreck. 1984a. Cortisol and its effects on plasmathyroid hormone and electrolyte concentrations in fresh water and during seawateracclimation in yearling coho salmon, Oncorhynchus kisutch. Gen. Comp. Endocrin.56:146-155.Redding, J.M., R. Patino and C.B. Schreck. 1984b. Clearance of corticosteroids inyearling coho salmon, Oncorhynchus kisutch, in freshwater and sea-water and afterstress. Gen. Comp. Endocrin. 54:433-443.Redding, J.M., C.B. Schreck and F.H. Everest. 1987. Physiological effects on cohosalmon and steelhead of exposure to suspended solids. Trans. Am. Fish. Soc.116:737-744.Richman III, N.H. and W.S. Zaugg. 1987. Effects of cortisol and growth hormone onosmoregulation in pre- and desmoltified coho salmon (Oncorhynchus kisutch). Gen.Comp. Endocrinol. 65:189-198.Rosewicz, S., A.R. McDonald, B.A. Maddux, I.D. Goldfine, R.L. Miesfeld and C.D.Logsdon. 1988. Mechanism of glucocorticoid receptor down-regulation byglucocorticoids. J. Biol. Chem. 263:2581-2584.Rydevik, M., B. Borg, C. Haux, H. Kawauchi and B.Th. Bjornsson. 1990. Plasmagrowth hormone levels increase during seawater exposure of sexually matureAtlantic salmon parr (Salmo salar L.). Gen. Comp. Endocr. 80:9-15.158Sakamoto, T., T. Ogasawara and T. Hirano. 1990. Growth hormone kinetics duringadaptation to a hyperosmotic environment in rainbow trout. J. Comp. Physiol. B160:1-6.Sakamoto, T. and T. Hirano. 1991. Growth hormone receptors in the liver andosmoregulatory organs of rainbow trout: characterization and dynamics duringadaptation to seawater. J. Endocrinol. 130:425-433.Sakamoto, T., M. Iwata and T. Hirano. 1991. Kinetic studies of growth hormone andprolactin during adaptation of coho salmon, Oncorhynchus kisutch, to differentsalinities. Gen. Comp. Endocr. 82:184-191.Salonius, K. 1991. Effects of early rearing history on endocrine and immune function injuvenile Pacific salmonids. MSc Thesis, Department of Animal Science, Universityof British Columbia.Sanchez, E.R., D.O. Toft, M.J. Schlesinger and W.B. Pratt. 1985. Evidence that the 90-kDa phosphoprotein associated with the untransformed L-cell glucocorticoidreceptor is a murine heat shock protein. J. Biol. Chem. 260:12398-12401.Sapolsky, R.M., L.C. Krey and B.S. McEwen. 1984. Stress down-regulates corticosteronereceptors in a site-specific manner in the brain. Endociin. 114:287-292.Saunders, R.L., J.L. Specker and M.P. Komourdjian. 1982. Effects of photoperiod ongrowth and smolting in juvenile Atlantic salmon (Salmo salar). Aquaculture 82:103-117.Scatchard, G. 1949. The attraction of protein for small molecules and ions. Ann N.Y.Acad. Sci. 51:661-672.Schreck, C.B. 1981. Stress and compensation in teleostean fishes: response to social andphysical factors. In: Stress and Fish. A.D. Pickering, ed. Academic Press, London.pp. 296-321.Schreck, C.B., R. Patino, C.K. Pring, J.R. Winton and J.E. Holway. 1985. Effects ofrearing density on indices of smoltification and performance of coho salmon,Oncorhynchus kisutch. Aquaculture 45:345-358.Sillence, M.N. and T.D. Etherton. 1989. Chronic effects of recombinant porcine growthhormone on adrenal weight and activity in pigs. J. Anim Sci. 67:1740-1743.Slater, E.P., J.D. Baxter and N.L. Eberhardt. 1986. Evolution of the growth hormonegene family. Amer. Zool. 26:939-949.159Smart, G.R. 1981. Aspects of water quality producing stress in intensive fish culture. In:Stress and Fish. A.D. Pickering, ed. Academic Press, London. pp. 103-119.Specker, J. and C.B. Schreck. 1982. Changes in plasma corticosteroids during smolt-ification of coho salmon, Oncorhynchus kisutch. Gen. Comp. Endocrin. 46:53-58.Strange, R.J., C.B. Schreck and R.D. Ewing. 1978. Cortisol concentrations in confinedjuvenile chinook salmon (Oncorhynchus tshawytscha). Trans. Am. Fish. Soc.107:812-819.Suarez, R.K., G.S. Brown and P.W. Hochachka. 1986. Metabolic sources of energy forhummingbird flight. Am. J. Pysiol. 251:R537-R542.Sumpter, J.P., P.Y. Le Bail, A.D. Pickering, T.G. Pottinger and J.F. Carragher. 1991.The effect of starvation on growth hormone concentrations of rainbow trout,Oncorhynchus mykiss. Gen. Comp. Endocr. 83:94-102.Svec, F. 1988. The bipotency of dexamethasone at causing hepatic glucocorticoid receptordown-regulation in the intact mouse. Biochim Biophys. Acta 970:90-95.Svec, F. and M. Rudis. 1981. Glucocorticoids regulate the glucocorticoid receptor in theAtT-20 cell. J. Biol. Chem. 256:5984-5987.Sweeting, R.M., G.F. Wagner and B.A. McKeown. 1985. Changes in plasma glucose,amino acid nitrogen and growth hormone during smoltification and seawateradaptation in coho salmon, Oncorhynchus kisutch. Aquaculture 45:185-197.Sweeting, R.M. and B.A. McKeown. 1987. Growth hormone and seawater adaptation incoho salmon, Oncorhynchus kisutch. Comp. Biochem. Physiol. 88A:147-151.Thorpe, J.E. 1977. Bimodal distribution of length of juvenile Atlantic salmon (Salmo salarL.) under artificial rearing conditions. J. Fish. Biol. 11:175-184.Thorpe, J.E., R.I.G. Morgan, E.M. Ottaway and M.S. Miles. 1980. Time of divergenceof growth groups between potential 1+ and 2+ smolts among sibling Atlanticsalmon. J. Fish Biol. 17:13-21.Thorpe, J.E., C. Talbot and C. Villarreal. 1982. Bimodality of growth and smolting inAtlantic Salmon, Salmo salar L. Aquaculture 28:123-132.Thorpe, J.E., M.G. McConway, M.S. Miles and J.S. Muir. 1987. Diel and seasonalchanges in resting plasma cortisol levels in juvenile Atlantic salmon, Salmo salar L.Gen. Comp. Endocrin. 65:19-22.160Tschaplinski, P.J. 1987. The use of estuaries as rearing habitats by juvenile coho salmon.In: Proceedings of the workshop: applying 15 years of Carnation Creek results.T.W. Chamberlain (ed). Pacific Biological Station, Nanaimo, BC. pp.123-142.Van Praag, D., S.J. Farber, E. Minkin and N. Primor. 1987. Production of eicosanoidsby the killifish gills and opercular epithelia and their effect on active transport ofions. Gen. Comp. Endo. 67:50-57.Varnavsky, V.S., T. Sakamoto and T. Hirano. 1992. Stunting of wild coho salmon(Oncorhynchus kisutch) in seawater: patterns of plasma thyroid hormones, cortisol,and growth hormone. Can. J. Fish. Aquat. Sci. 49:458-461.Wagner, G.F. and B.A. McKeown. 1986. Development of a salmon growth hormoneradioimmunoassay. Gen. Comp. Endocrin. 62:452-458.Wahle, R.J. and W.S. Zaugg. 1982. Adult coho salmon recoveries and their Na+-K+ATPase activity at release. Marine Fish. Rev. 44:11-13.Walters, C. 1988. Mixed-stock fisheries and the sustainability of enhancement productionfor chinook and coho salmon. In Salmon Production, Management, and Allocation.Biological, Economic, and Policy Issues, pp. 109-115. W.J. McNeil, ed. OregonState University Press, Corvalis, Oregon.Waples, R.S. 1991. Genetic interactions between hatchery and wild salmonids: lessonsfrom the Pacific Northwest. Can. J. Fish. Aquat. Sci. 48(Suppl. 1):124-133.Wedemeyer, G.A., R.L. Saunders and W.C. Clarke. 1980. Environmental factorsaffecting smoltification and early marine survival of anadromous salmonids. Mar.Fish. Rev. 42:1-14.Wedemeyer, G.A. and D.J. McLeay. 1981. Methods for determining the tolerance offishes to environmental stressors. In: Stress and Fish. A.D. Pickering, ed.Academic Press, London. pp.248-275.Weisbart, M., P.K. Chakraborti, G. Gallivan and J.G. Eales. 1987. Dynamics of cortisolreceptor activity in the gills of brook trout, Salvelinus fontinalis, during seawateradaptation. Gen. Comp. Endocrin. 68:440-448.Yada, T., K. Takahashi and T. Hirano. 1991. Seasonal changes in seawater adaptabilityand plasma levels of prolactin and growth hormone in landlocked sockeye salmon(Oncorhynchus nerka) and amago salmon (0. rhodurus). Gen. Comp. Endocr.82:33-44.161Yamauchi, K., R.S. Nishioka, G. Young, T. Ogasawara, T. Hirano and H.A. Bern.1991. Osmoregulation and circulating growth hormone and prolactin levels inhypophysectomized coho salmon (Oncorhynchus Icisutch) after transfer to fresh waterand seawater. Aquaculture 92:33-42.Young, G. 1988. Enhanced response of the interrenal of coho salmon (Oncorhynchuskisutch) to ACTH after growth hormone treatment in vivo and in vitro. Gen. Comp.Endocrin. 71:85-92.Young, G., B.T. Bjornsson, P. Prunet, R.J. Lin and H.A. Bern. 1989a. Smoltificationand seawater adaptation in coho salmon (Oncorhynchus kisutch): plasma prolactin,growth hormone, thyroid hormones, and cortisol. Gen. Comp. Endocrin. 74:335-345.Young, G., P. Prunet, T. Ogasawara, T. Hirano and H.A. Bern. 1989b. Growthretardation (stunting) in coho salmon: plasma hormone levels in stunts in seawaterand after transfer to fresh water. Aquaculture. 82:269-278.Zaugg, W.S. 1989. Migratory behaviour of underyearling Oncorhynchus tshawytscha andsurvival to adulthood as related to prerelease gill (Na+-K+)-ATPase development.Aquaculture 82:339-353.Zaugg, W.S. 1982. A simplified preparation for adenosine triphosphatase determinationin gill tissue. Can. J. Fish. Aquat. Sci. 39:215-217.Zaugg, W.S. and C.V.W. Mahnken. 1991. The importance of smolt development tosuccessful marine ranching of Pacific salmon. NOAA Tech. Rep. NMFS 102:89-97.Zaugg, W.S. and L.R. McLain. 1970. Adenosinetriphosphatase activity in gills ofsalmonids: seasonal variations and salt water influence in coho salmon Oncorhynchuskisutch. Comp. Biochem. Physiol. 35:587-596.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items