Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The fast-start and sprinting ability, and the effects of growth hormone (GH) upregulation on the muscle… Dimoulas, Peter Michael 2009

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

Item Metadata

Download

Media
24-ubc_2009_spring_dimoulas_peter.pdf [ 1.73MB ]
Metadata
JSON: 24-1.0067153.json
JSON-LD: 24-1.0067153-ld.json
RDF/XML (Pretty): 24-1.0067153-rdf.xml
RDF/JSON: 24-1.0067153-rdf.json
Turtle: 24-1.0067153-turtle.txt
N-Triples: 24-1.0067153-rdf-ntriples.txt
Original Record: 24-1.0067153-source.json
Full Text
24-1.0067153-fulltext.txt
Citation
24-1.0067153.ris

Full Text

THE FAST-START AND SPRINTING ABILITY, AND THE EFFECTS OF GROWTH HORMONE (GH) UPREGULATION ON MUSCLE  FUNCTIONING OF GH-TRANSGENIC COHO SALMON. BY  PETER MICHAEL DIMOULAS B.SC., NORTHEASTERN UNIVERSITY, 2005  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE IN  THE FACULTY OF GRADUATE STUDIES (ZOOLOGY)  THE UNIVERSITY OF BRITISH COLUMBIA (VANCOUVER)  APRIL 2009   PETER MICHAEL DIMOULAS, 2009  ABSTRACT If GH-transgenic coho salmon escaped into the natural environment would their performance during predator-prey encounters or spawning migrations be different from their wild conspecifics? We examined fast-start and sprinting performance to infer their prospective ability to evade predatory strikes and chasing predators, to chase prey, and to complete spawning migrations. Similarly, we addressed the effects of GH upregulation on muscle functioning. Fast-starts are rapid escape events, and are a summation of short-term behavior, white muscle intrinsic properties, musculoskeletal linking, and body shape. Sprinting additionally reveals the capacity of white muscle to support ATP turnover. We sought length-matched cohorts, and we examined fast-starts among juveniles, and sprinting among juveniles and adults. We used 3 groups of Coho salmon: GH-transgenic fed to satiation (GHf), wild fed to satiation, and GHtransgenic par-fed according to the fed consumption of wild Coho salmon. For fast-start performance we simulated predatory strikes and quantified the mechanics and escape velocity via high-speed video. For sprinting performance between juveniles we measured: time to exhaustion, anaerobic substrate usage, velocity, and tail-beat frequency. For sprinting between adults, who had similar body lengths (though GHf were significantly longer than wild Coho), we quantified time to exhaustion and velocity. During fast-starts we found similar mechanics and escape velocity between all groups. We conclude that juvenile GHf may have a similar ability to evade predatory strikes, and GH had no effect on the affectors of fast-starts (we additionally presented across-species comparisons of the effects of GH on muscle enzyme activity, fibre-types, and contractile properties). During sprinting between juveniles, we found similar velocity, tail-beat frequency, and substrate usage; however, GHf took significantly longer to exhaust. We concluded that juvenile GHf might have an enhanced ability to evade chasing predators and to chase prey, due to a higher capacity to support ATP turnover. Between adults, we found similar time to exhaustion, but GHf exhibited significantly lower velocities. While the scaling of velocity offered some insight, differences in body length complicated our analysis. Nevertheless, we suggest that adult GHf may have a reduced ability to evade chasing predators, to chase prey, and to complete spawning migrations.  ii  TABLE OF CONTENTS Abstract ............................................................................................................. ii Table of Contents ............................................................................................. iii List of Tables ...................................................................................................... v List of Figures ....................................................................................................vi List of Abbreviations ........................................................................................vii Acknowledgements ........................................................................................ viii Dedication ......................................................................................................... ix Co-Authorship Statement ..................................................................................x  1 Introduction ................................................................................................ 1 1.1 General Introduction ................................................................................ 1 1.2 References ................................................................................................ 6  2 Fast-Starts ................................................................................................ 12 2.1 Introduction ............................................................................................ 2.2 Methods .................................................................................................. 2.2.1 Fast-starts .......................................................................................... 2.2.2 Data analysis ..................................................................................... 2.2.3 Statistics ............................................................................................ 2.3 Results .................................................................................................... 2.4 Discussion .............................................................................................. 2.4.1 Fast-start results ................................................................................ 2.4.2 Effect of GH on muscle fibre-types and contractile properties............ 2.4.3 Ability to evade prospective predation ............................................... 2.5 References ..............................................................................................  12 17 17 18 19 20 30 30 34 40 42  3 Sprinting .................................................................................................... 54 3.1 Introduction ............................................................................................ 54 3.2 Methods .................................................................................................. 58 3.2.1 Sprinting ............................................................................................. 58 3.2.2 Statistics ............................................................................................ 66 3.3 Results .................................................................................................... 67 3.3.1 Net anaerobic substrate usage .......................................................... 73 3.3.2 Effects due to time to exhaustion ....................................................... 75 3.3.3 Velocity and the effects of body length .............................................. 80 3.4 Discussion .............................................................................................. 87 3.4.1 Exhaustive sprinting, fatigability, and muscle biochemistry ............... 88 3.4.2 The effects of GH on muscle enzymes .............................................. 95 3.4.3 Changes in sprint performance with changes in body length .......... 101 3.4.4 Ecological importance of sprinting and prospective fitness ............. 106 iii  3.5 References ............................................................................................ 113  4 Conclusion ................................................................................ 125 4.1 General Conclusion ............................................................................. 125 4.2 References ............................................................................................ 129  iv  LIST OF TABLES Table 2.1 Fast-start kinematics results .............................................................. 22 Table 2.2 MANOVA of fast-start results ............................................................. 24 Table 2.3 ANCOVA using BL as covariate ........................................................ 25 Table 2.4 ANCOVA using stimulus proximity and BL as covariates .................. 28 Table 2.5 Across-species comparison of the effects of GH administration on muscle fibre-types proportions ........................................................... 36 Table 3.1 ANCOVA for Y-intercept and median velocity using the slope (changes in velocity over time) as covariate ...................................... 65 Table 3.2 Sprinting kinematics results ............................................................... 68 Table 3.3 ANOVAs for sprinting results ............................................................. 70 Table 3.4 ANOVAs of substrate content within white muscle ............................ 75 Table 3.5 ANCOVAs of sprinting results using time as a covariate .................... 77 Table 3.6 ANCOVAs for median and maximum velocity, among individuals within the same season, using BL as covariate.................................. 81 Table 3.7 ANCOVAs for maximum velocity, length-specific and absolute, within and between seasons, using BL as a covariate ....................... 85 Table 3.8 Across-species comparison of the effects of GH administration on muscle enzyme activity....................................................................... 98  v  LIST OF FIGURES Figure 2.1 An image of a single fast-start, near the end of stage 1 .................... 19 Figure 2.2 Duration of fast-start stage 1 as a function of BL .............................. 26 Figure 2.3 Peak escape velocity as a function of BL ......................................... 27 Figure 2.4 Peak escape velocity as a function of stimulus proximity ................. 29 Figure 3.1 Instantaneous velocity as a function of time to exhaustion, for one individual .......................................................................................... 62 Figure 3.2 Y-intercept and median velocity as a function of the slope, or change in velocity over time ............................................................. 63 Figure 3.3 Average velocity as a function of tail-beat frequency, for one individual .......................................................................................... 64 Figure 3.4 Average time to exhaustion as a function of BL ............................... 71 Figure 3.5 Average velocity as a function of tail-beat frequency ....................... 72 Figure 3.6 ATP, CrP, and lactate content within rested and exhausted white muscle .............................................................................................. 74 Figure 3.7 The change in velocity over time, median, and maximum lengthspecific velocity as a function of time to exhaustion ......................... 79 Figure 3.8 Median and maximum length-specific velocity as a function of BL ... 82 Figure 3.9 Maximum length-specific and absolute velocity as a function of BL . 86  vi  ABBREVIATIONS ALT = alanine aminotransferase AST = asparate aminotransferase ATP = adenosine triphosphate BL = body length BMI = body mass index CCO = cytochrome c oxidase CrP = creatine phosphate CS = citrate synthase EDL = extensor digitorum longus muscle GDH = glutamate dehydrogenase GE = genetically engineered GH = growth hormone GHf = growth hormone-transgenic coho salmon fed to satiation GHr = growth hormone-transgenic coho salmon ration-restricted or par-fed according to the amount of feed consumed by the wild coho salmon GHR = growth hormone receptor glut. med. = gluteus medius muscle HAD = β-hydroxyacyl-CoA dehydrogenase IDH = isocitrate dehydrogenase IGF-1 = insulin-like growth factor 1 i.m. = intramuscular i.v. = intravenous LDH = lactate dehydrogenase long. dorsi = longissimus dorsi muscle MDH = malate dehydrogenase PFK = phosphofructokinase PK = pyruvate kinase s.c. = subcutaneous Ucrit = maximum sustained swimming velocity vastus intermed. = vastus intermedius muscle  vii  ACKNOWLEDGEMENTS Funding for this research was provided by grants from NSERC (to Dr. Robert E. Shadwick), the Canadian Regulatory Systems for Biotechnology (to Dr. Robert H. Devlin), and by SigmaXi grants in-aid-of research (to myself).  I would like to thank my committee members (including Dr. John Gosline and Dr. Jeffery G. Richards) for their invaluable oversight and feedback, especially my supervisor Dr. Robert E. Shadwick who afforded me the opportunity to study here and to do this work, and Dr. Robert H. Devlin who has been developing GHtransgenic salmonids for over 15 years. As well, many thanks to my labmates and peers for their fellowship and their own hard work and dedication, which served as an inspiration.  Finally, I would like to thank my family for their support over years.  viii  DEDICATION To my family and an unyielding desire to return home.  ix  CO-AUTHORSHIP STATEMENT The identification and design of this research was done in collaboration with my committee, especially my supervisor, Dr. Robert E. Shadwick and Dr. Robert H. Devlin. I performed all experiments, with some additional assistance from two undergraduate students, Amanda Rnic and Kassandra McFarlane. I performed all data analysis, also with some additional assistance from two undergraduate students, Amanda Rnic and Rachel Cadelina. I prepared the present manuscript with the input and feedback of my committee members, especially my supervisor, Dr. Robert E. Shadwick.  x  CHAPTER 1: GENERAL INTRODUCTION  Foodstuffs have been artificially selected and genetically engineered (GE) to yield more desirable traits such as improved growth rate and feed-conversion efficiency (Pew Initiative, 2007; Urbanik, 2007). Such modifications may hasten product turn-around times, thereby reducing overhead production costs.  GE  agricultural products are currently being used for human consumption, and GE agricultural as well as livestock and fish continue to be developed (Pew Initiative, 2007; 2003, respectively).  Additionally, alternative sources of aquaculture  species have the potential to mitigate declining fish stocks, particularly among wild salmonid populations (Lackey, 2003; MacKinlay et al., 2004). While those outcomes may be beneficial, would they also include any unintended negative consequences? For example, if fish with enhanced growth-rates escaped into the natural environment, how would they impact the ecosystems into which they escaped, or how would their escape impact existing species? Therefore, any associated cost-benefit analysis should consider the prospective ecological impact of the release of GE organisms (Muir and Howard, 2002; Pew Initiative, 2003; Devlin et al., 2006). In fact, producers seeking approval from the U.S. Food and Drug Administration (USFDA), to market GE animal products must confer not only food safety but also the prospective impact of GE animals escaping into natural environments (USFDA, 2009). That is, would the ecological impact of GE escapees be any different from that of their non-GE conspecifics?  1  Growth hormone (GH) regulates postembryonic muscle growth in a variety of species (Florini et al., 1996; Kopchick and Andry, 2000; Rescan, 2005; Velloso, 2008) and has been targeted as a means to improve growth rate and the efficiency of feed-conversion (Etherton and Bauman, 1998).  Indeed, Atlantic  (Salmo salar) and Coho (Oncorhynchus kisutch) salmon transgenic for GH upregulation realize dramatic increases in overall body and muscle growth (Du et al., 1992; Devlin et al., 1994, respectively). Thus, (in addition to other species of fish; Pew Initiative, 2003) they may be viable candidates for aquaculture. Our work addressed, in part, the prospective ecological fitness of GH-transgenic Coho salmon in relation to their wild conspecifics. The theoretical framework for assessing their prospective ecological fitness has been set out previously by Devlin et al. (2006).  Specifically, we were interested in their antipredator  behavior or their ability to evade predatory strikes, chasing predators, and to chase prey, as well as their prospective ability to successfully complete a spawning migration, whereby they might reach a spawning ground and yield the progeny of GE fish.  We investigated those aspects, using relatively simple  experiments, by assessing their fast-start and sprint performance. Fast-starts (or C-starts) are rapid, maximal escape maneuvers, which typically occur in response to a threatening stimulus or predatory strike (Domenici and Blake, 1997) and are completed within 100ms among salmonids (Webb, 1978; Hale et al., 2002). Moreover, faster escape velocities have been shown to increase the probability of evading predators (Taylor and McPhail, 1985; Walker  2  et al., 2005).  Thus, by comparing the fast-start performance between GH-  transgenic and wild coho salmon we can make inferences regarding their prospective ability to evade a predator strike, or their prospective fitness during such encounters. Previous studies on maximal sustained swimming velocity (Ucrit) revealed that GH-transgenic salmon have a lower Ucrit than their wild conspecfics throughout their life history (Farrell et al., 1997; Stevens et al., 1998; Lee et al., 2003; Deitch et al., 2006). This lower Ucrit in GH-transgenic salmon has the potential to impact their ability to successfully complete a spawning migration. During spawning migrations, salmonids average modest velocities, which typically fall below their Ucrit (Brown and Geist, 2002; Hinch et al., 2002). However, in order to successfully negotiate specific challenges, such as waterfalls, salmonids must use sprinting or bursting swimming velocities (Lauritzen et al., 2005), which are in excess of Ucrit and can only be maintained for short periods (< 2min.; Richards et al., 2002a). Relatively, few studies have examined fish sprinting, and no such studies have considered GH-transgenic fish. Therefore, we choose to address the prospective ability of GH-transgenic Coho salmon to successfully complete a spawning migration by comparing their performance during exhaustive sprinting to that of their wild conspecifics. We also presented evidence suggesting the use of sprinting during predator-prey interactions that involve chasing.  Thus, our work on sprint performance  3  additionally infers their prospective ability to evade chasing predators and to chase prey. GH upregulates post-embryonic muscle growth (Florini et al., 1996; Velloso, 2008) via net anabolism of contractile proteins (Pell and Bates, 1987). GH administration has typically been linked with the amelioration of reduced exercise capacity and muscle wasting, which may be induced by aging, disuse, or clinical disease-states such as hypopituitarism or GH insensitivity (Gibney et al., 2007). However, beyond upregulating muscle growth, the effect of GH on the functioning of normal muscle has not been well defined. Therefore, in addition to addressing the prospective abilities of GH-transgenic coho salmon, our work also addressed the effects of GH upregulation on muscle functioning in vivo. A mechanistic explanation for the lower Ucrit of GH-transgenic salmon has not been investigated. Sustained swimming is powered by red (type-1 or slowtwitch) lateral muscle (Coughlin, 2002; Richards et al., 2002b).  Sustained  swimming also employs multiple organs and systems (Johnston et al., 1977; McKenzie et al., 2004), which, in part support ATP turnover (Dobson et al., 1987). This is in addition to the musculoskeletal system, responsible for the development and transmission of muscle force to provide propulsion.  By  contrast, sprinting and fast-starts are two other swimming modes employed by most telost fish (Domenici, 2003), and are powered almost exclusively by the white (type-2 or fast-twitch, glycolytic) lateral muscle (Richards et al., 2002a). Because white muscle is more limited in its capacity to support ATP turnover  4  than red muscle (Dobson et al., 1987; Goolish et al., 1991), sprinting and faststarts offer an alternative approach whereby we may determine more specifically the effect of GH upregulation on the performance of the white lateral muscle of GH-transgenic Coho salmon in vivo. The present thesis is the first study to compare fast-start and sprinting performance between GH-transgenic and wild fish. As such, this work enabled us to infer the prospective ability of GH-transgenic coho salmon escapees to evade predatory strikes and chasing predators, to chase prey, and to successfully complete spawning migrations. Lastly, because the lateral white muscle powers fast-starts and sprinting, we were also able to determine more specifically the effect of GH upregulation on muscle functioning in vivo.  5  REFERENCES Brown, R.S. and D.R. Geist. 2002. Determination of Swimming Speeds and Energetic  Demands  of  Upriver  Migrating  Fall  Chinook  Salmon  (Oncorhynchus tshawytscha) in the Klickitat River, Washington. Bonnevill Power Administration, Portland, Ore. PNNL-12975. Coughlin, D.J. 2002. Aerobic muscle function during steady swimming in fish.  Fish and Fisheries, 3: 68-78. Deitch, E.J, G.L. Fletcher, L.H., L. H. Petersen, I. A. S. F. Costa, M. A. Shears, W. R. Driedzic, and A. K. Gamperl.  2006.  Cardiorespiratory  modifications, and limitations, in post-smolt growth hormone transgenic Atlantic salmon Salmo salar. Journal of Experimental Biology, 209: 131025. Devlin, R.H., T.Y. Yesaki, C.A. Biagi, E.M. Donaldson, P. Swanson, and W.-K. Chan. 1994. Extraordinary salmon growth. Nature, 371: 209-10. Devlin, R.H., L.F. Sundström, and W.M. Muir. 2006. Interface of biotechnology and ecology for environmental risk assessments of transgenic fish.  Trends in Biotechnology, 24(2): 89-97. Dobson, G.P., W.S. Parkhouse, and P.W. Hochachka.  1987.  Regulation of  anaerobic ATP-generating pathways in trout fast-twitch skeletal muscle.  American Journal of Physiology: Regulatory Integrative and Comparative Physiology, 253: R186-94.  6  Domenici, P. and R.W. Blake. 1997. The kinematics and performance of fish fast-start swimming. Journal of Experimental Biology, 200: 1165-78. Domenici, P. 2003. Habitat, body design, and swimming performance of fish.  Vertebrate Biomechanics and Evolution. Eds. V.L. Bels, J.P. Gasc and A. Casinos. BIOS Scientific Publishers Ltd, Oxford. Pg. 137-60. Du, S.J., Z. Gong, G.L. Fletcher, M.A. Shears, M.J. King, D.R. Idler, and C.L. Hew. 1992. Growth enhancement in transgenic Atlantic salmon by use of an “all fish” chimeric growth hormone gene construct. Biotechnology, 10: 176-81. Etherton, T.D. and D.E. Bauman. 1998. Biology of Somatotropin in Growth and Lactation of Domestic Animals. Physiological Reviews, 78(3): 745-61. Farrell, A.P., W. Bennett, and R.H. Devlin. 1997. Growth-enhanced transgenic salmon can be inferior swimmers. Canadian Journal of Zoology, 75: 3357. Florini, J.R., D.Z. Ewton, and S.A. Coolican. 1996. Growth Hormone and the Insulin-Like Growth Factor System in Myogenesis. Endocrine Reviews, 17(6): 481-517. Gibney, J., M-L. Healy, and P.H. Sönksen. 2007. The Growth Hormone/InsulinLike Growth Factor-I Axis in Exercise and Sport.  Endocrine Reviews,  28(6): 603-24. Goolish, E.M. 1991. Aerobic and anaerobic scaling in fish. Biological Reviews, 66: 33-56.  7  Hale, M.E., J.H. Long, M.J. McHenry, and M.W. Westneat. 2002. Evolution of behavior and neural control of the fast-start escape response. Evolution, 56(5): 993-1007. Hinch, S.G., E.M. Standen, M.C. Healey, and A.P. Farrell. 2002. Swimming patterns and behavior of upriver-migrating adult pink (Oncorhynchus  gorhuscha) and sockeye (O. nerka) salmon as assessed by EMG telemetry in the Fraser River, British Columbia, Canada. Hydrobiologia, 483: 147-60. Johnston, I.A., W. Davison, and G. Goldspink. 1977. Energy metabolism of carp swimming muscles. Journal of Comparative Physiology B, 114 (2): 20316. Kopchick, J.J. and J.M. Andry. 2000. Minireview – Growth Hormone (GH), GH Receptor, and Signal Transduction. Molecular Genetics and Metabolism, 71: 293-314. Lackey, R.T. 2003. Pacific Northwest Salmon: Forecasting Their Status in 2100.  Reviews in Fisheries Science, 11(1): 35-88. Lauritzen, D.V, F. Hertel, and M.S. Gordon. 2005. A kinematic examination of wild sockeye salmon jumping up natural waterfalls.  Journal of Fish  Biology, 67: 1010-20. Lee, C.G., R.H. Devlin, and A.P. Farrell. 2003. Swimming performance, oxygen consumption and excess post-exercise oxygen consumption in adult  8  transgenic and ocean-ranched coho salmon. Journal of Fish Biology, 62: 753-66. MacKinlay, D., S. Lehmann, J. Bateman, and R. Cook. 2004. Pacific salmon hatcheries in British Columbia. American Fisheries Society Symposium, 44: 57–75. McKenzie, D.J., S. Wong, D.J. Randall, S. Egginton, E.W. Taylor, and A.P. Farrell. 2004. The effects of sustained exercise and hypoxia upon oxygen tensions in the red muscle of rainbow trout.  Journal of Experimental  Biology, 207: 3629-37. Muir, W.M. and R.D. Howard. 2002. Assessment of possible ecological risks and hazards of transgenic fish with implications for other sexually reproducing organisms. Transgenic Research, 11: 101-14. Pell J.M. and P.C. Bates. 1987. Collagen and non-collagen protein turnover in skeletal muscle of growth hormone-treated rats. Journal of Endocrinology, 115: R1-4. Pew Initiative.  2003.  Future Fish:  Issue in Science and Regulation of  Transgenic Fish. Pew Initiative on Food and Biotechnology. <http://www.pewtrusts.org/our_work_detail.aspx?id=442> Pew Initiative. 2007. Application of Biotechnology for Functional Foods. Pew Initiative on Food and Biotechnology. <http://www.pewtrusts.org/our_work_detail.aspx?id=442>  9  Rescan, P.Y. 2005. Muscle growth patterns and regulation during fish ontogeny.  General and Comparative Endocrinology, 142: 111-6. Richards, J.G., G.J.F. Heigenhauser, and C.M. Wood.  2002a.  Glycogen  phosphorylase and pyruvate dehydrogenase transformation in white muscle of trout during high-intensity exercise.  American Journal of  Physiology: Regulatory Integrative and Comparative Physiology, 282: R828-36. Richards, J.G., A.J. Mercado, C.A. Clayton, G.J.F. Heigenhauser, and C.M. Wood. 2002b. Substrate utilization during graded aerobic exercise in rainbow trout. Journal of Experimental Biology, 206: 2067-77. Stevens, E.D. A. Sutterlin, and T. Cook. 1998. Respiratory metabolism and swimming performance in growth hormone transgenic Atlantic salmon.  Canadian Journal of Fisheries and Aquatic Sciences, 55: 2028-35. Taylor, E.B. and J.D. McPhail.  1985.  Burst Swimming and Size-Related  Predation of Newly Emerged Coho salmon Oncorhynchus kisutch.  Transactions of the American Fisheries Society; 114: 546-51. Urbanik, J. 2007. Locating the transgenic landscape: Animal biotechnology and politics of place in Massachusetts. Geoforum, 38: 1205-18. USFDA. 2009. Guidance for Industry: Regulation of Genetically Engineered Animals Containing Heritable Recombinant DNA Constructs.  US Food  and Drug Administration – Center for Veterinary Medicine. <http://www.fda.gov/cvm/geanimals.htm>  10  Velloso, C.P. 2008. Regulation of muscle mass by growth hormone and IFG-1.  British Journal of Pharmacology, 154: 557-68. Walker, J.A., C.K. Ghalambor, O.L. Griset, D. McKenney, and D.N. Reznick. 2005.  Do faster starts increase the probability of evading predators?  Functional Ecology, 19: 808-15. P.W. Webb. 1978. Fast-start performance and body form in seven species of teleost fish. Journal of Experimental Biology, 74: 211-26.  11  CHAPTER 2: FAST-STARTS1  INTRODUCTION Fast-starts are rapid escape responses typically exhibited in response to predator strikes in the wild (Domenici, 2003), and they have also been induced by perceived stimuli under more controlled conditions (Tytell and Lauder, 2002). The present study addressed the ability of GH-transgenic Coho salmon to evade predatory strikes by comparing their fast-start performance during simulated predatory attacks to that of their wild conspecifics. Additionally, as the effects of GH on muscle physiology beyond upregulating muscle growth are unclear (Gibney et al., 2007) this study also considers the effects of GH upregulation on muscle function during fast-starts. Previous work has shown that GH-transgenic salmon (both Atlantic and Coho) have a lower maximum sustained swimming velocity (Ucrit) throughout their life history (Farrell et al., 1997; Stevens et al., 1998; Lee et al., 2003; Deitch et al., 2006). In other words, GH-transgenic salmon may not perform as well as their wild conspecifics during long-distance migrations. Sustained swimming is powered by red lateral muscle (Coughlin, 2002; Richards et al., 2002b). Red muscle can also employ organs and systems to support ATP turnover (Johnston et al., 1977; McKenzie et al., 2004), thus enabling sustained swimming. Because  1  A version of this chapter will be submitted for publication. Dimoulas, P.M., R.H. Devlin, and R.E. Shadwick. Fast-start performance of GH-transgenic Coho salmon.  12  Ucrit fails to account for ATP turnover supported by other such factors, it is currently not possible to determine specifically why GH-transgenic salmon exhibit a lower Ucrit or to partition the effect of GH on muscle functioning in these fish. In contrast to sustained swimming, fast-starts last < 100 ms and are powered predominately by white, burst, or fast-twitch lateral or axial muscle fibres (Jayne and Lauder, 1993; Coughlin, 2002; Wakeling, 2006), which uses stored CrP, ATP, and glycogen via glycolysis (Richards et al., 2002a). Fast-starts can be considered a summation of: 1) short-term behavior (Eaton et al., 2001), 2) muscle contractile properties and excitation contraction coupling of their white muscle (Syme, 2006), 3) musculoskeletal linking (Gemballa and Vogel, 2002; Long et al., 2002; Summers and Long, 2006; for additional reviews of muscletendon organization among fish please see Shadwick and Gemballa, 2006), and 4) body shape (Wakeling, 2006). White muscle is more limited in its ability to support ATP turnover (Dobson et al., 1987; Goolish, 1991). measuring  fast-start  performance  simultaneously  Even though  addresses  the  four  aforementioned factors, because white muscle is more limited in its ability to support ATP turnover we are likely to determine more specifically the effects of GH on muscle functioning. Fast-starts are completed in three sequential stages (Wakeling, 2006): 1) rapid lateral muscle contraction induces bending of the body into a C-shape, 2) a propulsive wave of muscle contraction, along the convex-side, provides initial thrust away from the center-of-mass, and 3) the fish employs undulatory  13  swimming to continue or hasten its continued escape. We chose to quantify the mechanics of the C-bend by quantifying the time duration and the magnitude of muscle shortening achieved during or at the end of stage 1, and the resultant peak escape velocity achieved during stages 2 or 3. Additionally, because closer stimuli invoked faster escape responses (Tytell and Lauder, 2002) we also quantified the effect of stimulus proximity. Regarding  the  four  aforementioned  factors  influencing  fast-start  performance, there is some evidence that suggests that GH-transgenic and wild fish might differ. GH has also been shown to influence the proportion of muscle fibre-types among such faster growing fish. Firstly, among most teleost fish, red muscle typically constitutes 10% of lateral muscle among adults, and less among juveniles (the remaining area being composed, almost exclusively of white muscle; Johnston, 2001). At approximately 9g and 9cm body length (BL), the proportions of red and white lateral muscle of size-matched GH-transgenic Coho salmon differed significantly, with a total of 1.6 – 1.9 % red muscle in GHf compared with 1% in wild Coho salmon (Hill et al., 2000). There is also evidence to suggest that GH-transgenic Coho salmon might also differ in their body shape (Ostenfeld et al., 1998). Both muscle fibre proportions and body shape have the potential to affect fast-start performance, but the functional significance of those results remains unclear. Interestingly, the twitch properties of 7 week-old mice that received daily injections of GH for 4 weeks were not changed in the extensor digitorum longus muscle (EDL) (Kim et al., 2005). However, GH has been shown  14  to influence neural functioning among mice (Chen et al., 1997; Harvey and Hull, 2003) and bone development among mice (Ohlsson et al., 1998) as well as fish (Nordgarden et al., 2006).  Yet how those findings may influence fast-start  performance of fish also remain unclear. The present study compared the fast-start performance or kinematics between length-matched GH-transgenic Coho salmon fed to satiation (GHf), wild Coho salmon fed to satiation, and GH-transgenic Coho salmon par-fed according to the amount of feed consumed by the wild group (GHr). GHr exhibit a similar growth rate as wild coho salmon (Devlin et al., 2001). Since GHf grow faster, GHf are younger than their length-matched conspecifics (Devlin et al., 2004). Therefore we chose to use GHr to control for potential differences due to age as well as GH upregulation.  Additionally, we chose to use length-matched  individuals because muscle contractile properties are known to vary according to BL (James and Johnston, 1998). We used high-speed video to determine the time duration and magnitude of muscle shortening achieved at the end of stage 1, the resultant peak escape velocity, and the proximity of the stimulus. Even though there are differences in muscle fibre proportions (Hill et al., 2000) and a preliminary study showed the potential for differences in body shape between GHf and wild coho salmon (Ostenfeld et al., 1998), we speculated that the magnitude of the actual differences might have little functional significance. Furthermore, while GH has been shown to influence bone development and neural functioning, such evidence does not clearly support a hypothesis  15  regarding fast-start performance of GH-transgenic fish.  Therefore, we  hypothesized that there would be no differences in fast-start performance between all three groups. Our results showed that we did not find any significant differences in fast-start performance between juvenile GHf, GHr, and wild Coho salmon. Thus, we suggest that GH has no effect on the fast-start performance of juvenile GH-transgenic Coho salmon, which may be equally capable of evading predatory strikes as their wild conspecifics.  16  METHODS Transgenic coho salmon were produced and all fish were raised according to Devlin et al. (2004) at the Department of Fisheries and Oceans Canada laboratory in West Vancouver. Pre-smolts (parr) from each group were housed in a 200-L fiberglass aquarium with fresh, flow-through well water maintained at 11 oC ± 1 (during our experiments) and kept under natural light conditions. GHf and wild Coho salmon were fed to satiation using a commercial salmon diet (Skretting Canada Ltd., Vancouver, BC), and GHr were par-fed (averaged per individual) according to the amount of feed consumed by the wild coho salmon.  Fast-Starts Fast-starts were conducted on fish, approximately 15 cm BL, between 26 August – 6 October 2006 when wild and GHr were 17-19 months old, and GHf were 5-7 months old.  Fish were netted and transferred to a fiberglass oval  aquarium (1.2 x 0.6 x 0.25 m) with flow-through, oxygenated well water kept at 11-13oC. Fish were allowed 20 minutes to acclimate and were given 5 minutes of rest between replicates. With a given fish at least 1 BL from the nearest wall, fast-starts were initiated by dropping a 50g weight into the tank while high-speed video (500 Hz) was recorded from above via a PC-based high-speed camera (Fastcam X1280 PCI 16k, Photron USA Inc., San Diego, CA).  17  Data Analysis Image sequences were exported and digitized into 30 points along either side of a given fish using NIH Image (National Institutes of Health, http://rsb.info.nih.gov/nih-image/) and Quick Image, a modification of NIH Image, by Jeffrey Walker (http://www.usm.maine.edu/%7Ewalker/software.html), for Mac OS 9. The outline points were used to map the position of the backbone and determine its curvature as a function of time (see fig. 2.1). Curvature of the backbone was used to calculate the magnitude of fractional shortening within the white lateral muscle (Goldbogen et al., 2005). Peak white muscle shortening (or the magnitude of shortening) was calculated at 0.4 and 0.7 BL as well as the time elapsed (or duration of stage 1) from the beginning of a given fast-start, which was indicated by the first detectable movement of the rostrum. The average rate of shortening was obtained by dividing the magnitude of shortening by the time elapsed. Maximum escape velocity (averaged over 10ms) was determined by tracing the nose, of a given fish, frame-by-frame using NIH-Image, as above, following the realization of peak shortening at 0.4 BL. We choose this method because the end of stage 1, at 0.4 BL, requires the beginning of the second propulsive stage or development of contralateral muscle shortening (Hale et al., 2002). Finally, the proximity of the stimulus was obtained by determining the distance from the nose of a given fish to the closest edge of the 50g weight in the horizontal plane, upon the onset of detectable movement of the rostrum.  18  Figure 2.1 An image of a fast-start at approximately the end of stage 1. Arrows indicate the positions of 0.4 and 0.7 BLs, the stimulus dropped from above. Arrows also point to the position of the rostrum and the original body axis, and the approximate position of the backbone, both of which are indicated by a dotted line.  Statistics ANCOVAs and MANCOVAs were performed using Minitab (Release 13.32, Minitab Inc., State College, PA,) via the “general linear trend” option for ANOVA analysis on a Windows XP – based personal computer (as described in Dytham, 2003). P-values less than 0.05 were deemed significant.  19  RESULTS A summary of the average time elapsed, the shortening magnitude, and the rate of shortening at both 0.4 and 0.7 BL as well as the escape velocity among all individual responses per group has been provided (table 2.1).  A  MANOVA among all parameters and replicates from table 2.1, using BL as a covariate, revealed no significant differences between the groups (table 2.2). However, the associated ANCOVAs revealed that BL exerted a significant effect upon the time elapsed at both 0.4 and 0.7 BL as well as escape velocity (table 2.3). Both time elapsed and escape velocity have been plotted as a function of BL to which linear regressions were fit (figs. 2.2 and 2.3, respectively). Among all individuals from the three groups, longer individuals took more time to realize peak shortening at 0.4 and 0.7 BL, respectively (fig. 2.2), and also exhibited a slower length-specific escape velocity (fig. 2.3). A second MANOVA among all parameters and replicates from table 2.1, using BL and stimulus proximity as covariates revealed no significant differences between the groups (table 2.2).  Because we could not obtain the stimulus  proximity from all replicates, we used reduced sample sizes (GHf n= 4 indviduals, 2-3 replicates; wild n= 4, 1-4; and GHr n= 4,4).  The associated ANCOVAs  revealed that BL exerted a significant effect upon the time elapsed at 0.7 BL and escape velocity, and that stimulus proximity also exerted a significant effect upon escape velocity (table 2.4). The escape velocity has been plotted as a function of stimulus proximity to which a linear regression was fit (fig. 2.4), whereby  20  individuals responded to a closer, more proximal stimulus with a lower lengthspecific escape velocity.  21  Table 2.1  22  Table 2.1 Fast-start parameters compared between all 3 groups. Averages ± S.E.M. of the time elapsed (ms), the shortening magnitude achieved within the white muscle (half the distance between the skin and the backbone, as a percent), and the rate of shortening (percent ms-1) are shown, at 0.4 and 0.7 BL, as well as escape velocity (BL s-1) for each group. The averages were calculated from all of the replicates, per group.  23  Source  Pillai Trace  F  d.f.  P  BL (covariate) BL Group  0.278 0.167  2.10 0.51  7, 38 14, 78  0.068 0.922  1.69 1.24 1.09  7, 27 7, 27 14, 56  0.153 0.315 0.385  BL & Stimulus Proximity (covariates) BL 0.305 Stimulus prox. 0.244 Group 0.429  Table 2.2 Comparison of fast-start parameters as a function of BL, as well as BL and stimulus proximity, between the groups. MANOVA results for all the variables (from table 2.1), and the effects due to body length (BL) and group (ANCOVA from table 2.3) as well as the effects due to BL, stimulus proximity, and group (ANCOVA from table 2.4).  24  Source  d.f.  Time elapsed at 0.4BL BL 1 Group 2 Error 44 Shortening Magnitude at 0.4BL BL 1 Group 2 Error 44 Shortening Rate at 0.4BL BL 1 Group 2 Error 44 Time elapsed at 0.7BL BL 1 Group 2 Error 44 Shortening Magnitude at 0.7BL BL 1 Group 2 Error 44 Shortening Rate at 0.7BL BL 1 Group 2 Error 44 Escape Velocity BL 1 Group 2 Error 44 * *  MS  F  P  565.13 22.13 70.55  8.01 0.31  0.007** 0.73  15.08 14.05 5.64  2.76 2.57  0.10 0.09  0.026 0.007 0.015  1.75 0.45  0.19 0.64  1108.6 115.1 126.4  8.77 0.91  0.005** 0.41  46.1 13.85 17.4  2.65 0.80  0.11 0.46  0.0002 0.002 0.010  0.01 0.17  0.91 0.85  37.42 2.06 6.69  5.59 0.31  0.02* 0.74  Indicates significance level < 0.05 Indicates significance level < 0.01  Table 2.3 Comparison of individual fast-start parameters as a function of BL between the groups. ANCOVA results between the groups for the time elapsed, the magnitude of shortening, and the average shortening rate achieved at 0.4 and 0.7 BL, at the stage 1 BL as the covariate.  25  Time to Peak Shortening (ms)  GHf 0.4 GHf 0.7 wild 0.4 wild 0.7 GHr 0.4 GHr 0.7 regression 0.4 regression 0.7  80  60  40  20  0 10  12  14  16  18  Body Length (cm) Figure 2.2 The time to peak shortening (ms) at 0.4 and 0.7 BL are shown, for each replicate per individual, as a function of BL (cm). Filled blue squares represent GHf at 0.4 BL, open blue squares represent GHf at 0.7 BL, filled red circles represent wilds at 0.4 BL, open red circles represent wilds at 0.7 BL, filled green triangles represent GHr at 0.4, and open green triangles represent GHr at 0.7 BL. The linear regressions among all replicates at 0.4 BL is represented by the solid line (peak shortening = 1.6*BL - 2.5, r2= 0.38), and 0.7 BL is represented by the dotted line (peak shortening = 2.33*BL - 3.5, r2= 0.36).  26  GHf wild GHr regression  14  -1  Escape Velocity (BL s )  16  12  10  8  6  4 10  12  14  16  18  Body Length (cm) Figure 2.3 The average maximal escape velocity (BL s-1) achieved following the completion of stage 1 (of a given fast-start) at 0.4 BL has been plotted as a function of BL (cm) for each replicate per individual. Filled blue squares represent GHf, filled red circles represent wilds, and filled green triangles represent GHr. The resultant linear regression (escape velocity = 0.65*BL - 19.62, r2= 0.32) was obtained from all replicates shown.  27  Source  d.f.  Time elapsed at 0.4BL BL 1 Stimulus prox. 1 Group 2 Error 33 Shortening Magnitude at 0.4BL BL 1 Stimulus prox. 1 Group 2 Error 33 Shortening Rate at 0.4BL BL 1 Stimulus prox. 1 Group 2 Error 33 Time elapsed at 0.7BL BL 1 Stimulus prox. 1 Group 2 Error 33 Shortening Magnitude at 0.7BL BL 1 Stimulus prox. 1 Group 2 Error 33 Shortening Rate at 0.7BL BL 1 Stimulus prox. 1 Group 2 Error 33 Escape Velocity BL 1 Stimulus prox. 1 Group 2 Error 33  MS  F  P  132.27 24.63 15.03 64.84  2.04 0.38 0.23  0.16 0.54 0.79  0.39 6.19 6.30 4.35  0.09 1.42 1.45  0.77 0.24 0.25  0.004 0.023 0.020 0.013  0.31 1.82 1.56  0.58 0.19 0.23  472.7 0.00 19.5 108.0  4.388 0.00 0.18  0.044* 0.99 0.84  36.87 34.83 20.58 12.29  3.0 2.83 1.67  0.09 0.10 0.20  0.001 0.020 0.009 0.010  0.13 2.0 0.85  0.72 0.17 0.44  38.22 32.15 16.68 6.0  6.37 5.36 2.78  0.017* 0.027* 0.077  * Indicates significance level < 0.05 Table 2.4 Comparison of fast-start parameters as a function of BL and stimulus proximity between the groups. ANCOVA results between the groups for the time elapsed, the magnitude of shortening, and the average shortening rate achieved at 0.4 and 0.7 BL, at the stage 1 with stimulus proximity and BL as covariates.  28  16  -1  Escape Velocity (BL s )  14  12  GHf 1 GHf 2 GHf 3 GHf 4 wild 1 wild 2 wild 3 wild 4 GHr 1 GHr 2 GHr 3 GHr 4 regression  10  8  6  4 0.0  0.5  1.0  1.5  Stimulus Proximity (percent BL)  Figure 2.4 The average maximal escape velocity (BL s-1) achieved following the completion of stage 1 (of a given fast-start), at 0.4 BL, has been plotted as a function of the stimulus proximity (percent BL) for each replicate per individual. Filled blue symbols represent GHf individuals, open red symbols represent wild individuals, and filled green symbols represent GHr individuals (see figure legend for symbols assigned to a given individual). The resultant linear regression (escape velocity = (2.7 * stimulus proximity) + 8.56, r2= 0.28) represents all replicates shown.  29  DISCUSSION By assessing the fast-start performance of GH-transgenic and wild coho salmon, the present study infers the prospective ability of GH-transgenic coho salmon to evade predatory strikes, and it is the first study to do so using GHtransgenic fish.  Similarly, we also sought to determine more specifically the  effects of GH on muscle functioning in vivo. We did not find any significant differences in their fast-start performance between the 3 groups used (see tables 2.3, 2.4, & 2.5).  Therefore we suggest that GH does not affect short-term  behavior, muscle contractile properties, excitation contraction coupling of white muscle, or musculoskeletal linking among GHf, GHr, and wild coho salmon. Our results also suggest that GH-transgenic and wild coho salmon may be equally capable of evading predatory strikes.  Fast-Start Results Our fast-starts results (table 2.1) are similar to those obtained among related species. The time duration of stage 1 is comparable to that of similarly sized rainbow trout, Salmo gairdneri (Webb, 1978) and Oncorhynchus mykiss (Hale et al., 2002), angelfish (Domenicic and Blake, 1993), and short-horn sculpin (James and Johnston, 1998).  Our data are somewhat different from  those seen in bichirs, as these fish took slightly longer to realize their maximum C-bend, (Tytell and Lauder, 2002), which may have been due to their lower body stiffness. The shortening magnitudes achieved at the end of stage 1 (present  30  study) were also comparable to those of rainbow trout, O. mykiss (Goldbogen et al., 2005) and, though slightly lower in magnitude, short-horn sculpin (James and Johnston, 1998). As well, the escape velocity was comparable to that of rainbow trout (Webb, 1976; Hale et al., 2002) and angelfish (Domenici and Blake, 1993). The escape velocities observed in the present study were slightly higher than those found among short-horn sculpin (James and Johnston, 1998) and bichir (Tytell and Lauder, 2002).  While there could be important changes in body  shapes between short-horn sculpins and bichirs, for which we have not accounted, we may also attribute such differences to their lower muscle shortening magnitude and longer stage 1 duration, respectively. Overall, we found no significant differences between either of the groups, but we did find a significant effect of BL on stage 1 duration (table 2.3) and a significant effect of both BL and stimulus proximity on escape velocity (table 2.4). Regarding the duration of stage 1, as fish BL increased, stage 1 duration also increased (fig. 2.2). During fast-starts, as well as during undulatory swimming, the wave of body bending (or muscle shortening) must travel posteriorly (Goldbogen et al., 2005). Therefore, for a given fast-start, the muscle shortening magnitude, or the apex of the C-bend, must take more time to be realized at a given BL segment in a longer fish. Such an effect due to body position has also been obtained among similar species, which were noted previously (Webb, 1978; Domenici and Blake, 1993; James and Johnston, 1998).  Additionally, the  relationship for the stage 1 time duration increasing with increasing BL differed  31  between 0.4 and 0.7 BL (fig. 2.2). The relationship for 0.7 BL (exhibited by the resultant linear regression) had a higher y-intercept and slope than was exhibited at 0.4 BL. We attribute this difference to the need for nerve impulses to travel further along the body to reach 0.7 BL, whereby the muscle shortening magnitude was achieved at 0.7 BL later than at 0.4 BL. Escape velocity (length-specific) decreased as fish BL increased (fig. 2.2). This result has also been demonstrated previously among juvenile and adult fish for not only fast-starts (Webb, 1976; James and Johnston, 1998) but also for Ucrit (Goolish, 1991) and sprinting (Bainbridge, 1960). Length-specific swimming velocity decreases as BL increases because longer-larger fish employ lower mass-specific power outputs to realize the same absolute velocity. Although, Domenici and Blake (1993) did not find such a relationship during angelfish faststarts, they only included maximal escapes responses for their data analysis. Rather, we suggest that their inclusion of only high-performance responses may introduce an experimenter bias and not accurately reflect the diversity of escape responses observed by a given fish. We also found that escape velocity correlated with changes in stimulus proximity (fig. 2.4).  More specifically, among all individuals, escape velocity  decreased with a closer stimulus, which contradicts previous work among bichirs (Tytell and Lauder, 2002). Unfortunately, we were not able to obtain the stimulus proximity from all replicates. Nonetheless, we chose to include this relationship with a reduced number of replicates and applied equal statistical weighting to  32  each replicate. Within the present study, upon inspection of the aforementioned relationship for a single individual (see fig. 2.4), escape velocity did not appear to decrease with a more proximal stimulus. In fact, some individuals responded to a closer stimulus with a higher escape velocity, which is consistent with findings from Tytell and Lauder (2002). Therefore, it is possible that the relationship we obtained, escape velocity decreasing with a closer stimulus, may be attributed to the application of equal statistical weight to each replicate. We suggest that future studies should be undertaken, using more individuals or at least more replicates, which would provide more statistical power to address the potential for differences in the behavioral aspects of fast-starts between GHf, GHr, and wild Coho salmon. Regarding the MANOVAs, obtained for BL, and BL and stimulus proximity (table 2.2), there were no differences in fast-start performance between the groups. However, it is remarkable that the ANCOVAs for escape velocity using BL (table 2.3), and BL and stimulus proximity as covariates (table 2.4), realized a reduction in the P-value (predictive of prospective group differences) from 0.74 – 0.077, respectively. Moreover, on average the stimulus proximity was closer for GHr than it was for GHf or wild individuals; however, no difference in average escape velocity was obtained. Further investigation of the behavioral aspects of fast-starts between GH-transgenic and wild Coho salmon is warranted. No other fast-start studies have been conducted using GH-transgenic fish; however, contrasting findings have been obtained regarding the effect of  33  changes in growth rate (induced by variable feeding) on fast-start escape velocity, and burst swimming. Billerbeck et al. (2005), found that ration-restricted populations of Nova Scotia, Atlantic silverside, M. medidia, had higher bursting speeds (it was unclear whether they addressed sprinting or fast-start escape velocity) than their conspecifics, which were fed ad libitum. In contrast, Royle et al. (2006) found no effect of changing growth rates, via ration-restriction on the fast-start performance of Green sword-tails, X. helleri. Neither study proposed a viable physiological explanation.  Effects of GH on muscle fibre-types and contractile properties Even though fast-starts are a summation of 1) short-term behavior (Eaton et al., 2001), 2) muscle contractile properties and excitation contraction coupling of the white muscle (Syme, 2006), 3) musculoskeletal linking (Gemballa and Vogel, 2002; Long et al., 2002; Summers and Long, 2006) and 4) body shape (Wakeling, 2006), unless either one of the differing factors are compensating for one another, no significant differences in fast-start performance suggests that GH had no effect on average of the 4 factors.  GH upregulation did induce an  increase in the proportion of type I (red) muscle fibres among GHf, at 10 g and 9 cm, however, the absolute difference was less than 1 % (Hill et al., 2000). Faststarts are powered by white muscle, which among coho salmon from Hill et al. (2000), constituted approximately 98 % of their axial muscle.  The functional  significance of a 1 % difference in available white muscle to power fast-starts  34  seems unlikely. Interestingly, Hill et al. (2000) also found significant differences in the number and size of white muscle fibres; however, the functional significance of such a finding is not apparent. At 15-20 cm BL we found no differences in fast-start performance.  That said, significant changes in the  proportion of muscle fibre-types have been reported in other species (table 2.5). Table 2.5 provides a summary of the effects GH on muscle fibre-type proportions, including studies that used comparable methods.  GH injections  yielded an increase in the proportion of type IIb muscle fibres in the semispinalis of pigs, and the semitendinosus of cows (see table 2.5).  By contrast, GH  injections yielded an increase in type I muscle fibres in the EDL of rats. GHtransgenesis (for GH upregulation) also yielded an increase in type I muscle fibres in the soleus, plantaris, and gastrocnemius of mice and in the axial muscle of fish. Finally, the absence of a functional GHR did not change muscle fibretype proportions of mice. Among pigs, cows, and rats the increases in muscle fibre-type proportions, found within a given species, were not observed within all muscles (table 2.5). Among pigs and cows, affected muscles realized an increase in type IIb proportions, in contrast to rats, mice, and fish wherein affected muscles realized increases in the proportions of type 1 muscle fibres. Interestingly, GH was injected into pigs, cows, and rats as opposed to having been induced via transgenesis among mice and fish.  35  Table 2.5  36  Table 2.5 This table summarizes the effect of GH treatment on changes in muscle fibretype proportions from all available published data, which used comparable and  37  appropriate staining methods – ie: studies that froze muscle sections in liquid nitrogen directly were not included, as such a procedure can induce changes in muscle cell morphology. We have also not included studies considering aging or clinical disease states such as hypopituitarism, GH deficiency, etc. as they might be associated with other affectors of muscle physiology. We have only considered supraphysiological levels of GH as induced by transgenesis resulting in increased GH production or overexpression within skeletal muscle, externally derived GH (such as via injections) given to normal or wild-type individuals, and targeted GH receptor (GHR) lesions (as opposed to systemic GH insensitivity), all of which may address the effects of GH more specifically. Moreover, we have only considered changes in the proportion of muscle fibre-types among species in which GH upregulates muscle growth (and for which data was available): cows and pigs (Etherton and Bauman, 1998), rats, mice, and humans (Florini et al., 1996), and fish (Rescan, 2005). Interestingly, GH does not regulate muscle growth among guinea pigs (Baumman, 1997) and may even negatively regulate such growth among chickens (Vasilatos-Younken et al., 2000), both of which have been excluded. Column 2 indicates whether the associated study used age or size-matched controls. Column 3 indicates the size or age at which treatment was initiated for non-transgenic GH augmentation (among humans, pigs, and cows) and the sample size and age for GH-transgenic fish. Column 7 indicates the muscle fibre-types that differed significantly from their proportions among wild-type or saline-controls (↑ for significant increases). Abbreviations: i.v.= intravenous, i.m.= intramuscular, long. dorsi= longissimus dorsi, GHR= growth hormone receptor, glut. med.= gluteus medius, semitend.= semitendinosus, s.c.= subcutaneous, and vastus intermed.= vastus intermedius.  Changes in muscle fibre-type proportions were only observed across all muscles sampled among mice (fish are excluded from this comparison as only axial muscle was sampled in Hill et al., 2000). The reasons for such differences are not apparent. A number of signal transduction pathways are affected by GH, in addition to those responsible for muscle growth (Kopchick and Andry, 2000; Kopchick et al., 1999; Florini et al., 1996), which may not exhibit homogeneity from one species to another (Etherton and Bauman, 1998). Furthermore, among  38  studies using rats, mice, and fish (table 2.5), all subjects were fed ad libitum or to satiation, while GH-treated pigs and cows were fed less than their wild or untreated conspecifics and the GH-treated subjects still realized accelerated growth. GH significantly effects liver metabolism (Costa et al., 1998; Rise et al., 2006; Raven et al., 2008, Devlin et al., 2009). Changes in liver metabolism may influence or alter substrate availability (Valera et al., 1993; Kopchick et al., 1999; Gibney et al., 2007), and perhaps may therefore induce changes in muscle-fibre type proportions (though this has not been demonstrated experimentally). Moreover, the differences found across species and muscles may also be attributed to any one of the following affectors of GH functioning, in which differences between the studies were also apparent (see table 2.5):  GH  administration protocol, age, mass, breed, species, starting proportions of a given fibre-type within a given muscle, and diet. Muscle contractile properties are an additional affector of fast-start performance and one such study has considered the effects of GH on subjects administered GH not in an aged or clinical diseased state. Kim et al., (2005) found that the EDL of 7 week-old mice, which had been injected with 3mg (kg*day)-1 for 3 weeks, of recombinant bovine GH, did not differ in specific (per cross-sectional area) twitch or tetanic force. In addition to GH, GH-transgenic coho salmon also exhibit elevated levels of IGF-I within their blood plasma as well as increased expression (of IGF-I) within their epaxial muscle (Raven et al.,  39  2008); IGF-I also upregulates muscle growth (Florini et al., 1996; Velloso, 2008). Two studies have examined the effects of IGF-1 on the muscle contractile properties of subjects administered IGF-1, not in an aged or clinical diseased state, and obtained conflicting results. Anderson et al. (2006) injected IGF-1 into the superior rectus (extraocular) muscle of rabbits, which realized an increase in specific twitch and tetanic force one week later. In contrast, 3 month-old mice transgenic for IGF-I overexpression exhibited no change in specific tetanic force from the EDL (Musarò et al., 2001). In spite of some of the contrasting findings concerning the effects of GH (and IGF-1) on muscle fibre-type proportions and muscle contractile properties, we found no differences in fast-start performance between all three groups. Therefore, neither GH upregulation (as evidenced by GHr) nor GH upregulation and accelerated growth (as evidenced by GHf) affected the average of short-term behavior, muscle contractile properties, excitation contraction coupling, or musculoskeletal linking among GH-transgenic Coho salmon.  Ability to evade prospective predation Predicting the prospective ecological fitness of GH-transgenic fish remains one of the challenges regarding the use of such fish in aquaculture, with unknown potential consequences should such species escape into the natural environment (Devlin et al., 2006). Swimming performance studies, including the present study on fast-starts, as well as previous work on Ucrit, suggest that GH-  40  transgenic Coho salmon would be equally capable of evading predatory strikes but might under-perform during long-distance migrations (Farrell et al., 1997; Lee et al., 2003) relative to their wild conspecifics. Regarding other factors determining ecological fitness, foraging and predator avoidance behaviors, hatchery-reared GH-transgenic Atlantic salmon and Brown trout exhibited an increased propensity for foraging (Abrahams and Sutterlin, 1999; Johnsson and Björnsson, 2001, respectively). Additionally GHtransgenic Atlantic salmon exhibited decreased antipredator behavior when in the presence of both a predator and food pellets (Abrahams and Sutterlin, 1999). Moreover, Sundström et al. (2007a) found that GHf Coho salmon exhibited a lower propensity for schooling following simulated predator attacks, which may further suggest an increased susceptibility to predator attacks.  However,  experiments using hatchery-reared fish, conducted under laboratory conditions, may not accurately predict outcomes in a natural setting. In fact, Sundström et al. (2007b) investigated the effects of juvenile hatchery-raised GH-transgenic and wild coho salmon living within the same stimulated natural stream environment and found that GH-transgenic Coho salmon exhibited a reduced growth advantage relative to GH-transgenic Coho salmon fed to satiation under hatchery conditions. The results concerning changes in foraging and predator avoidance behavior suggest that while GH-transgenic fish will realize increased growth rates, albeit less than has been demonstrated in hatchery environments, they may also realize increased exposure to predators.  41  REFERENCES Abrahams, M.V. and A. Sutterlin. 1999. The foraging and antipredator behavior of growth-enhanced transgenic Atlantic salmon.  Animal Behavior, 58:  933-42. Anderson, B.C., S.P. Christiansen, S. Grandt, R.W. Grange, and L.K. McLoon. 2006.  Increased Extraocular Muscle Strength with Direct Injection of  Investigative Ophthalmology and Visual  Insulin-like Growth Factor-I.  Science, 47(6): 2461-7. Aroniadou-Anderjaska, V., P.W.R. Lennon, J. Gilloteaux.  1996.  Effects of  exogenous growth hormone on skeletal muscle of young female rats.  Tissue and Cell, 28(6): 719-24. Ayling, C.M., J.M. Zanelli, B.M. Moreland, and D. Schulster. 1992. Effect of Human Growth Hormone Injection on the Fibre Type Composition and Metabolic  Activity  in  a  Skeletal  Muscle  From  Normal  and  Hypophysectomized Rats. Growth Regulation, 2: 133-43. Bainbridge, R. 1960. Speed and stamina in three fish. Journal of Experimental  Biology, 37: 129-53. Baumman, G. 1997. Editorial: Growth without a Pituitary? – Lessons from the Guinea Pig. Endocrinology, 138(9): 3575-6. Billerbeck, J.M., T.E. Lankford, and D.O. Conover. 2005. Evolution of intrinsic growth and energy acquisition rates.  I.  Trade-offs with swimming  performance in Menidia menidia. Evolution, 55(9): 1863-72.  42  Carani, F.R., P.S.A. Moreira, R.F. Carvalho, C.R. Padovani, L.A.L. Chardulo, and M.D.P. Silva. 2006. Histochemistry and growth characteristics of bovine  semitendinosus muscle exposed to recombinant bovine somatotropin (rbST). Brazilian Journal of Morphological Science, 23(2): 263-70. Chen, L., P.K. Lund, S.B. Burgess, B.E. Rudisch, and D.L. McIlwain.  1997.  Growth Hormone, Insulin-like Growth Factor I, and Motoneuron Size.  Journal of Neurobiology, 32: 202-12. Costa, C., G. Solanes, J. Visa, and F. Bosch.  1998.  Transgenic rabbits  overexpressing growth hormone develop acromegaly and diabetes mellitus. Journal of the Federation of American Societies for Experimental  Biology, 12: 1455-60. Coughlin, D.J. 2002. Aerobic muscle function during steady swimming in fish.  Fish and Fisheries, 3: 68-78. Deitch, E.J, G.L. Fletcher, L.H., L. H. Petersen, I. A. S. F. Costa, M. A. Shears, W. R. Driedzic, and A. K. Gamperl.  2006.  Cardiorespiratory  modifications, and limitations, in post-smolt growth hormone transgenic Atlantic salmon Salmo salar. Journal of Experimental Biology, 209: 131025. Devlin, R.H., T.Y. Yesaki, C.A. Biagi, E.M. Donaldson, P. Swanson, and W.-K. Chan. 1994. Extraordinary salmon growth. Nature, 371: 209-10. Devlin, R.H., C.A. Biagi, T.Y. Yesaki, D.E. Smailus, and J.C. Byatt.  2001.  Growth of domesticated transgenic fish. Nature, 409: 781-2.  43  Devlin, R.H., C.A. Biagi, and T.Y. Yesaki. 2004. Growth, viability and genetic characteristics of GH transgenic coho salmon strains. Aquaculture, 60732. Devlin, R.H., L.F. Sundström, and W.M. Muir. 2006. Interface of biotechnology and ecology for environmental risk assessments of transgenic fish.  Trends in Biotechnology, 24(2): 89-97. Devlin, R.H., D. Sakhrani, W.E. Tymchuk, M.L. Rise, and B. Goh.  2009.  Domestication and growth hormone transgenesis cause similar changes in gene expression in coho salmon (Oncohynchus kisutch). Proceedings of  the National Academy of Sciences, 106(9): 3047-52. Dobson, G.P., W.S. Parkhouse, and P.W. Hochachka.  1987.  Regulation of  anaerobic ATP-generating pathways in trout fast-twitch skeletal muscle.  American Journal of Physiology: Regulatory Integrative and Comparative Physiology, 253: R186-94. Domenici, P. and R.W. Blake. 1993. The effect of size on the kinematics and performance of angelfish (Pterophyllum eimekei) escape responses.  Canadian Journal of Zoology, 71: 2319-26. Domenici, P. and R.W. Blake. 1997. The kinematics and performance of fish fast-start swimming. Journal of Experimental Biology, 200: 1165-78. Domenici, P. 2003. Habitat, body design, and swimming performance of fish.  Vertebrate Biomechanics and Evolution. Eds. V.L. Bels, J.P. Gasc and A. Casinos. BIOS Scientific Publishers Ltd, Oxford. Pg. 137-60.  44  Du, S.J., Z. Gong, G.L. Fletcher, M.A. Shears, M.J. King, D.R. Idler, and C.L. Hew. 1992. Growth enhancement in transgenic Atlantic salmon by use of an “all fish” chimeric growth hormone gene construct. Biotechnology, 10: 176-81. Dudley, G.A. and R. Portanova. 1987. Histochemical Characteristics of Soleus Muscle in hGH Transgenic Mice (42561). Proceedings of the Society for  Experimental Biology and Medicine, 185: 403-8. Dytham, C. 2003. Choosing and Using Statistics: A biologist's guide. Blackwell Publishing, Malden, MA, pp. 195-8. Eaton, R.C., R.K.K. Lee, and M.B. Foreman. 2001. The Mauthner cell and other identified neurons of the brainstem escape network of fish. Progress in  Neurobiology, 63: 467-85. Etherton, T.D. and D.E. Bauman. 1998. Biology of Somatotropin in Growth and Lactation of Domestic Animals. Physiological Reviews, 78(3): 745-61. Farrell, A.P., W. Bennett, and R.H. Devlin. 1997. Growth-enhanced transgenic salmon can be inferior swimmers. Canadian Journal of Zoology, 75: 3357. Florini, J.R., D.Z. Ewton, and S.A. Coolican. 1996. Growth Hormone and the Insulin-Like Growth Factor System in Myogenesis. Endocrine Reviews, 17(6): 481-517. Gemballa, S. and F. Vogel. 2002. Spatial arrangement of white muscle fibers and myoseptal tendons in fishes.  Comparative Biochemistry and  45  Physiology Part A, 133: 1013-37. Gibney, J., M-L. Healy, and P.H. Sönksen. 2007. The Growth Hormone/InsulinLike Growth Factor-I Axis in Exercise and Sport.  Endocrine Reviews,  28(6): 603-24. Goldbogen, J.A., R.E. Shadwick, D.S. Fudge, and J.M. Gosline. 2005. Fast-start muscle dynamics in the rainbow trout Oncorhynchus mykiss: phase relationship of white muscle shortening and body curvature. Journal of  Experimental Biology, 208: 929-38. Goolish, E.M. 1991. Aerobic and anaerobic scaling in fish. Biological Reviews, 66: 33-56. Hale, M.E., J.H. Long, M.J. McHenry, and M.W. Westneat. 2002. Evolution of behavior and neural control of the fast-start escape response. Evolution, 56(5): 993-1007. Harvey, S. and K. Hull. 2003. Neural growth hormone. Journal of Molecular  Neuroscience, 20(1): 1-13. Hill, J.A., A. Kiessling, and R.H. Devlin. 2000. Coho salmon (Oncorhynchus  kisutch) transgenice for a growth hormone gene construct exhibit increased rates of muscle hyperplasia and detectable levels of differential gene expression. Canadian Journal of Fish and Aquatic Sciences, 57: 939-50. James, R.S. and I.A. Johnston. 1998. The scaling of muscle performance during escape responses in the fish Myoxocephalus scorpius L. Journal of  46  Experimental Biology, 201:913-23. Jayne, B.C. and G.V. Lauder.  1993.  Red and white muscle activity and  kinematics of the escape response of the bluegill sunfish during swimming. Journal of Comparative Physiology A, 173: 495-508. Johnston, I.A. 2001. Genetic and environmental determinants of muscle growth patterns.  Muscle Growth and Development. Fish Physiology Vol. 18.  Eds. I.A. Johnston. Academic Press. Pg. 141-86. Johnsson, J.I. and B.TH. Björnsson.  2001.  Growth-enhanced fish can be  competitive in the wild. Functional Ecology, 15: 654-9. Johnston, I.A., W. Davison, and G. Goldspink. 1977. Energy metabolism of carp swimming muscles. Journal of Comparative Physiology B, 114 (2): 20316. Kim, H., E. Barton, N. Muja, S. Yahar, P. Pennisi, and D. LeRoith. 2005. Intact Insulin and Insulin-Like Growth Factor-I Receptor Signallin Is Required for Growth Hormone Effects on Skeletal Muscle Growth and Function in Vivo.  Endocrinology, 146(4): 1772-9. Kopchick, J.J., L.L. Bellush, and K.T. Coschigano. 1999. Transgenic models of growth hormone action. Annual Reviews of Nutrition, 19: 437-61. Kopchick, J.J. and J.M. Andry. 2000. Minireview – Growth Hormone (GH), GH Receptor, and Signal Transduction. Molecular Genetics and Metabolism, 71: 293-314. Lee, C.G., R.H. Devlin, and A.P. Farrell. 2003. Swimming performance, oxygen  47  consumption and excess post-exercise oxygen consumption in adult transgenic and ocean-ranched coho salmon. Journal of Fish Biology, 62: 753-66. Lefaucheur, L., A. Missohou, P. Ecolan, G. Monin, and M. Bonneau.  1992.  Performance, Plasma Hormones, Histochemical and Biochemical Muscle Traits, and Meat Quality of Pigs Administered Exogenous Somatotropin Between 30 or 60 Kilograms and 100 Kilograms Body Weight. Journal of  Animal Science, 70: 3401-11. Long, J.H., B. Adcock, and R.G. Root. tendons in undulating fish:  2002.  Force transmission via axial  a dynamic analysis.  Comparative  Biochemistry and Physiology Part A, 133: 911-29. Lundvall, D., R. Svanbäck, L. Persson, and P. Byström. 1999. Size-dependent predation in piscivores: interactions between predator foraging and prey avoidance abilities. Canadian Journal of Fish and Aquatic Sciences, 56: 1285-92. McKenzie, D.J., S. Wong, D.J. Randall, S. Egginton, E.W. Taylor, and A.P. Farrell. 2004. The effects of sustained exercise and hypoxia upon oxygen tensions in the red muscle of rainbow trout.  Journal of Experimental  Biology, 207: 3629-37. Musarò, A., K. McCullagh, A. Paul, L. Houghton, G. Dobrowolny, M. Molinaro, E.R. Barton, H.L. Sweeney, and N. Rosenthal.  2001.  Localized Igf-1  transgene expression sustains hypertrophy and regeneration in senescent  48  skeletal muscle. Nature Genetics, 27: 195-200. Nordgarden, U., P.G. Fjelldal, T. Hansen, B.T. Björnsson, and A. Wargelius. 2006. Growth hormone and insulin-like growth factor-I act together and independently when regulating growth in vertebral and muscle tissue of atlantic salmon postsmolts.  General and Comparative Endocrinology,  149: 253-60. Ohlsson, C., B. Bengtsson, O.G.P. Isaksson, T.T. Andreassen and M.C. Slootweg. 1998. Growth Hormone and Bone. Endocrine Reviews, 19(1): 55-79. Oksbjerg, N., J.S. Petersen, M.T. Sórensen, P. Henckel, N. Agergaard, C. Bejerholm, and E. Erlandsen. 1995. The Influence of Porcine Growth Hormone on Muscle Fibre Characteristics, Metabolic Potential and Meat Quality. Meat Science, 39: 375-85. Ostenfeld, T.H., E. McLean, and R.H. Devlin. 1998. Transgenesis changes body and head shape in Pacific salmon. Journal of Fish Biology, 52: 850-4. Rescan, P.Y. 2005. Muscle growth patterns and regulation during fish ontogeny.  General and Comparative Endocrinology, 142: 111-6. Raven, P.A., M. Uh, D. Sakhani, B.R. Beckman, K. Cooper, J. Pinter, E.H. Leder, J. Silverstein, and R.H. Devlin.  2008.  Endocrine effects of growth  hormone overexpression in transgenic coho salmon.  General and  Comparative Endocrinology, 159: 26-37. Richards, J.G., G.J.F. Heigenhauser, and C.M. Wood.  2002a.  Glycogen  49  phosphorylase and pyruvate dehydrogenase transformation in white muscle of trout during high-intensity exercise.  American Journal of  Physiology: Regulatory Integrative and Comparative Physiology, 282: R828-36. Richards, J.G., A.J. Mercado, C.A. Clayton, G.J.F. Heigenhauser, and C.M. Wood. 2002b. Substarte utilization during graded aerobic exercise in rainbow trout. Journal of Experimental Biology, 206: 2067-77. Rise, M.L., S.E. Douglas, D. Sakhrani, J. Williams, K.V. Ewart, M. Rise, W.S. Davidson, B.F. Koop, and R.H. Devlin.  2006.  Multiple microarray  platforms utilized for hepatic gene expression profiling of GH transgenic coho salmon with and without ration restriction.  Journal of Molecular  Endocrinology, 37: 259-82. Royle, N.J., N.B. Metcalfe, and J. Lindström. 2006. Sexual selection, growth compensation and fast-start swimming performance in Green Swordtails,  Xiphophorus helleri. Functional Ecology, 20: 662-9. Schuenke, M.D., J.J. Kopchick, R.S. Hikida, W.J. Kraemer, and R.S. Staron. 2008. Effects of Growth Hormone overexpression vs. growth hormone receptor gene disruption on mouse hindlimb muscle fibre type composition. Growth Hormone and IGF Research, 18: 479-86. Shadwick, R.E. and S. Gemballa.  2006.  Structure, kinematics, and muscle  dynamics in undulatory swimming. Fish Biomechanics. Fish Physiology Volume 23. Eds. R.E. Shadwick and G.V. Lauder. Academic Press. San  50  Diego, CA. Pg. 241-80. Sørensen, M.T., N. Oksbjerg, N. Agergaard, and J.S. Petersen. 1996. Tissue Deposition Rates in Relation to Muscle Fibre and Fat Cell Characteristics in Lean Female Pigs (Sus scrofa) Following Treatment with Porcine Growth Hormone (pGH). Comparative Biochemical Physiology, 113A(2): 91-6. Sotiropoulos, A., M. Ohanna, C. Kedzia, R.K. Menon, J.J. Kopchick, P.A. Kelly, and M. Pende. 2006. Growth hormone promotes skeletal muscle cell fusion  independent  of  insulin-like  growth  factor  1  up-regulation.  Proceedings of the National Academy of Science, 103(19): 7315-20. Stevens, E.D. A. Sutterlin, and T. Cook. 1998. Respiratory metabolism and swimming performance in growth hormone transgenic Atlantic salmon.  Canadian Journal of Fisheries and Aquatic Sciences, 55: 2028-35. Summers, A.P. and J.H. Long. 2006. Skin and bones, sinew and gristle: the mechanical behavior of fish skeletal tissues. Fish Biomechanics. Fish Physiology Volume 23. eds. R. E. Shadwick and G.V. Lauder. Academic Press. San Diego, CA. Pg. 141-77. Sundström, L.F., M. Lõhmus, J.I. Johnsson, and R.H. Devlin. 2007a. Dispersal Potential is Affect by Growth-Hormone Transgenesis in Coho Salmon (Oncorhynchus kisutch). Ethology, 113: 403-10. Sundström, L.F., M. Lõhmus, W.E. Tymchuk, and R.H. Devlin. 2007b. Geneenvironment interactions influence ecological consequences of transgenic  51  animals.  Proceedings of the National Academy of Sciences, 104(10):  3889-94. Syme, D.A. 2006. Functional properties of skeletal muscle. Fish Biomechanics. Fish Physiology Volume 23. eds. R. E. Shadwick and G.V. Lauder. Academic Press. San Diego, CA. Pg. 179-211. Tytell, E.D. and G.V. Lauder. 2002. The C-start escape response of Polypterus  senegalus: bilateral muscle activity and variation during stage 1 and 2. Journal of Experimental Biology, 205: 2591-2603. Valera, A., J.E. Rodriguez-Gil, J.S. Yun, M.M. Mcgrane, R.W. Hanson, and F. Bosch.  1993.  Glucose metabolism in transgenic mice containing a  chimeric P-enolpyruvate carboxykinase/bovine growth hormone gene.  Journal of the Federation of American Societies for Experimental Biology, 7: 791-800. Vasilatos-Younken, R., Y. Zhou, X. Wang, J.P. McMurtry, R.W. Rosebrough, E. Decuypere, N. Buys, V.M. Darras, S. Van der Geyten, and F. Tomas. 2000.  Altered chicken thyroid hormone metabolism with chronic GH  enhancement in vivo: consequences for skeletal muscle growth. Journal  of Endocrinology, 166: 609-20. Velloso, C.P. 2008. Regulation of muscle mass by growth hormone and IFG-1.  British Journal of Pharmacology, 154: 557-68. Vestergaard, M., S. Purup, P. Henckel, E. Tonner, D.J. Flint, L.R. Jensen, and K. Sejrsen.  1995.  Effects of Growth Hormone and Ovariectomy on  52  Performance Serum Hormones, Insulin-Like Growth Factor-Binding Proteins, and Muscle Fiber Properties of Prepubertal Friesian Heifers.  Journal of Animal Science, 73: 3574-84. Wakeling, J.M. 2006. Fast-start mechanics. Fish Biomechanics. Fish Physiology Volume 23. eds. R.E. Shadwick and G.V. Lauder. Academic Press. San Diego, CA. Pg. 333-63. Walker, J.A., C.K. Ghalambor, O.L. Griset, D. McKenney, and D.N. Reznick. 2005.  Do faster starts increase the probability of evading predators?  Functional Ecology, 19: 808-15. P.W. Webb. 1976. The effect of size on the fast-start performance of Rainbow trout Salmo gairdneri, and a consideration of piscivorous predator-prey interactions. Journal of Experimental Biology, 65: 157-77. P.W. Webb. 1978. Fast-start performance and body form in seven species of teleost fish. Journal of Experimental Biology, 74: 211-26.  53  CHAPTER 3: SPRINTING2  INTRODUCTION Coho salmon transgenic for growth hormone (GH) upregulation realize dramatic acceleration of overall body and muscle growth (Devlin et al., 1994), dependent upon a sufficient supply of rations relative to their wild conspecifics (Raven et al., 2006; Oakes et al., 2007). Accelerated muscle growth may be beneficial for aquaculture because of decreased product turnaround times; however, large-scale aquaculture would also raise the probability of such fish escaping into a natural habitat. Therefore, it is necessary to understand the differences between GH-transgenic and wild fish in order to make inferences regarding their prospective impact on such habitats or ecosystems (Pew Initiative, 2003; Devlin et al., 2006; USFDA, 2009). We were interested in assessing the prospective ability of GH-transgenic coho salmon to evade chasing predators, to chase prey and to successfully complete spawning migrations. Previous work has shown that GH-transgenic salmon have a lower Ucrit than their wild conspecifics throughout their life history (Farrell et al., 1997; Stevens et al., 1998; Lee et al., 2003; Deitch et al., 2006), which suggests that they might exhibit lower performance during long-distance migrations if data from smaller fish can be extrapolated to later life history stages.  2  A version of this chapter will be submitted for publication. Dimoulas, P.M., R.H. Devlin, and R.E. Shadwick. The sprinting ability of GH-transgenic Coho salmon.  54  In fact, sub-Ucrit speeds are used by Salmonids during spawning migrations (Hinch et al., 2002); however, sprinting is used to negotiate the challenging reaches such as waterfalls, which are typically encountered during such migrations (Hinch et al., 2002, Brown and Geist, 2002; Laurtizen et al., 2005). Furthermore, sprinting may even be used during predator-prey interactions (Savitz & Bardygula-Nonn, 1997).  Thus, the present study considered the  differences in sprinting performance between GH-transgenic and wild coho salmon.  Moreover, because sustained swimming is powered by red lateral  muscle, which has a relatively high capacity to support ATP turnover, a mechanistic explanation why GH transgenic salmon have a lower Ucrit remains elusive. In this regard, we also sought to determine more specifically the effects of GH upregulation on muscle functioning of fish in vivo. Sprinting is characterized as maximal and unsteady or burst and glide swimming (Bainbridge, 1960; Nelson et al., 2008).  Sprinting encompasses  swimming velocities faster than Ucrit (Goolish, 1991; Katz et al., 1999; Reidy et al., 2008). No study has compared sprinting with fast-start escape velocities, though rainbow trout are reported to realize higher velocities 1s after acceleration rest than during fast-starts, which last < 100ms (Webb, 1976). Initially (during the first 10s from 210g rainbow trout), sprinting is fueled by ATP and CrP stores and subsequently via glycolysis within the white muscle (Richards et al., 2002a). In addition to the affectors of fast-start performance (musculoskeletal linking, shortterm or twitch muscle contractile properties and excitation-contraction coupling of  55  their white lateral muscle, body shape, and short-term behavior), sprinting is also affected by the capacity of the white lateral muscle for continual maximal power output and to support ATP turnover. Because white muscle is more limited in its capacity to support ATP turnover than red lateral muscle (Dobson et al., 1987; Goolish et al., 1991), sprinting offers an alternative approach to examine the effect of GH upregulation on the exhaustive exercise capacity of GH-transgenic Coho salmon. We examined the sustained capacity of white muscle by comparing the kinematics of sprinting to exhaustion between length-matched GH-transgenic and wild coho salmon.  The objective of these studies was to determine more  specifically the effects of GH-upregulation on sustained capacity of white muscle, as well as the prospective ability of GH-transgenic Coho salmon to evade chasing predators, to chase prey, and to successfully complete spawning migrations. We used three populations of coho salmon: GH-transgenic fed to satiation (GHf), wild fed to satiation, and GH-transgenic ration-restricted or par fed according to the amount of feed consumed by the wild Coho (GHr) (Devlin et al., 2001b). The latter population, GHr, enabled us to control for potential growthrate-related differences because GHf are younger than length-matched wild and GHr (Devlin et al., 2004). We sought length-matched cohorts because muscle contractile properties vary with body length (BL) (James and Johnston, 1998). Additionally, to account for potential age-related differences we examined sprinting at two different body lengths.  56  There are few studies upon which we may base a hypothesis. In fact, the present study is the first to address the kinematics of exhaustive sprinting in any fish.  Two studies that examined muscle enzyme activities between length-  matched GHf and wild Coho salmon found that the GHf have higher aerobic and anaerobic enzyme activities within their white muscle at 10 cm BL (Hill et al., 2000) and at 17 cm BL (Blier et al., 2002). Therefore, we hypothesized that higher muscle enzyme activity might enable juvenile GHf to outperform their conspecifics. In the absence of any such studies that used larger adult GHf, concerning the performance of their white muscle, we offer the following null hypothesis: larger, length-matched GH-transgenic and wild Coho salmon will perform similarly during sprinting. Our results showed that juveniles exhibited similar sprinting velocities but GHf could sprint longer before exhaustion than their conspecifics. Adults, which unfortunately were not length-matched, realized similar times to exhaustion, and longer GHf exhibited lower velocities than smaller wild Coho salmon.  57  METHODS Transgenic coho salmon were produced and all fish were raised according to Devlin et al. (2004) at the Department of Fisheries and Oceans Canada (DFO) laboratory in West Vancouver. Pre-smolts (parr), juveniles from each group were housed in a 200-L fiberglass, flow-through aquarium with fresh well water maintained at 11±1 oC (during our experiments), and kept under natural light conditions. Post-smolts, adults were housed in 500-L fiberglass aquaria with flow-through seawater obtained off the coast from the DFO West Vancouver facility at 11±1 oC (during our experiments, though temperatures vary seasonal 513 oC), and kept under natural light conditions. GHf and wild were fed to satiation using a commercial salmon diet (Skretting Canada Ltd., Vancouver, BC) though GHr (which do not thrive post-smolt) were par-fed (averaged per individual) according to the amount of feed consumed by the wild coho salmon.  For  experiments conducted during the fall that used juveniles, wild and GHr were 18 months old while GHf were 6 months old. For experiments conducted during the summer that used adults, wild coho were 26 months old while GHf were 14 months old.  Sprinting Fall experiments were conducted between 3 – 17 November 2007, using juvenile male GHf, GHr, and wild. Fish were sexed rather females were identified via a PCR test for a Y-chromosome specific sequence (Devlin et al., 2001a)  58  using tissue obtained from their adipose fin.  Blind trials were conducted on  individuals selected at random with regard to their group. Individuals were netted and transferred to a clear-bottomed, fiberglass, oval aquarium (1.2m x 0.6m and 0.2m water depth) with flow-through, oxygenated well water kept at 11-13oC. Once transferred, fish were subsequently allowed 3 minutes to acclimate to the experimental aquarium. Blind trials were also conducted using adults GHf female and wild male coho salmon during the summer, on 9 and 10 July 2007. Those individuals were netted and transferred to a donut-shaped, fiberglass aquarium (0.1m radius, 0.05m wide track, and 0.3m water depth) with flow-through, oxygenated seawater kept between 12-14oC. Once transferred, they were also allowed 3 minutes to acclimate to the experimental aquarium. Once acclimated, all individuals were induced to sprint via continuous manual tail-grabs (Richards et al., 2002a) until they failed to respond to 5 consecutive tail-grabs, at which point they were considered exhausted (though some individuals underwent only 20s of sprinting, and were not exhausted as defined previously). However, all adults were sprinted to exhaustion. The difference between the two experimental tanks, used for juveniles and adults, complicated our analysis (yielding an apparent discontinuity in the results obtained between these cohorts, which will be discussed later). We were only able to use female adult GHf, and all other cohorts were males due to the requirements of other studies on GH-transgenic and wild Coho salmon. That said, sex-related differences have not been found during sprinting (Reidy et al.,  59  2000).  As well, allowing only 3 minutes of acclimation time having netted,  bucketed, and transferred fish into the experimental tank is less than ideal for examining exhaustive exercise (sprinting) in fish. It is beyond doubt that our method of capture and transfer might affect resting fish by lowering resting metabolic substrate levels, and that this effect could differ between the groups. We confirmed as much by preliminary assessment of anaerobic substrate content (CrP, ATP, and lactate) from red and white lateral muscle before and after capture and transfer among all 3 groups (biochemistry methods to follow, see below).  However, due to logistical reasons (time constraints) we were  unsure whether individual fish could realize a full return to their resting state. Therefore, we choose to allow individual fish only 3 minutes to acclimate to the experimental conditions prior to the commencement of sprinting. Thereby limiting the potential for differences in recovery mechanisms between the groups, which were known to be different (Hill et al., 2000; Blier et al., 2002). The two latter points will not be discussed further. Video recordings at 60 Hz were obtained for all individuals sprinted (for 20s and to exhaustion) with a Canon XL1 digital camera (Canon U.S.A., Inc., Lake Success, NY). Among juveniles images were obtained from underneath the experimental tank via a mirror angled at 45o, and among adults from overhead. Instantaneous velocity (calculated by the linear distance traveled between successive frames) was obtained by tracing the nose (fig. 3.1a), and tail-beat frequency was obtained by counting each tail-beat (only among juveniles; fig.  60  3.3a) frame-by-frame using NIH-Image for Mac Os 9 (http://rsbweb.nih.gov/ij/). This was done for all movements of the head and complete tail-beats so long as their movement was not affected by a given tail-grab.  For each individual,  instantaneous velocity was averaged over every successive 5s interval, from which we obtained a linear regression as a simple way to summarize this relationship and offer a means of comparison (fig 3.1b).  The slope and median  velocity (calculated from a given regression, at half the time to reach exhaustion) was used for subsequent analysis.  We did not use the Y-intercept as this  measurement was correlated with the slope (fig. 3.2a), while median velocity was not correlated with the slope (table 3.1 and fig 3.2b). Maximum velocity was obtained by identifying the 100ms period during which the average velocity was at a maximum, per individual. Lastly, instantaneous velocity was binned by tailbeat frequency for each frequency with at least 5 data points (fig. 3.3b).  61  Instantaneous Velocity (BL s-1)  12  10  8  6  4  2  0  Average Instantaneous Velocity (BL s-1)  A  0  20  0  20  40  60  80  40  60  80  100  8  6  4  2  0  B  100  Time Sprinting (s)  Figures 3.1 A) Instantaneous velocity (BL s-1) plotted as a function of time sprinting (s), for one individual. B) The average instantaneous velocity (BL s-1) over every 5s interval until exhaustion from the preceding figure (fig. 3.1a), ± 1 s.e.m. The dotted line represents a linear regression fitted to the means (BL s-1 = 6.38 0.050 * (time), r2 = 0.716).  62  -1  Y-intercept (b, from fig. 3.1b) (BL s )  8  6  4  2  -0.10  A -1  Median Velocity (from fig. 3.1b) (BL s )  8  0.00  0.05  GHf fall GHf summer GHr fall Wild fall Wild summer  6  4  2  -0.10  B  -0.05  -0.05  0.00  0.05  Slope (change in ave. velocity over time)  Figures 3.2 Data points represent A) the y-intercept (b, obtained for each individual as fig. 1b) and B) the median velocity (calculated from linear regressions for each individual, from fig. 3.1b), both are plotted as a function of the slope (the change in average velocity, fig. 3.1b). Closed blue squares represent fall GHf, open blue squares represent summer GHf, closed green triangles represent fall GHr, close red circles represent fall wild, and opened red circles represent summer wild coho salmon for each graph.  63  8  -1  Average Velocity (BL s )  10  6  4  2  0 0  A  5  10  15  5  10  15  8  -1  Average Velocity (BL s )  10  6  4  2  0 0  B  Tail-beat Frequency (Hz)  Figures 3.3 The average velocity (BL s-1) plotted as a function of tail-beat frequency (Hz), from one individual. Each data point indicates A) the average velocity achieved over a given tail-beat and B) the average velocity binned by tail-beat frequency for each frequency with at least 5 data points, ± 1 standard deviation.  64  Source Y-intercept Slope Group Error Median Velocity Slope Group Error  d.f.  MS  F  P  1 4 27  8.425 6.883 0.462  18.24 14.90  < 0.001** < 0.001**  1 4 27  0.804 5.975 0.399  2.02 14.99  0.167 < 0.001**  * Indicates significance level < 0.05 ** Indicates significance level < 0.01 Table 3.1 ANCOVA results for Y-intercept and median velocity (from fig. 3.1b) among all cohorts with the slope, of average instantaneous velocity as a function of time, as a covariate tested among all individuals sprinted to exhaustion (table 3.2).  During the fall 2006, using length-matched juveniles from all 3 groups at approximately 15 cm BL, we performed a similar study (though the experimenter was aware of fish group identifications and no kinematics were obtained) during which we exhaustively sprinted fish. Following the same acclimation period fish were induced to sprint via manual tail-grabs until exhaustion. Once exhausted fish were removed from the experimental tank and stunned by a sharp blow to their head. Fish were subsequently sectioned in a sagittal plane at both cloaca and caudal peduncle. Then this entire section was placed between aluminum clamps, which were cooled in liquid nitrogen, and submerged in liquid nitrogen to freeze those sections. It took no more than 5 seconds to take the fish out of the water, section their axial muscle, and cool the section between the clamps.  65  Samples were lyophilized and stored at -80 oC prior to analysis. For resting samples, fish were induced in their holding tanks via benzocaine dissolved in 95% ethanol. The final concentration of benzocaine in the housing tanks was 0.05g/L. Lateral muscle was harvested (as described above) following a 20minute induction period. Red and white lyophilized lateral muscle tissue was analyzed separately for anaerobic substrate content (CrP, ATP, and lactate) via spectroscopic assays (as in Richards et al., 2002a).  Statistics Linear regressions, and one-way ANOVAs (Tukey HSD pair-wise comparison) were performed using Igor Pro software (version 5, Wavemetrics Inc., Lake Oswego, OR) for Macintosh Os 10. ANCOVAs, via the “general linear trend” option for ANOVA analysis, were performed using Minitab (Release 13.32, Minitab Inc., State College, PA,) on a Windows XP – based personal computer (as described in Dytham, 2003).  A t-test for significance of slopes was  performed as described in Glantz, “How to compare two regression lines” (1997).  P-values less than 0.05 were deemed statistically significant.  66  RESULTS While sprinting to exhaustion, individuals employed a range of speeds (fig. 3.1a).  We found that a linear regression offered a reasonable comparison  between cohorts and among individuals.  Table 3.2 summarizes the time to  exhaustion, BL, the change in average velocity over time (fig. 3.1b), and the median, as well as the maximum (length specific, BL s-1) velocity achieved for individuals sprinted to exhaustion. One-way ANOVAs were performed on each dataset within a given season (from table 3.2, e.g. time, BL, slope…) and are summarized in table 3.3.  Between juveniles, there were no significant  differences in BL or the change in average velocity over time. However, GHf took significantly longer to exhaust than the wild (fig. 3.4), and GHr exhibited a significantly higher median velocity than wild Coho salmon. Among adults, there were no significant differences in time to exhaustion or the (slope) change in average velocity over time; however, GHf were significantly longer and exhibited a significantly lower median and maximum velocity than wild Coho salmon. To summarize the kinematics among juveniles, we included the average velocity achieved at a given tail-beat frequency (fig. 3.3a and 3.3b), among all individuals, for each cohort (fig. 3.5).  One-way ANOVAs at each tail-beat  frequency revealed no statistical differences between the GHf, GHr, and wild Coho salmon from 3 to 15 Hz.  67  Season/ Group  Time (s)  BL (cm)  Fall Cohorts (juveniles) GHf 78 16.9 63 20.9 72 22.8 88 16.5 46 26.0 58 16.6 72 20.9 Average 68.14 20.08 S.E.M. 5.63 1.48 GHr  Average S.E.M. Wild Coho  Average S.E.M.  Slope  Median BL s-1  Max. BL s-1  r2  -0.063 -0.016 -0.064 -0.027 -0.060 -0.049 -0.008 -0.041 0.01  3.463 4.316 2.368 3.757 3.071 4.381 3.981 3.62 0.29  9.132 7.915 8.090 12.760 8.480 10.155 8.821 9.34 0.69  0.779 0.306 0.924 0.467 0.658 0.724 0.345  32 92 59 42 59 39 58 54.43 8.10  23.9 16.5 25.9 19.8 15.9 18.2 14.2 19.2 1.76  0.029 -0.050 -0.015 -0.015 -0.088 -0.053 -0.084 -0.039 0.017  4.752 4.060 3.085 4.361 3.773 5.017 4.424 4.21 0.26  8.239 11.830 7.129 7.780 11.171 10.843 12.820 9.97 0.91  0.540 0.716 0.376 0.151 0.881 0.625 0.790  31 54 33 56 40 59 43.43 5.12  24.7 27.2 21.7 20.5 17.8 24.3 22.71 1.5  -0.029 -0.065 -0.103 -0.080 -0.113 -0.045 -0.072 0.015  3.793 2.203 3.112 3.417 2.634 2.946 3.02 0.25  8.292 7.581 9.704 11.909 10.217 9.059 9.46 0.68  0.498 0.825 0.635 0.671 0.845 0.694  Table 3.2 (table continued on the following page)  68  Season/ Group  Time (s)  BL (cm)  Summer Cohorts (adults) GHf 129 52 60 47 115 50 39 42 46 48 120 42 90 45 Average 85.57 45.5 S.E.M. 15.23 1.56 Wild Coho  Average S.E.M.  72 92 80 73 49 33 55 64.86 8.24  33 33 31 37 34 36 35 34.14 0.83  Slope  Median BL s-1  Max. BL s-1  r2  0.003 -0.021 -0.003 -0.050 -0.062 -0.005 -0.022 -0.023 0.01  3.024 2.415 3.034 3.374 3.457 2.616 3.699 3.09 0.19  5.266 5.481 4.797 7.021 6.280 4.837 7.714 5.91 0.46  0.240 0.733 0.431 0.683 0.700 0.436 0.714  -0.034 -0.021 -0.028  4.258 5.519 5.673  8.527 11.079 11.294  0.767 0.588 0.517  -0.007 -0.106 -0.066 -0.044 0.016  6.581 4.860 5.566 5.41 0.35  10.439 10.482 9.532 10.23 0.46  0.120 0.788 0.638  Table 3.2 Individual and group comparisons of sprinting parameters. Included is the season during which experiments were conducted (fall for juveniles or summer for adults), group ID (GHf, GHr, or wild) with a numerical assignment for statistical ID, BL (cm), time to exhaustion (s), the slope (the change in average velocity over time, fig. 3.1b) of decreasing average velocity over time, the median velocity (BL s-1) (calculated from fig. 3.1b, at half the time to exhaustion), and the r2 of the linear regression (obtained from fig. 3.1b) for each individual sprinted to exhaustion. Averages (± s.e.m.) are also shown, for each cohort.  69  Season  Groups Tested  Time  BL  Slope  Median BL s-1  Maximum BL s-1  Fall  GHf & GHr  0.258  0.902  1.000  0.234  0.992  GHf & wild  0.023 *  0.450  0.242  0.250  0.996  GHr & wild  0.410  0.251  0.210  0.011 *  1.0  GHf & wild  0.221  < 0.001 **  0.243  < 0.001 **  < 0.001**  Summer  * Indicates significance level < 0.05 ** Indicates significance level < 0.01 Table 3.3 Seasonal and group comparisons of sprinting parameters. P-values, for one-way ANOVAs for individual data from table 3.2 are shown for BL, time (to exhaustion), slope (M), median and maximum velocity. Tests were run on data from the same season (fall or summer) according to group (GHf, GHr, or wild), as indicated.  70  Time to Exhaustion (s)  100  80  60  Wild GHf GHr  40  20 20  30  40  50  Body Length (cm)  Figure 3.4 The average time to exhaustion (s) ± 1 s.e.m. plotted as a function of BL (cm) ± 1 s.e.m. Red circles represent wild, blue squares represent GHf, and green triangles represent GHr (N= 7 per cohort).  71  -1  Average Velocity (BL s )  8  6  4  2  GHf GHr wild  0 5  10  15  Tail-beat Frequency (Hz)  Figure 3.5 The average velocity (BL s-1) binned by tail-beat frequency (Hz), for all individuals from the fall (based on procedure presented in figure 2), grouped by cohort, ± 1 s.e.m. Blue squares represent GHf (n= 10), green triangles represent GHr (n= 10), and red circles represent wild coho salmon (n= 10).  72  Net anaerobic substrate usage Among juveniles, we found significant differences in time to exhaustion between the 3 groups (ANOVA, P < 0.001): GHf took 143s ± 22.9 (1 s.e.m.), GHr took 70s ± 4.9, and wild took 73s ± 5.2 (n= 10 for each group; Post hoc tukey assessment for GHf versus GHr: P value = 0.002, GHf versus wild: P value = 0.003). We were able to assess substrate content from 3 individuals for each group. The average time to exhaustion for those 3 individuals differed: GHf 173s ± 47.2 (1 s.e.m.), GHr 75s ± 14.5, and wild 60s ± 11.4. We found no significant differences between the groups in terms of absolute content per condition for ATP, CrP, and lactate within the white lateral muscle (see figs. 3.9a, b, & c, respectively).  However, within each group we found significant  accumulation of lactate and significant depletion of CrP and ATP (table 3.4). We also assessed anaerobic substrate content (CrP, ATP, and lactate) in the red muscle of rested (n= 3, per group) and exhausted (n= 3, per group) GHf and wild coho salmon. We found a significant accumulation of lactate between rested and exhausted GHf (ANOVA, P = 0.036). All other results were not significant, and there were no differences between the substrate levels among the red muscle of GHf and wild Coho.  73  µmol ATP g-1 ww  4  • Wild • GHf • GHr  3  2  1  0  White at Rest  White Exhausted  A  -1  µmol CrP g ww  15  10  5  0  White at Rest  White Exhausted  White at Rest  White Exhausted  B  µmol lactate g-1 ww  40  30  20  10  0  C Figures 3.6 Bars (n= 3) represents the average substrate content: A) µmol ATP, B) µmol CrP, C) µmol lactate per gram of white muscle (wet weight) at rest and exhaustion, ± s.e.m. Red bars represent wild, blue bars represent GHf, and green bars represent GHr coho salmon from Fall 2006.  74  Comparison  Group(s)  CrP  ATP  Lacate  Within Resting  GHf vs. GHr  0.40  0.62  1.0  GHf vs. wild  0.99  0.70  1.0  GHr vs. wild  0.58  1.0  1.0  GHf vs. GHr  1.0  0.57  0.99  GHf vs. wild  0.99  0.49  0.72  GHr vs. wild  1.0  1.0  0.96  GHf  0.026*  < 0.001**  < 0.001**  GHr  < 0.001**  < 0.001**  < 0.001**  wild  0.026*  < 0.001**  < 0.001**  Within Exhaustion  Resting & Exhaustion  * Indicates significance level < 0.05 ** Indicates significance level < 0.01 Table 3.4 Anaerobic substrate usage compared between groups, and resting and exhaustion. P-values for one-way ANOVAs of substrate content (CrP, ATP, and lactate) within the white muscle. Tests were run between the groups at rest and exhaustion and within each group between rested and exhausted individuals.  Effects due to time to exhaustion To examine differences related to time to exhaustion, we performed ANCOVAs on BL, the (slope) change in average velocity over time, median and maximum velocity (from table 3.2) using time to exhaustion as a covariate (table 3.5). The associated figures for the change in average velocity over time, and median and maximum velocity have been presented as a function of time to exhaustion for each cohort (figs. 3.7a, b, and c). Among juveniles, there was no  75  significant effect of BL, slope, median or maximum velocity on time to exhaustion. Additionally, there was no significant effect of group on BL or the change in velocity over time, though there were significant differences in median velocity between juveniles. Wild Coho salmon exhibited lower median velocities than GHr (see tables 3.3 and 3.5). Between adult GHf and wild Coho salmon there was no significant effect of BL, median or maximum velocity with time to exhaustion; however, there was a significant effect of the (slope) change in average velocity over time with time to exhaustion (table 3.5). The associated linear regression has been presented (fig. 3.7a), which depicts how adults that took longer to exhaust, also exhibited a slower rate of decrease in average instantaneous  velocity  (than  individuals  that  exhausted  more  quickly).  Additionally, there was no significant effect of group on the change in average velocity over time, but there were significant differences in BL, median and maximum velocity between adults. The longer, GHf exhibited lower median and maximum length-specific velocities than the smaller and faster wild Coho salmon.  76  Source  d.f.  F  P  2.01 1.0  0.175 0.391  0.45 2.08  0.513 0.157  1.68 6.40  0.214 0.009**  3.71 1.0  0.072 0.388  3.3 447.32 9.93  0.33 45.03  0.576 < 0.001**  0.00544 0.0000726 0.000479  11.36 0.15  0.007** 0.705  0.0986 14.313 0.428  0.23 33.43  0.642 < 0.001**  1.403 46.418 1.156  1.21 40.16  0.296 < 0.001**  MS  GHf, GHr, & wild from fall (juveniles) Body Length Time 1 27.56 Group 2 13.65 Error 16 13.71 Slope Time 1 0.000523 Group 2 0.00243 Error 16 0.00117 Median Velocity Time 1 0.682 Group 2 2.60 Error 16 0.406 Maximum Velocity Time 1 10.975 Group 2 2.972 Error 16 GHf & wild from summer (adults) Body Length Time 1 Group 1 Error 11 Slope Time 1 Group 1 Error 10 Median Velocity Time 1 Group 1 Error 10 Maximum Velocity Time Group Error  1 1 10  * Indicates significance level < 0.05 ** Indicates significance level < 0.01 Table 3.5  77  Table 3.5 Comparison of sprinting parameters (from table 3.2) as a function of time to exhaustion. ANCOVA results for body length, slope, median and maximum velocity with time to exhaustion as a covariate, among juvenile GHf, GHr, and wild and adult GHf and wild coho salmon.  78  Slope (M, from fig. 3.1b)  0.05  0.00  -0.05  GHf, fall GHf, summer wild, fall wild, summer GHr, fall regression summer cohorts  -0.10  -0.15  A  40  60  80  100  40  60  80  100  40  60  80  100  120  Median Velocity (BL s-1)  7  6  5  4  3  2  B  120  Maximum Velocity (BL s-1)  14  12  10  8  6  4  C  120  Time to Exhaustion (s)  Figures 3.7  79  Figure 3.7 Calculated values obtained from fig. 3.1b, for each individual sprinted to exhaustion: A) the slope B) median velocity (BL s-1) and C) maximum velocity (BL s-1) are plotted as a function time to exhaustion, using the same x-axis. Closed blue squares represent fall GHf (n= 7), open blue squares represent summer GHf (n= 7), closed green triangles represent fall GHr (n= 7), close red circles represent fall wild (n= 6), and opened red circles represent summer wild coho salmon (n= 6) for each graph. The dotted line depicted in fig. 3.7a is represented by the linear regression: the change in velocity over time = (0.000713 * (time)) - 0.08639.  Velocity and the effects of body length To examine differences in both median and maximum velocity, we performed ANCOVAs (table 3.6), using BL as a covariate, among datasets obtained within the same season (see figs. 3.8a & b). Between juveniles there was no significant effect of BL on median velocity, although there was a significant effect of BL on maximum velocity. The associated linear regression represents all juveniles and depicts how longer individuals exhibited a lower length-specific maximum velocity (fig. 3.8b). Additionally, between the juveniles, we found no significant effect of group on either median or maximum velocity (see table 3.6). Between adult GHf and the wild Coho salmon we found no significant effect of BL on either median or maximum velocity. We also found no significant effect of group on maximum velocity; however, GHf exhibited a significantly lower median velocity than wild Coho salmon with BL as a covariate (table 3.6).  80  Source  d.f.  F  P  2.22 3.49  0.155 0.055  44.01 0.18  < 0.001** 0.84  0.0116 2.48 0.437  0.03 5.67  0.874 0.039*  1.52 3.911 1.144  1.33 3.42  0.276 0.094  MS  GHf, GHr, and wild from fall (juveniles) Median Velocity BL 1 0.876 Group 2 1.38 Error 16 0.394 Max. Velocity BL 1 68.5 Group 2 0.273 Error 32 1.56 GHf and wild from summer (adults) Median Velocity BL 1 Group 1 Error 10 Max. Velocity BL 1 Group 1 Error 10 * Indicates significance level < 0.05 ** Indicates significance level < 0.01  Table 3.6 Seasonal and group comparison of median and maximum velocity as a function of BL. ANCOVA results for median and maximum velocity among GHf, GHr (only from the fall), and wild with BL as a covariate, tested among individuals within the same season.  81  7  GHf, fall GHf, summer wild, fall wild, summer GHr, fall  -1  Median Velocity (BL s )  6  5  4  3  2  10  20  30  40  50  A fall cohorts regression fall cohorts summer cohorts regression (constrained) summer cohorts  -1  Maximum Velocity (BL s )  14  12  10  8  6  4  10  B  20  30  40  50  Body Length (cm)  Figures 3.8 Length-specific median (A) and maximum velocity (B), from the fall and summer cohorts are plotted as a function of BL (cm). A) Median velocity (BL s-1) among the fall and summer cohorts: red circles represent fall wild (n= 6), blue squares represent fall GHf (n= 7), green triangles represent fall GHr (n= 7), open red circles represent summer wild (N= 6), and open blue squares represent summer GHf (N= 7). B) Maximum velocity (BL s-1) among the fall cohorts, represented by the open circles and closed squares represent all individuals from the summer. A linear regression among the fall cohorts is represented by the solid line 82  (maximum velocity = (17.35 ± 2.38, 1 standard deviation) – ((0.3712 ± 0.11) *BL), r2 = 0.758) and among the summer cohorts, which was constrained by the slope obtained from the fall cohorts, by the dotted line (maximum velocity = (22.983 ± 0.35) – (0.3712 * BL), r2 = 0.889). An unconstrained linear regression among the summer cohorts: maximum velocity = (20.086 ± 1.92) – ((0.30 ± 0.05)*BL), r2 = 0.889.  We used the relationship obtained for maximum velocity as a function of BL among juveniles, to further examine the same relationship among adults. By constraining the slope of a linear regression, we obtained a similar relationship among the summer cohorts (see fig. 3.8b). We also obtained an unconstrained linear regression among the summer cohorts.  The two slopes were not  significantly different (t = 1.08, 45 degrees of freedom, P-value > 0.20). We were not sure what to expect regarding changes in (maximum) velocity with changes in BL, or whether the apparent discontinuity in velocity between juveniles and adults was due to the use of two distinctive protocols (see fig. 3.8a & b). Thus, we also included comparisons, or ANCOVAs using BL as a covariate, of maximum length-specific and absolute velocity between all 3 groups from both juveniles and adults (table 3.7, figs. 3.9a & b). Using maximum lengthspecific velocity we found no significant differences between the 3 groups, but we did find a significant relationship for length-specific maximum velocity decreasing with increasing BL among all cohorts (see table 3.7 and fig. 3.9a).  Using  83  maximum absolute velocity we found no significant differences between the 3 groups of juveniles or between adult GHf and wild Coho salmon, nor did we find a significant effect regarding changes in maximum absolute velocity with changes in BL (see table 3.7). Using maximum absolute velocity again, we tested all the individuals (from both juveniles and adults) from the 3 groups, and found significant differences between the groups (table 3.7 and fig. 3.9b; post hoc tukey assessment for GHf versus GHr: P value = 0.72, GHf versus Wild: P value = 0.003 and for GHr versus Wild:  P value = 0.06).  In addition, we found a  significant relationship for maximum absolute velocity increasing with increasing BL (see fig. 3.9b).  84  F  P  Maximum BL s-1, all cohorts (juveniles and adults) BL 1 78.44 Group 2 4.69 Error 45 2.46  31.84 1.90  < 0.001** 0.16  Maximum cm s-1, fall cohorts (juveniles only) BL 1 955.8 Group 2 353.8 Error 32 485.7  2.0 0.73  0.17 0.49  Maximum cm s-1, summer cohorts (adults only) BL 1 109 Group 1 3590 Error 10 1865  0.06 1.93  0.81 0.20  Maximum cm s-1, all cohorts BL 1 Group 2 Error 45  44.89 6.73  < 0.001** 0.003**  Source  d.f.  MS  75865 11367 1690  * Indicates significance level < 0.05 ** Indicates significance level < 0.01 Table 3.7 Group comparisons, within and between seasons, of maximum absolute and maximum length-specific velocity (BL s-1) and absolute velocity (cm s-1). ANCOVA results for maximum length-specific velocity with BL as a covariate between all cohorts, as well as for maximum absolute velocity between the cohorts from the fall and the summer separately (see fig. 3.9a), and between all cohorts (see fig. 3.9b).  85  GHf Wild GHr GHf regression Wild regression  -1  Maximum Velocity (BL s )  14  12  10  8  6  4  10  20  A  30  40  50  Body Length (cm) GHf Wild GHr GHf regression Wild regression  -1  Maximum Velocity (cm s )  400  300  200  100 10  20  30  40  50  Body Length (cm) B Figure 3.9 A) Maximum length-specific velocity (BL s-1) as a function of body length (cm), for both the fall and summer cohorts. Blue squares represent GHf, red circles represent wild coho salmon, and green triangles represent GHr. Linear regressions for wild coho, represented by the red line: maximum velocity = ((11.19 ± 1.9, 1 standard deviation)*BL) + (-0.055 ± 0.07), r2 = 0.19; and GHf, represented by the blue line: maximum velocity = ((12.44 ± 0.68)*BL) + (-0.140 ± 0.02), r2 = 0.86. B) Maximum absolute velocity (cm s-1) has also been plotted as a function of body length (cm). Linear regression for GHf: maximum velocity = ((3.283 ± 0.569)*BL) + (119.66 ± 18.8), r2 = 0.814; and for wild coho salmon: maximum velocity = ((9.382 ± 1.62)*BL) + (7.676 ± 42.7), r2 = 0.823.  86  DISCUSSION We hypothesized that among the smaller, length-matched juveniles, the GHf would out perform their conspecifics and that there would be no differences in sprinting performance between the larger cohorts, or adults.  Indeed, at  approximately 20 cm BL, GHf took significantly longer to sprint to exhaustion than the wild Coho salmon (see fig. 3.4 and table 3.3). This is discussed in the first section that follows, “Exhaustive sprinting, fatigability, and muscle biochemistry.” Among juveniles we also found no significant differences in maximum velocity (see tables 3.3 & 3.5), their relationship of average velocity as a function of tailbeat frequency (see fig. 3.5), or net anaerobic substrate usage (table 3.4 and figs. 3.6a, b, & c). However, we did find significant differences in median velocity between GHr and wild Coho salmon (see table 3.3). We were unable to obtain length-matched adults, and GHf were significantly longer than wild Coho salmon (see table 3.5).  Nevertheless,  between adults, we found significant differences in both median and maximum velocity but no significant differences in time to exhaustion (see tables 3.3, 3.5, & 3.6 and figs. 3.4, 3.7b & c).  Upon considering the relationship of maximum  velocity as a function of BL, among both juveniles and adults, between the 3 groups, adult GHf may have realized a lower maximum velocity than the wild coho salmon (tables 3.3 & 3.7 and figs. 3.9a & b). This is discussed in the section below titled, “Changes in sprinting performance with changes in body length.” We suggest that juvenile GHf may realize an enhanced ability to evade  87  chasing predators and to chase prey, in contrast to adult GHf that may realize similar or even a reduced ability thereof. Finally, we speculate that adult GHf may have a reduced ability to successfully negotiate challenging reaches encountered during prospective spawning migrations.  Exhaustive sprinting, fatigability, and muscle biochemistry Consistent with our hypothesis, using juveniles obtained during the fall, GHf took significantly longer to sprint to exhaustion than wild Coho salmon, but neither differed significantly in time to exhaustion from GHr (table 3.3 and fig. 3.4). Between GHf and wild Coho (both 17cm BL), Blier et al. (2002) found that GHf had significantly higher enzyme activities within their white lateral or axial muscle for all enzymes tested: citrate synthase (CS), pyruvate kinase (PK), and lactate dehydrogenase (LDH). This finding suggests a possible mechanism by which GHf may take longer to reach exhaustion: GHf had a higher capacity to metabolize aerobic and anaerobic substrates, and support ATP turnover within their white axial muscle during sprinting. A similar finding was obtained between 10cm BL GHf and wild Coho (Hill et al., 2000), whereby GHf had significantly higher enzyme activities for phosphofructokinase (PFK) and cytochrome C oxidase (CCO), also within their white axial muscle. No data of muscle enzyme activity is presently available for GHr. Our results on anaerobic substrate usage (see table 3.4 and figs. 3.6a, b, & c) suggest that regardless of time to exhaustion, all 3 groups exhibited a similar  88  net usage of CrP, ATP, and anaerobic glycolysis within their white muscle. We did not assess aerobic metabolism within their white muscle, which is known to contribute to exhaustive sprinting (Richards et al., 2002a). Therefore, we do not know the extent to which there may be differences in the contribution of aerobic metabolism or oxidative phosphorylation during sprinting.  We did assess  anaerobic substrate content (CrP, ATP, and lactate) in the red muscle, between rested and exhausted wild and GHf, and only found a significant accumulation of lactate between rested and exhausted GHf. While this may suggest that the red lateral muscle significantly contributed to exhaustive sprinting, at least among the GHf, it is also possible that the lactate levels measured in the red muscle were the result of higher blood lactate, as opposed to higher lactate levels within the red muscle fibres themselves. Red muscle is thought to be unable to support positive muscle work during active bouts of sprinting, given its lower contraction force-velocity characteristics relative to white muscle (Coughlin, 2002). On the other hand, studies of muscle activation patterns suggest that red muscle is recruited during fast-starts of rainbow trout (O. mykiss) and bluegill sunfish (Lepomis macrochirus) (Elerby and Altringham, 2001; Jayne and Lauder, 1993, respectively), and during sprinting of rainbow trout (Elerby and Altringham). However, electromyographic leads that were implanted in red lateral muscle may have also recorded muscle activity from the underlying white muscle. Our profiles of instantaneous velocity during sprinting suggest that the red lateral muscle contributed during exhaustive sprinting (see fig. 3.1a). Firstly, we  89  know from previous studies that wild salmon (Coho and Atlantic) have a higher Ucrit than length-matched GH-transgenic salmon throughout their life history (Farrell et al., 1997; Stevens et al., 1998; Lee et al., 2003; Deitch et al., 2006). Specifically, at 22cm BL, GHf (coho) exhibited a Ucrit of 2.14 BL s-1 and at 20cm BL the wild Coho exhibited a Ucrit of 3.5 BL s-1 (Farrell et al., 1997).  All  individuals, from the present study employed a range of velocities, which were often less than 2 BL s-1 (see fig. 3.1a). The median velocity from the wild Coho was 3.02 BL s-1 (table 3.2), which was approximately 85% of their Ucrit. However, periods of low velocity may be indicative of acceleration, a reduced ability for maximum power output (due to fatigue), or even gliding or deceleration. In fact, during experiments fish often sprinted from one end of our experimental tank to the other, which constitutes approximately 5 body lengths (or 100cm), having to stop and start with each pass. Even though we could not distinguish between instances of active (bursting) and passive (gliding) swimming, these data suggests that the red muscle may have contributed during exhaustive sprinting.  That said, our data does not offer insight regarding red muscle  recruitment during actual bouts of sprinting, and further studies are therefore suggested. Juvenile GHf took longer to sprint to exhaustion while exhibiting similar net usage of anaerobic substrates, which may have been due an elevated capacity to support ATP turnover than their length-matched wild conspecifics (Blier, et al., 2002). We also reported no differences in their kinematics, or the relationship of  90  velocity to tail-beat frequency during sprinting. Velocity increased with increasing tail-beat frequency, during sprinting, and this relationship was similar between the three groups (see fig. 3.5). The relationship of velocity to tail-beat frequency, at similar velocities, has only been studied once previously. Bainbridge (1958) used a “fish wheel” to sprint dace, goldfish, and trout. The water within the “fish wheel” was not propelled by a motor but by spinning its platform. It was unclear how the fish were induced to sprint; presumably the spinning of the platform may have startled the fish, such to induce sprinting. The results of Bainbridge (1958) suggested that the relationship for velocity as a function of tail-beat frequency was linear (this was not statistically defined in his paper).  We actually found that a power relationship more  accurately represented our results (not reported). We attribute the differences between our studies as due to sprinting fish within a confined area (present study) in contrast to sprinting fish within a “fish wheel.”  Individuals sprinting  within a confined area must stop and start, as opposed to fish swimming or sprinting around a circular “fish wheel.” In other words, we speculate that the highest tail-beat frequencies (reported presently) may have been used to accelerate from a lower velocity, thus lowering the average speed attained at such frequencies de facto. In contrast to accelerating from a glide or higher starting velocity, which may have been facilitated by continuous sprinting within a circular or donut-shaped tank. Thus, we suggest that “real” the relationship for  91  increasing average velocity with increasing tail-beat frequency may, in fact, be linear. Among juveniles, we also determined their maximal and median sprinting velocities and found no differences in maximal velocity (see table 3.3). However, we found that GHr exhibited a significantly higher median velocity than wild Coho salmon (see table 3.3). We measured median velocity because it represents the average velocity when fish are half-exhausted, while median velocity itself was not significantly correlated with changes in average velocity over time (see fig. 3.2 and table 3.1). Interestingly, the observed differences in median velocity (see table 3.2) were not correlated with differences in time to exhaustion (see table 3.5) nor were they correlated with changes in BL (see table 3.6). Thus we are left with the additional conclusion, simply that GHr may be employing a different strategy when undergoing exhaustive sprinting than wild Coho salmon. Between adult GHf and wild Coho salmon (summer cohorts) we found no significant differences in time to exhaustion (table 3.3). No data are available on muscle enzyme activities from adult GHf and wild Coho salmon, which could have potentially accounted for the differences in time to exhaustion between the juveniles. We also found that adult GHf exhibited significantly lower median and maximal velocities than adult wild Coho salmon (see table 3.3). Unfortunately, we were unable to obtain length-matched individuals (see table 3.2 and 3.5), which complicates our analysis. Briefly, the differences in BL may explain the differences in maximal velocity, but they did not explain the differences in median  92  velocity (see table 3.6). Thus, even though we found no differences in time to exhaustion, it seems that GHf and wild Coho (adults, from the summer) used different strategies when undergoing exhaustive sprinting, much as we found between juvenile GHr and wild Coho salmon (from the fall). Among the adults neither median nor maximum velocity correlated with time to exhaustion; however, the change in average velocity over time did (correlate with time to exhaustion). Adults that took longer to sprint to exhaustion exhibited a less pronounced decrease in average velocity over time, but there were no differences between the groups (table 3.5).  Even though most  individuals realized a decrease in average velocity over time, we did not find such a correlation among juveniles. To assess instantaneous velocity over time, average velocity was calculated or binned over each 5s interval and linear regressions were fit to this averaged data (see fig. 3.1a and 3.1b). Because the juveniles were sprinted within a confined area they often employed stop and start sprinting, though sometimes they did sprint around the perimeter of the tank. Stops and starts necessitates acceleration from lower velocities, and thus, lower velocities would have been used prior to any onset of fatigue. Therefore the associated shift in their linear regression would have reduced the perceived impact of fatigue (or decreasing average velocity over time) de facto, perhaps because those individuals had to accelerate from lower velocities. It was unclear if maximum velocity was retained or reduced over time. Some individuals did appear to retain  93  their maximum ability (over time); however, this was not clearly associated with changes in their fatigability or the group, by which they were identified. Regarding the decrease in average sprinting velocity over time, which was obtained from all but two individuals (see table 3.2), ATP may be limited in exhausted individuals, thereby defining the end point of exhaustive sprinting (Richards et al., 2002a). Initially, sprinting is powered by the hydrolysis of stored CrP and subsequently by stored ATP (the first 10s of sprinting from 210g rainbow trout; Richards et al., 2002a). As individuals continue to sprint, ATP stores are depleted and may be replenished via glycolysis. With the resultant depletion of muscle glycogen, glycolysis would be increasingly inhibited. ATP turnover must thereon be supported via oxidative phosphorylation. With each subsequent step or reaction, ATP turnover occurs more slowly (Dobson et al., 1987), therefore, as fish sprint longer the probability that a given white muscle cell would be ATP limited increases. Less ATP within a muscle cell means less ATP for crossbridge cycling (the process by which muscle forces are produced). Fewer crossbridges cycling in parallel and in series result in lower force and velocity of shortening, respectively, which yields lower tail-beat amplitude and frequency, and therefore lower velocity.  This biochemical explanation may offer a  reasonable hypothesis or a mechanism by which white muscle realizes exhaustion. Fitts (1994) reviewed other plausible or coinciding explanations that could also define the end point of exhaustive exercise on the basis of 1) fatigue within excitation-contraction-coupling or a reduced ability for the sarcolemma to  94  depolarize, 2) a reduction in peak tetanic and twitch muscle forces 3) as well as accumulation of inorganic phosphates.  The effects of GH on muscle enzymes At both 10 cm and 17 cm BL, GHf exhibited higher muscle enzyme activity than length-matched wild Coho salmon (Hill et al., 2000; Blier et al., 2002), which offers an explanation of how juvenile GHf exhibited similar kinematics, velocity, and net anaerobic substrate usage while taking significantly longer than wild Coho salmon to reach exhaustion. By contrast at approximately 11cm BL, GHf Atlantic salmon had significantly lower aerobic and glycolytic enzyme activities than wild Atlantic salmon, within their red and white lateral muscle (Levesque et al., 2008). Also between GHf and wild Atlantic salmon, both 56cm BL, Dietch et al. (2006) found no difference in CCO or CS activity but found significantly lower CS activity within the white lateral muscle of GHf. The reasons for contrasting differences obtained between GHf and wild Coho and GHf and wild Atlantic salmon are not apparent.  Furthermore, ration-restricted adult Atlantic cod  exhibited lower aerobic and glycolytic enzyme activity within their white lateral muscle, relative to their satiated conspecifics (Pelletier et al., 1993a; 1993b). Other previous studies have demonstrated increases in LDH activity with increases in BL and mass among a variety of teleost fish (Goolish, 1991). Thus, it seems that changes in muscle enzyme activity with changes in growth rate, due  95  to feed restriction or GH upregulation, have different effects on different species of fish. Table 3.8 offers a summary of the effects of GH administration on muscle enzyme activity among humans, pigs, cows, and fish. Among adult humans (males and females), administered intravenous (i.v.) GH for 14 hours overnight, Short et al. (2008) found an increase in CS activity from their vastus lateralis muscle. Not included in table 3.8, Peyreigne et al. (2002) administered GH via daily s.c. (subcutaneous) injections for 2 weeks that did not change muscle mitochondrial respiration among normal, ~280g, adult rats (from their entire hindlimb). The results from Peyreigne et al. (2002) contrast those from GHf Coho salmon (Hill et al., 2000; Blier et al. 2002) but agree with those from 56cm BL GHf Atlantic salmon (Dietch et al., 2006), which exhibited an increase and decrease, respectively, in CCO activity (CCO is involved in mitochondrial respiration). Interestingly, Semsarian et al. (1999) found an increase in LDH and alanine aminotransferase (ALT) activity among cultured murine myoblasts that were provided media supplemented with IGF-1 (IGF-1 is also upregulated in GHf coho salmon; Raven et al., 2008). Collectively, and with the inclusion of various breeds of pigs and cows, these results do not offer a clear understanding of the effects of GH on muscle enzyme activity, which is perhaps not surprising given the complex nature of GH’s signal transduction cascade (Kopchick and Andry, 2000). Moreover, GH does not upregulate muscle growth in all animals. In fact, GH has no effect on muscle growth in guinea pigs (Baumman, 1997), and  96  supraphysiological levels (of GH) negatively regulate muscle growth in chickens (Vasilatos-Younken, 2000). Also, different effects of GH-transgenesis were seen in different strains of trout with low and high growth rates due to domestication (Devlin et al., 2001). The regulation of muscle enzymes is thought to occur in concert with other physiological processes (Moyes and LeMoine, 2005).  Other factors, besides  differences in species, which might account for differences in changes in muscle enzyme activity include, but are not limited to the following: the mode of GH administration, the starting age and mass of the subjects, the duration of treatment, and the muscles sampled (see table 3.8). That said, the findings revealing that the white muscle enzymes from GHf have higher activity levels than wild Coho salmon (Hill et al., 2000; Blier et al., 2002) offer an explicit explanation as to why juvenile GHf may have taken longer to sprint to exhaustion while exhibiting similar kinematics, velocity, and net anaerobic substrate usage: they had an increased capacity to support ATP turnover in their white muscle.  97  Table 3.8  98  Table 3.8  99  Table 3.8 Changes in muscle enzyme activities, associated with GH treatment, among species in which GH upregulates skeletal muscle growth. Species include: humans (Florini et al., 1996), pigs and cows (Etherton and Bauman, 1998), as well as fish (Rescan, 2005). Data was included from all available studies in which GH was either injected into normal or wild-type subjects, induced via transgenesis yielding an increase in GH expression or production within skeletal muscle, as well as studies using targeted GHR (GH receptor) lesioning (in contrast to systemic GH insensitivity). Column 2 indicates whether the associated study used age or size-matched controls. Column 3 indicates the size or age at which treatment was initiated for nontransgenic GH augmentation (among humans, pigs, and cows) and the sample size and age for GH-transgenic fish. Column 8 indicates the muscle enzymes that differed significantly from their wild-type or saline-controls: ↑ for significant increases and ↓ for significant decreases in activity. Some muscle enzymes have been color-coded: red for aerobic enzymes, which are involved in either the citric acid cycle or mitochondrial respiration, and green for glycolytic enzymes. Enzyme abbreviations: ATPase= myofibrillar ATPase, AST= aspartate aminotransferase (amino acid metabolism), ALT= alanine aminotransferase (amino acid metabolism), CCO= cytochrome c oxidase (mitochondrial respiration), CS= citrate synthase (citric acid cycle), GDH= glutamate dehydrogenase (amino acid metabolism), HAD= β-hydroxyacyl-CoA dehydrogenase (fatty acid metabolism), IDH= isocitrate dehydrogenase (citric acid cycle), LDH= lactate dehydrogenase (glycolysis), MDH= malate dehydrogenase (citric acid cycle), PFK= phosphofructokinase (glycolysis), and PK= pyruvate kinase (glycolysis). Other abbreviations: BMI= body mass index, i.v.= intravenous, i.m.= intramuscular, long. dorsi= longissimus dorsi, glut. med.= gluteus medius, and s.c.= subcutaneous. * Values reported, from Oksbjerg et al. (1995) were presented as a sum of enzyme activity among all three muscles.  100  Changes in sprinting performance with changes in body length Between the fall (juveniles) and summer (adults) cohorts, on average the adults took longer to sprint to exhaustion than the juveniles (see fig. 3.4 and table 3.2). However, because we used two different protocols (for the two different cohorts) this finding should be interpreted with caution. Moreover, there is no precedent for the changes that we observed in time to exhaustion during sprinting with changes in BL. McDonald et al. (1998) found an increase in time to exhaustion with an increase in BL using juvenile Atlantic salmon, as did Gregory and Wood (1998), who similarly compared time to fatigue and growth rate (augmented by ration-restriction, half-satiation) among juvenile rainbow trout. However, both of those studies swam multiple fish, simultaneously, within the same swim tunnel, and thus no adjustment was made for smaller or larger individuals. Therefore, larger individuals that took longer to exhaust would have employed a lower mass specific power output than the smaller individuals to achieve the same absolute velocity. Thus, the larger individuals might have also exhibited slower kinetics of substrate usage. The differences in average time to exhaustion, between the fall and the summer cohorts, might be explained by the differences between the two protocols used to exhaustively sprint the fish. During the summer, continuous sprinting (or burst and glide swimming), facilitated by the use of a circular or donut-shaped tank, would have enabled individuals to accelerate from a higher starting velocity, in contrast to individuals from the fall that employed stops and  101  starts (within a confined area).  Continuous sprinting would also enable  individuals to continually escape or swim away from the experimenter. Thus during gliding individuals could recover while continuing to escape. Therefore, the adults may have performed the same amount of net work, but because their forward progress was not inhibited they took longer to sprint to exhaustion. Unfortunately, we were unable to use the same protocol and are unable make any conclusions regarding changes in time to exhaustion with changes in BL during sprinting. Further research is recommended to demonstrate the efficacy of the two protocols. There is, seemingly, a dichotomy in the relationship of BL with maximum and possibly median velocity (figs. 3.8a & b), between the fall and summer cohorts. We did not find a scaling effect for median velocity, which is a function of both time and BL (table 3.6). We did, however, obtain a significant correlation between maximum length-specific velocities as a function of BL, among the fall cohorts (table 3.6 and fig. 3.8b). Changes in sprinting velocity as a function of BL have only been considered once previously.  Goolish (1991) compiled data from Bainbridge  (1960) and derived the following linear regression for sprinting velocities which were averaged over 1s (we used 100ms in the present study), among goldfish, dace, and rainbow trout, all < 30cm BL: BL s-1 = 11.68 – 0.056 * (BL). Averaging sprinting velocities over 1s may underestimate maximum velocities. Additionally, speeds reported by Bainbridge (1960) were obtained via calculations used to  102  describe the velocity of the water within the “fish wheel,” which indirectly predicts swimming velocities. In fact, fish are known to migrate forwards and backwards when they employ burst and glide swimming within a given swim tunnel. Moreover, from a separate study (Nelson et al., 2008), 3.5g dace, estimated to be < 5cm BL (actual BL was not reported), when induced to sprint via manual tailgrabs, exhibited speeds in excess of 20 BL s-1, which are almost twice the speeds reported by Bainbridge (1960). We suggest that maximum velocity (from the present study) represents a period during which individuals would have overcome drag forces, turbulent water, deceleration due to tail-grabs, as well as lost momentum due to stops and starts within a confined area, to realize their peak velocity. That said, a precise conclusion, regarding changes in maximum velocity with changes in BL remains elusive. On one hand, the fact that a similar relationship (constant, or slope) could be used to predict the maximum velocity between two different protocols and 2 different cohorts of different body lengths is perhaps a coincidence (see fig. 3.8b).  However, if we accept that relationship as real, then the fall cohorts  underperformed. If they underperformed, then it may have been because such individuals would have had to work to accelerate from lower velocities while the summer cohorts would have done so from a higher starting velocity. Moreover, by accepting the scaling effect for maximum velocity (see fig. 3.8b), we conclude that the differences in maximum velocity between adult GHf and wild Coho salmon may be explained by differences in BL.  103  Changes in Ucrit and fast-start (escape) velocities with changes in BL have been reported previously (Goolish, 1991; James and Johnston, 1998, respectively), but perhaps the aforementioned relationship (from the present study) was exaggerated by a tank effect (larger fish realize lower velocities within the same swim tunnel as smaller fish). Additionally, consideration of maximum velocity between adults is complicated by the use of GHf and wild Coho salmon of different body lengths. Moreover, it is unclear whether or not the changes observed are also due to the effects of using a different protocol (to sprint fish) or even GH upregulation. To that end, we offer 3 ways to consider changes in maximum sprinting velocity with changes in BL: 1) We assume that length-specific maximum velocity should decrease with increasing BL and that the dichotomy (see fig. 3.8b) is due to the differences in the two protocols. We find such a scaling effect among the fall cohorts, and conclude that the differences between GHf and wild coho salmon from the summer are due to their differences in BL. 2) We assume that there is no dichotomy in maximum length-specific sprinting velocity:  maximum velocity from the fall is comparable to  maximum velocity from the summer (fig. 3.9a). The ANCOVA, using BL as a covariate, revealed no significant differences between the 3 groups; however, there was a significant relationship for maximum length-specific velocity decreasing with increasing BL (table 3.7).  104  3) We firstly, assume that there is no dichotomy in maximum sprinting velocity (as per option #2, above) and secondly, that we should be comparing absolute velocities (see fig. 3.9b). The ANCOVA, using BL as a covariate, revealed a significant relationship for maximum absolute velocity increasing with increasing BL, and significant differences between the GHf and the wild Coho salmon (see table 3.7). The GHf realized lower maximum absolute sprinting velocities than the wild Coho salmon.  Few studies have examined sprinting and fewer still have examined sprinting in a laboratory setting to which we may compare our results. Nelson and Claireaux (2005) induced juvenile European Sea bass, averaging 16 cm BL, to sprint via a single manual tail-grab, and obtained sprinting velocities of approximately 100-300 cm s-1 or 6-18 BL s-1. The range of velocities reported from the Sea bass are similar to those reported presently, among the fall cohorts (see table 3.2).  Colavecchia et al. (1998) swam wild adult Atlantic salmon,  averaging 51.2 cm BL, in a swimming flume and obtained burst speeds of 350 cm s-1 or approximately 7 BL s-1, which is similar to sprinting speeds reported among the summer cohorts (see table 3.2).  Briefly, when we considered  velocities reported from salmonids undergoing spawning migrations, the relationships that compared among both the fall (juveniles) and summer (adults) cohorts (see figs. 3.9a & b) predicted or corroborated such findings.  105  Ecological importance of sprinting and prospective abilities Sprinting has been characterized as maximal and unsteady or burst and glide swimming (Bainbridge, 1960; Nelson et al., 2008), encompassing velocities faster than Ucrit (Goolish, 1991; Katz et al., 1999; Reidy et al., 2000). Ecologically, sprinting may be used during predator-prey interactions by both attacking predators and escaping prey. Savitz and Bardygula-Nonn (1997) used video analysis to determine the escape velocity of alewives, yellow perch, rainbow smelt, bloaters, fathead minnows, and spottail shiners (approximately 4-5 cm BL) undergoing predatory attacks from Coho and Chinook salmon (approximately 30 cm BL). The prey utilized escape velocities ranging 28-70 BL s-1, which exceed any predictions based on our results (see figs. 3.8b, 3.9a & b). Predatory chinook salmon used average attack velocities ranging (for each prey species) 8.7-10 BL s-1, and coho salmon used average attack velocities ranging 8.7-12 BL s-1, both which are similar to sprinting velocities reported presently for adult wild coho salmon (see table 3.2). Furthermore, Ucrit studies on both chinook and coho salmon also support the usage of sprinting during those attacks, reported in Savitz and Bardygula-Nonn (1997). Experiments by Savitz and Bardygula-Nonn (1997) were conducted at 15 o  C, and used Chinook and coho salmon. Gallaugher et al. (2001) and Randall et  al. (1987) performed Ucrit studies on chinook salmon approximately 30 cm BL at 10 oC. They obtained Ucrit of 2.3 and 2.7 BL s-1, respectively. We speculate that  106  at 15 oC Ucrits would not be more than double what was obtained at 10 oC. Glova and McInerney (1977) obtained Ucrit of coho salmon 12 cm BL over a range of temperatures. Ucrits at 23, 18, 13, 8, and 3 oC were approximately, 5.0, 5.0, 4.5, 4.0, and 3.5 BL s-1, respectively. Farrell et al. (1997) reported Ucrit of 20 cm BL wild coho salmon at 11-13 oC of 3.5 BL s-1. Even though coho salmon from Glova and McInerney (1977) were 12 cm BL, as opposed to 30 cm BL from Savitz and Bardygula-Nonn (1997), Ucrit has been shown to decrease with increasing BL (Goolish, 1991).  Based our results (see table 3.2) and the  aforementioned studies on Ucrit, we suggest that predatory chinook and coho salmon from Savitz and Bardygula-Nonn (1997) were using sprinting to attack their prey. Lundvall et al. (1999) determined that adult Eurasian perch (Perca  fluviatilis; approximately 15 cm BL) used attack velocities approximately 10 BL s-1 when attacking “young-of-the-year” perch (approximately 4 cm BL), which used escape velocities approximately 30 BL s-1. Those escape velocities of young-ofthe-year perch are in excess of velocities we reported, while attack velocities of adult Eurasian perch were similar (see figs. 3.8b, 3.9a, & b; Lundvall et al.,1999). Interestingly, from both Lundvall et al. (1999) and Savitz and Bardygula-Nonn (1997) predator attack velocities were similar to sprinting velocities that we observed, by contrast, prey escape velocities exceed velocities that we observed. Prey may have used fast-starts to escape their respective predators, however, fast-start velocities from rainbow trout are actually similar sprinting velocities  107  reported herein (see tables 2.1 and 3.2; Webb, 1976; Domenici and Blake, 1997). To further demonstrate the use of sprinting by escaping prey, we have also considered the time dependency of predator-prey interactions. Predator-prey interactions can involve chasing. Turesson and Brönmark (2004) examined predator-prey interactions among pikeperch and perch (both were approximately 20 cm BL) while they were attacking roach (approximately 6 cm BL) within a confined area. Reportedly, pikeperch took an average of 13 min, from the time of their first attack, to capture their prey. Pike took an average of 5 min to capture their prey. These fish negotiated several attacks throughout the attack period; however, it was unclear to what extent either the predators or prey were sprinting, nor was it clear for how long the response, from a given attack, lasted.  Interestingly, Fuiman (1991) determined that attacks from yearling  herring, (approximately 7 cm BL) induced 0.5-2s of active flight or swimming among larvae herring (approximately 2 cm BL). More prolonged responses during predator-prey interactions have been observed in the wild among schools of herring, which were attacked by Killer whales (Nøttestand and Axelson, 1999) and Atlantic puffins (Axelson et al., 2001), as well as for Killer whales chasing Bluefin tuna (Guinet et al., 2007). From laboratory-based experiments, prolonged predator-prey interactions have also been reported for Eurasian perch chasing roach (Christensen, 1996) and bottlenose dolphins chasing tilapia (Marten et al., 2001).  108  Christensen (1996) conducted predator-prey experiments of Eurasian perch attacking roach within a 3 m3 pool.  Piscivorous Eurasian perch  (approximately 24 cm BL) pursued roach (5.9-11.6 cm BL) for 18-20s on average, negotiating 5-10 attacks within a 10-minute period (Christensen, 1996). Dolphins have been reported to chase tilapia “at high speed” for 7s, within an enclosed tank, while maintaining a distance of 10-100 cm (Marten et al., 2001). Marten et al. (2001) also describes other predator-prey interactions as ‘chases’ between killer whales and bottlenose dolphins with anchovies, tilapia, and salmon (actual species was not reported) in natural habitats; however, they did not report time durations. Nøttestand and Axelson (1999) used sonar to observe 54 antipredator responses by schools of herring during 10 whale-herring interactions, which lasted from 2-80 min. Antipredator events occurred every 3.9 min, on average, and each event lasted from 1-8s. Similarly, Axelson et al. (2001) reported that the average antipredator response to puffins attacking schools of herring lasted 42s. While possible, whether or not such responses involved sprinting, among an entire school or for individual herring is unknown, as no velocities, for the antipredator maneuvers, were reported in these studies. Guinet et al. (2007) modeled predatory chases by Killer whales, which based on observations reported in the wild, may last up to 30 min and are used to exhaust their prey, Bluefin tuna. Their model integrated available data on swimming velocities from Killer whales and Bluefin tuna, as well as tuna muscle physiology. They reported  109  that, indeed, Killer whales have the ability to induce exhaustion among smaller Bluefin tunas by chasing them.  However, this extraordinarily prolonged  interaction may be exclusive to tunniform swimmers, as they exhibit markedly different morphological features (e.g. medial-centralization of white muscle, the use of a hydrofoil caudal peduncle for propulsion…), which may facilitate such continuous high-performance swimming (Shadwick, 2005). Sprinting, and thus white muscle performance is important during predator-prey interactions.  The aforementioned studies confer its use by  attacking predators, and by escaping prey undergoing a prolonged chase from a given predator. Thus, since juvenile GHf took significantly longer to sprint to exhaustion while exhibiting similar performance among all other parameters, GHf might have an enhanced ability to evade a chasing predator and to chase prey, relative to wild Coho salmon.  However, juvenile GHf also exhibited a lower  propensity for antipredator behavior when in the presence of a predator (Sundström et al., 2007; Abrahams and Sutterlin, 1999).  These opposing  potential responses to predation therefore make difficult any conclusions regarding overall mortality. No data on antipredator behavior among adult GHtransgenic Coho salmon is available. Thus, because GHf adult Coho salmon exhibited similar time to exhaustion but lower velocities than their wild conspecifics we suggest that adult GHf may have a reduced ability to evade chasing predators and to chase prey.  110  Wild salmon are renown for their use of sprinting during spawning migrations. Hinch et al. (2002) implanted EMGs and trackers into wild adult Sockeye salmon (Oncorhynchus nerka), ranging 50-62cm BL, undergoing spawning migration up the Fraser River (British Columbia, Canada). Throughout their spawning migration Sockeye salon averaged < 3 BL s-1.  However, at  specific points or reaches they used sprint or burst swimming speeds for up to 10% of the time, and the speeds reported were > 9-10 BL s-1 or 450-620 cm s-1. Brown and Geist (2002) used similar methods as Hinch et al. (2002) to examine adult Chinook salmon (Oncorhynchus tshawytscha), 67-99 cm BL, undergoing spawning migration up the Klickitat River (Washington, USA). Brown and Geist (2002) also reported relatively modest average swimming speeds, approximately 2.1 BL s-1, throughout Chinook salmon migrations.  Furthermore, Brown and  Geist (2002) estimated that in order to successfully negotiate or jump Lyle Falls, the Chinook would need to exceed 8 BL s-1 or 536-792 cm s-1. Lauritzen et al. (2005) obtained digital video recordings of wild Alaskan Sockeye salmon (O.  nerka), 60-70cm BL, jumping over Brooks Falls in the Russian River (Katmui National Park, Alaska).  They determined that 10% of Sockeye jumps were  successful, and that during successful attempts Sockeye realized takeoff velocities of approximately 8 BL s-1 or 550 cm s-1. Velocities reported from all three of the preceding studies are similar to those that would be predicted based on the relationship obtained from wild Coho salmon from the fall and the summer (see figs. 3.9a & b), and exceed velocities based on relationships obtained when 111  either the fall (juveniles) or summer (adults) cohorts were assessed separately (see fig. 3.8b). We therefore speculate that our comparison of sprinting velocities among both juveniles and adults may be valid (see figs. 3.9a & b). Consequentially, we also suggest that adult GHf may perform either similarly or exhibit lower maximum sprinting speeds than length-matched wild Coho salmon. If we can extrapolate our sprinting results to later life history stages, then when sprinting to negotiate challenging reaches such as waterfalls, adult GHf might exhibit similar or reduced performance relative to their wild conspecifics. Unfortunately, the unavailability of length-matched adults did not allow us to test this notion unequivocally. Although Lee et al. (2003) compared among oceanranched wild or non-transgenic coho salmon, they found that adult GHf have a significantly lower Ucrit, at approximately 53 cm BL, than length-matched wild Coho salmon. If we assume that ocean-ranched wild coho salmon are similar to hatchery-reared wild coho salmon and extrapolate the findings from Lee et al. (2003) to later life history stages, because Ucrit velocities are employed during spawning migrations (Hinch et al., 2002) we speculate that GHf would be less likely to successfully complete a spawning migration than wild Coho salmon. Interestingly, Sundström et al. (2007) found that at approximately 4 cm BL, GHf had a lower propensity to migrate upstream. Whether or not this finding is also indicative of the behavior of adult GHf Coho salmon remains to be determined.  112  REFERENCES Abrahams, M.V. and A. Sutterlin. 1999. The foraging and antipredator behavior  Animal Behavior, 58:  of growth-enhanced transgenic Atlantic salmon. 933-42.  Axelsen, B.E., T. Anker-Nilssen, P. Fossum, C. Kvamme, and L. Nøttestad. 2001. Pretty patterns but a simple strategy: predator-prey interactions between juvenile herring and Atlantic puffins observed with multibeam sonar. Canadian Journal of Zoology, 79: 1586-96. Bainbridge, R. 1958. The speed of swimming of fish as related to size and to the frequency and amplitude of the tail beat. Journal of Experimental Biology, 35: 109-33. Bainbridge, R. 1960. Speed and stamina in three fish. Journal of Experimental  Biology, 37: 129-53. Baumman, G. 1997. Editorial: Growth without a Pituitary? – Lessons from the Guinea Pig. Endocrinology, 138(9): 3575-6. Blier, P.U., H. Lemieux, and R.H. Devlin. 2002. Is the growth rate of fish set by digestive enzymes or metabolic capacity of the tissues?  Insight from  transgenic coho salmon. Aquaculture, 209: 379-84. Brown, R.S. and D.R. Geist. 2002. Determination of Swimming Speeds and Energetic  Demands  of  Upriver  Migrating  Fall  Chinook  Salmon  (Oncorhynchus tshawytscha) in the Klickitat River, Washington. Bonnevill Power Administration, Portland, Ore. PNNL-12975.  113  Christensen, B. behaviors:  1996.  Predator foraging capabilities and prey antipredator  pre- versus postcapture constraints on size-dependent  predator-prey experiments. Oikos, 76: 368-80. Colavecchia, M., C. Katopodis, R. Goosney, D.A. Scruton, and R.S. Mckinley. 1998.  Measurement of burst swimming performance in wild atlantic  salmon (Salmo salar L.) using digital telemetry.  Regulated Rivers:  Research and Management, 14: 41-51. Coughlin, D.J. 2002. Aerobic muscle function during steady swimming in fish.  Fish and Fisheries, 3: 68-78. Deitch, E.J, G.L. Fletcher, L.H., L. H. Petersen, I. A. S. F. Costa, M. A. Shears, W. R. Driedzic, and A. K. Gamperl.  2006.  Cardiorespiratory  modifications, and limitations, in post-smolt growth hormone transgenic Atlantic salmon Salmo salar. Journal of Experimental Biology, 209: 131025. Devlin, R.H., B.K. McNeil, I.I. Solar, and E.M. Donaldson. 1994. A rapid PCRbased test for Y-chromosomal DNA allows simple production of all-female strains of chinook salmon. Aquaculture, 128: 211-20. Devlin, R.H., C.A. Biagi, and D.E. Smailus.  2001a.  Genetic mapping of Y-  chromosomal DNA markers in Pacific salmon. Genetica; 111: 43-58. Devlin, R.H., C.A. Biagi, T.Y. Yesaki, D.E. Smailus, and J.C. Byatt.  2001b.  Growth of domesticated transgenic fish. Nature, 409: 781-2. Devlin, R.H., C.A. Biagi, and T.Y. Yesaki. 2004. Growth, viability and genetic  114  characteristics of GH transgenic coho salmon strains. Aquaculture, 60732. Devlin, R.H., L.F. Sundström, and W.M. Muir. 2006. Interface of biotechnology and ecology for environmental risk assessments of transgenic fish.  Trends in Biotechnology, 24(2): 89-97. Dobson, G.P., W.S. Parkhouse, and P.W. Hochachka.  1987.  Regulation of  anaerobic ATP-generating pathways in trout fast-twitch skeletal muscle.  American Journal of Physiology: Regulatory Integrative and Comparative Physiology, 253: R186-94. Domenici, P. and R.W. Blake. 1997. The kinematics and performance of fish fast-start swimming. Journal of Experimental Biology, 200: 1165-78. Dytham, C. 2003. Choosing and Using Statistics: A biologist's guide. Blackwell Publishing, Malden, MA, pp. 195-8. Ellerby, D.J. and J.D. Altringham. 2001. Spatial variation in fast muscle function of the rainbow trout Oncorhynchus mykiss during fast-starts and sprinting.  Journal of Experimental Biology, 204: 2239-50. Etherton, T.D. and D.E. Bauman. 1998. Biology of Somatotropin in Growth and Lactation of Domestic Animals. Physiological Reviews, 78(3): 745-61. Farrell, A.P., W. Bennett, and R.H. Devlin. 1997. Growth-enhanced transgenic salmon can be inferior swimmers. Canadian Journal of Zoology, 75: 3357. Fitts, R.H.  1994.  Cellular Mechanisms of Muscle Fatigue.  Physiological  115  Reviews, 74 (1): 49-93. Florini, J.R., D.Z. Ewton, and S.A. Coolican. 1996. Growth Hormone and the Insulin-Like Growth Factor System in Myogenesis. Endocrine Reviews, 17 (6): 481-517. Fuiman, L.A. 1991. Influence of temperature on evasive responses of Atlantic herring larvae attacked by yearling herring, Clupea harengus L. Journal of  Fish Biology, 39: 93-102. Gallaugher, P.E., H. Thorarensen, A. Kiessling, and A.P. Farrell. 2001. Effects of high intensity exercise training on cardiovascular function, oxygen uptake, internal oxygen transport and osmotic balance in chinook salmon (Oncorhynchus tshawytscha) during critical speed swimming. Journal of  Experimental Biology, 204: 2861-2872. Glantz, S.A. 1997. Primer of Bio-Statistics (4th ed.). McGraw-Hill, New York, NY. pp. 238-9. Glova, G.J. and J.E. McInerney.  1977.  Critical swimming speeds of coho  salmon (Oncorhynchus kisutch) fry to smolt stages in relation to salinity and temperature. Journal of the Fish Research Board of Canada, 34: 151-4. Gregory, T.R. and C.M. Wood. 1998. Individual variation and interrelationships between swimming performance, growth rate, and feeding in juvenile rainbow trout (Oncorhynchus mykiss). Canadian Journal of Fisheries and  Aquatic Sciences, 55 (7): 1583-9.  116  Gibney, J., M-L. Healy, and P.H. Sönksen. 2007. The Growth Hormone/InsulinLike Growth Factor-I Axis in Exercise and Sport.  Endocrine Reviews,  28(6): 603-24. Goolish, E.M. 1991. Aerobic and anaerobic scaling in fish. Biological Reviews, 66: 33-56. Guinet, C., P. Domenici, R. de Stephanis, L. Barrett-Lennard, J.K.B. Ford, and P. Verborgh. 2007. Killer whale predation on bluefin tuna: exploring the hypothesis of the endurance-exhaustion technique.  Marine Ecology  Progress Series, 347: 111-9. Hale, M.E., J.H. Long, M.J. McHenry, and M.W. Westneat. 2002. Evolution of behavior and neural control of the fast-start escape response. Evolution, 56(5): 993-1007. Hill, J.A., A. Kiessling, and R.H. Devlin. 2000. Coho salmon (Oncorhynchus  kisutch) transgenice for a growth hormone gene construct exhibit increased rates of muscle hyperplasia and detectable levels of differential gene expression. Canadian Journal of Fish and Aquatic Sciences, 57: 939-50. Hinch, S.G., E.M. Standen, M.C. Healey, and A.P. Farrell. 2002. Swimming patterns and behavior of upriver-migrating adult pink (Oncorhynchus  gorhuscha) and sockeye (O. nerka) salmon as assessed by EMG telemetry in the Fraser River, British Columbia, Canada. Hydrobiologia, 483: 147-60.  117  James, R.S. and I.A. Johnston. 1998. The scaling of muscle performance during escape responses in the fish Myoxocephalus scorpius L. Journal of  Experimental Biology, 201:913-23. Jayne, B.C. and G.V. Lauder.  1993.  Red and white muscle activity and  kinematics of the escape response of the bluegill sunfish during swimming. Journal of Comparative Physiology A, 173: 495-508. Johnston, I.A., W. Davison, and G. Goldspink. 1977. Energy metabolism of carp swimming muscles. Journal of Comparative Physiology B, 114 (2): 20316. Katz, S.L. R.E. Shadwick, and H.S. Rapoport. 1999. Muscle strain histories in swimming milkfish in steady and sprinting gaits. Journal of Experimental Biology, 202: 529-41. Kopchick, J.J. and J.M. Andry. 2000. Minireview – Growth Hormone (GH), GH Receptor, and Signal Transduction. Molecular Genetics and Metabolism, 71: 293-314. Lauritzen, D.V, F. Hertel, and M.S. Gordon. 2005. A kinematic examination of wild sockeye salmon jumping up natural waterfalls.  Journal of Fish  Biology, 67: 1010-20. Lee, C.G., R.H. Devlin, and A.P. Farrell. 2003. Swimming performance, oxygen consumption and excess post-exercise oxygen consumption in adult transgenic and ocean-ranched coho salmon. Journal of Fish Biology, 62: 753-66.  118  Lefaucheur, L., A. Missohou, P. Ecolan, G. Monin, and M. Bonneau.  1992.  Performance, Plasma Hormones, Histochemical and Biochemical Muscle Traits, and Meat Quality of Pigs Administered Exogenous Somatotropin Between 30 or 60 Kilograms and 100 Kilograms Body Weight. Journal of  Animal Science, 70: 3401-11. Levesque, H.M., M.A. Shears, G.L. Fletcher, and T.W. Moon. 2008. Myogenesis and muscle metabolism in juvenile Atlantic salmon (Salmo salar) made transgenic for growth hormone.  Journal of Experimental Biology, 211:  128-37. Lundvall, D., R. Svanbäck, L. Persson, and P. Byström. 1999. Size-dependent predation in piscivores: interactions between predator foraging and prey avoidance abilities. Canadian Journal of Fish and Aquatic Sciences, 56: 1285-92. Marten, K., D. Herzing, M. Poole, and K.N. Allman. predation hypothesis:  2001.  The acoustic  linking underwater observations and recordings  during odontocete predation and observing the effects of loud impulsive souns on fish. Aquatic Mammals, 27(1): 56-66. McDonald, D.G., W.J. McFarlane, and C.L. Milligan. 1998. Anaerobic capacity and swim performance of juvenile salmonids. Canadian Journal of  Fisheries and Aquatic Sciences, 55 (5): 1198-1207. McKenzie, D.J., S. Wong, D.J. Randall, S. Egginton, E.W. Taylor, and A.P. Farrell. 2004. The effects of sustained exercise and hypoxia upon oxygen  119  tensions in the red muscle of rainbow trout.  Journal of Experimental  Biology, 207: 3629-37. Moyes, C.D. and C.M.R. LeMoine. 2005. Control of muscle bioenergetic gene expression: implications for allometric scaling relationships of glycolytic and oxidative enzymes. Journal of Experimental Biology, 208: 1601-10. Nelson, J.A. and G. Claireaux. 2005. Sprint Swimming Performance of Juvenile European Sea Bass.  Transactions of the American Fisheries Society,  134: 1274-84. Nelson,  J.A.,  P.S.  Gotwalt,  C.A.  Simonetti,  J.W.  Snodgrass.  2008.  Environmental Correlates, Plasticity, and Repeatability of Differences in Performance among Blacknose Dace (Rhinichthys atratulus) Populations across a Gradient of Urbanization.  Physiological and Biochemical  Zoology, 81 (1): 24-42. Nøttestad, L and B.E. Axelsen. Herring schooling manoeuvres in response to killer whale attacks. Canadian Journal of Zoology, 77: 1540-6. Oakes, J.D., D.A. Higgs, J.G. Eales, and R.H. Devlin. 2007. Influence of ration level on the growth performance and body composition of non-transgenic and growth-hormone-transgenic coho salmon (Oncorhynchus kisutch).  Aquaculture, 265: 309-24. Oksbjerg, N., J.S. Petersen, M.T. Sórensen, P. Henckel, N. Agergaard, C. Bejerholm, and E. Erlandsen. 1995. The Influence of Porcine Growth Hormone on Muscle Fibre Characteristics, Metabolic Potential and Meat  120  Quality. Meat Science, 39: 375-85. Raven, P.A., R.H. Devlin, and D.A. Higgs. 2006. Influence of dietary digestible energy utilization and body composition of growth hormone transgenic and non-transgenic coho salmon (Oncorhynchus kisutch). Aquaculture, 254: 730-47. Raven, P.A., M. Uh, D. Sakhani, B.R. Beckman, K. Cooper, J. Pinter, E.H. Leder, J. Silverstein, and R.H. Devlin.  2008.  Endocrine effects of growth  hormone overexpression in transgenic coho salmon.  General and  Comparative Endocrinology, 159: 26-37. Randal, D.J., D. Mense, and R.G. Boutilier. 1987. The effects of burst swimming on aerobic swimming in chinook salmon (Oncorhynchus tshawytscha). Marine Behavioral Physiology, 13: 77-88. Reidy, S.P., S.R. Kerr, and J.A. Nelson. 2000. Aerobic and anaerobic swimming performance of individual atlantic cod. Journal of Experimental Biology, 203: 347-57. Pell J.M. and P.C. Bates. 1987. Collagen and non-collagen protein turnover in skeletal muscle of growth hormone-treated rats. Journal of Endocrinology, 115: R1-4. Pelletier, D., H. Guderley, and J.D. Dutil. 1993a. Does aerobic capacity of fish muscle change with growth rates? Fish Physiology and Biochemistry, 12 (2): 83-93. Pelletier, D., H. Guderley, and J.D. Dutil.  1993b.  Effects of Growth Rate,  121  Temperature, Season, and Body Size on Glycolytic Enzyme Activities in the White Muscle of Atlantic Cod (Gadus morhua).  The Journal of  Experimental Zoology, 265: 477-87. Peyreigne, C., E. Raynaud, C. Fedou, C. Prefaut, J.F. Brun, and J. Mercier. 2002.  Does Growth Hormone Treatment Alter Skeletal Muscle  Mitochondrial Respiration in Rats? Hormone Research, 58: 287-91. Pew Initiative.  2003.  Future Fish:  Issue in Science and Regulation of  Transgenic Fish. Pew Initiative on Food and Biotechnology. http://www.pewtrusts.org/our_work_detail.aspx?id=442 Rescan, P.Y. 2005. Muscle growth patterns and regulation during fish ontogeny.  General and Comparative Endocrinology, 142: 111-6. Reidy, S.P., S.R. Kerr, and J.A. Nelson. 2000. Aerobic and anaerobic swimming performance of individual Atlantic cod. Journal of Experimental Biology, 203: 347-57. Richards, J.G., G.J.F. Heigenhauser, and C.M. Wood.  2002a.  Glycogen  phosphorylase and pyruvate dehydrogenase transformation in white muscle of trout during high-intensity exercise.  American Journal of  Physiology: Regulatory Integrative and Comparative Physiology, 282: R828-36. Richards, J.G., A.J. Mercado, C.A. Clayton, G.J.F. Heigenhauser, and C.M. Wood. 2002b. Substarte utilization during graded aerobic exercise in rainbow trout. Journal of Experimental Biology, 206: 2067-77.  122  Savitz, J. and L. Bardygula-Nonn. 1997. Behavioral interactions between coho (Oncorhynchus kisutch) and chinook salmon (Oncorhynchus shawytscha) and prey fish species. Ecology of Freshwater Fish, 6: 190-5. Semsarian, C., P. Sutrave, D.R. Richmond, and R.M. Graham. 1999. Insulin-like growth factor (IGF-I) induces myotube hypertrophy associated with an increase in anaerobic glycolysis in a clonal skeletal-muscle cell model.  Biochemical Journal, 339: 443-51. Shadwick, R.E. 2005. How tunas and lamnid sharks swim: An evolutionary convergence. American Scientist, 93(6): 524-31. Short, K.R., N. Moller, M.L. Bigelow, J.Coene-Schimke, and K.S. Nair. 2008. Enhancement of Muscle Mitochondrial Function by Growth Hormone.  Journal of Clinical Endocrinology and Metabolism, 93(2): 597-604. Stevens, E.D. A. Sutterlin, and T. Cook. 1998. Respiratory metabolism and swimming performance in growth hormone transgenic Atlantic salmon.  Canadian Journal of Fisheries and Aquatic Sciences, 55: 2028-35. Sundström, L.F., M. Lõhmus, J.I. Johnsson, and R.H. Devlin. 2007. Dispersal Potential is Affected by Growth-Hormone Transgenesis in Coho salmon (Oncorhynchus kisutch). Ethology, 113: 403-10. Turesson, H. and C. Brönmark. 2004. Foraging behavior and capture success in perch, pikeperch and pike and the effects of prey density. Journal of Fish  Biology, 65: 363-75. USFDA. 2009. Guidance for Industry: Regulation of Genetically Engineered  123  Animals Containing Heritable Recombinant DNA Constructs.  US Food  and  Medicine.  Drug  Administration  –  Center  for  Veterinary  <http://www.fda.gov/cvm/geanimals.htm> Vasilatos-Younken, R., Y. Zhou, X. Wang, J.P. McMurtry, R.W. Rosebrough, E. Decuypere, N. Buys, V.M. Darras, S. Van der Geyten, and F. Tomas. 2000.  Altered chicken thyroid hormone metabolism with chronic GH  enhancement in vivo: consequences for skeletal muscle growth. Journal  of Endocrinology, 166: 609-20. Velloso, C.P. 2008. Regulation of muscle mass by growth hormone and IFG-1.  British Journal of Pharmacology, 154: 557-68. Vestergaard, M., S. Purup, P. Henckel, E. Tonner, D.J. Flint, L.R. Jensen, and K. Sejrsen.  1995.  Effects of Growth Hormone and Ovariectomy on  Performance Serum Hormones, Insulin-Like Growth Factor-Binding Proteins, and Muscle Fiber Properties of Prepubertal Friesian Heifers.  Journal of Animal Science, 73: 3574-84. P.W. Webb. 1976. The effect of size on the fast-start performance of rainbow trout Salmo gairdneri, and a consideration of piscivorous predator-prey interactions. Journal of Experimental Biology, 65: 157-77. P.W. Webb. 1978. Fast-start performance and body form in seven species of teleost fish. Journal of Experimental Biology, 74: 211-26.  124  CHAPTER 4: CONCLUSION  Salmon transgenic for GH upregulation may be used in aquaculture in the future. However, in order to market such GE animals, the ecological impact of potential GH escapees, relative to their non-GE conspecifics must be considered (Pew Initiative, 2003; Devlin et al., 2006; USFDA, 2009). To that end, we used fast-starts and sprinting, to infer the prospective ability of GH-transgenic Coho salmon to evade predatory strikes and chasing predators, to chase prey, and to successfully complete spawning migrations, respectively.  Because their  transgenesis employs GH we have also considered the effects of GH upregulation on muscle functioning during fast-starts and sprinting. The present study is the first to consider the kinematics of sprinting to exhaustion, and as such we discussed the potential mechanisms of exhaustion during sprinting. Lastly, because sprinting has been seldom discussed within an ecological context, we also presented the potential uses for and the importance of sprinting in the wild. Fast-starts are used to evade predatory strikes (Domenici, 2003). Between juvenile GH-transgenic and wild Coho salmon we found no differences in the mechanics of and the resultant escape velocity achieved during fast-starts. Thus, we suggest that GH-transgenic and wild Coho salmon have a similar ability to evade predatory strikes; however, because GHf also exhibited a lower propensity for antipredator behavior when in the presence of a predator  125  (Sundström et al., 2007) they may realize higher mortality. Additionally, due to their similar fast-start performance, GH may have no effect on affectors of faststart performance:  short-term behavior, intrinsic muscle properties and  excitation-contraction coupling, musculoskeletal linking, and body shape. Muscle fibre-type changes have been reported between juvenile GHf and wild Coho salmon (Hill et al., 2000); however, we suggest the absolute magnitude of those differences had little or no functional significance. No such data is available among adults to which we might compare our findings. Changes in muscle fibre-types have the potential to impact locomotor performance (such as fast-starts) and may offer insight concerning the mechanisms of GH action on skeletal muscle.  We summarized changes in  muscle fibre-types due to GH administration across a variety of species, relative to their normal, untreated, or wild conspecifics (see table 2.5). Among affected muscles from rats, mice, and fish, GH treatment resulted in higher proportions of type 1, slow-twitch, or red muscle fibres. In contrast, the affected muscles of GH treated cows and pigs realized higher proportions of type 2b or fast-twitch glycolytic muscle fibre-types. Interestingly, rodents and fish were fed ad libitum or to satiation while cows and pigs were fed less than their untreated conspecifics. We suggest that the observed changes in muscle fibre-types may occur in concert with the effect of GH administration and feed-levels on liver metabolism by altering the availability of metabolic substrates in circulation.  126  Sprinting is used during spawning migrations by salmonids (at least those which undergo such migrations, such as Coho salmon) to successfully negotiate challenging reaches such as waterfalls (Hinch et al., 2002). We also presented evidence, from previous studies that demonstrate the importance of sprinting during predator-prey interactions by both chasing predators and escaping prey. In addition to the affectors of fast-starts, sprinting is also affected by the capacity of the white axial muscle to support ATP turnover.  We found that during  exhaustive sprinting individuals employed a range of velocities, which suggests that the type 1, slow-twitch, or red lateral muscle also contributed. Juvenile GHtransgenic and wild Coho salmon exhibited similar kinematics, velocity, and net anaerobic substrate usage; however, GHf (GH-transgenic fed to satiation) took significantly longer to reach exhaustion. While previous research showed that GHf exhibited a lower propensity for antipredator behavior when in the presence of a predator (Sundström et al., 2007), we speculate that juvenile GH-transgenic Coho escapees may have an enhanced ability to evade chasing predators and to chase prey relative to their wild conspecifics. Previous research also revealed that length-matched juvenile GHf have higher activity levels of aerobic and anaerobic enzymes within their white axial muscle (Hill et al., 2000; Blier et al., 2002). This offers an explicit explanation as how juvenile GHf might realize similar kinematics, velocity, and net anaerobic substrate usage while still taking longer to reach exhaustion: they had a higher capacity to support ATP turnover. No such data are presently available among  127  adults.  We also summarized the effects of GH administration on muscle  enzymes among humans, cows, pigs, and fish, and found no clear patterns regarding changes in activity levels thereof (see table 3.8). Between adult, GHf and wild Coho salmon, we found no differences in time to exhaustion; however, because we were unable to obtain length-matched cohorts, a concise conclusion concerning velocity, particularly maximum velocity, remains elusive. That said, at best, adult GH-transgenic Coho escapees might exhibit similar sprinting performance compared to wild Coho salmon. Therefore, because no information is available concerning antipredator behavior among adults, we conclude that, at best, GH-transgenic Coho may exhibit a similar ability to evade chasing predators, as well as to chase prey. Moreover, if we extrapolate our maximum velocity results to later life history stages, at best, GHf may be equally capable of successfully negotiating challenging reaches encountered during spawning migrations. Finally, because GH-transgenic Coho exhibited a lower Ucrit than wild Coho salmon throughout their life history (Farrell et al., 1997; Lee et al., 2003), if we extrapolate those results to later life history stages, GHf may have a reduced ability to successfully complete spawning migrations.  128  REFERENCES Blier, P.U., H. Lemieux, and R.H. Devlin. 2002. Is the growth rate of fish set by digestive enzymes or metabolic capacity of the tissues?  Insight from  transgenic coho salmon. Aquaculture, 209: 379-84. Devlin, R.H., L.F. Sundström, and W.M. Muir. 2006. Interface of biotechnology and ecology for environmental risk assessments of transgenic fish.  Trends in Biotechnology, 24(2): 89-97. Domenici, P. 2003. Habitat, body design, and swimming performance of fish.  Vertebrate Biomechanics and Evolution. Eds. V.L. Bels, J.P. Gasc and A. Casinos. BIOS Scientific Publishers Ltd, Oxford. Pg. 137-60. Farrell, A.P., W. Bennett, and R.H. Devlin. 1997. Growth-enhanced transgenic salmon can be inferior swimmers. Canadian Journal of Zoology, 75: 3357. Hill, J.A., A. Kiessling, and R.H. Devlin. 2000. Coho salmon (Oncorhynchus  kisutch) transgenice for a growth hormone gene construct exhibit increased rates of muscle hyperplasia and detectable levels of differential gene expression. Hinch, S.G., E.M. Standen, M.C. Healey, and A.P. Farrell. 2002. Swimming patterns and behavior of upriver-migrating adult pink (Oncorhynchus  gorhuscha) and sockeye (O. nerka) salmon as assessed by EMG telemetry in the Fraser River, British Columbia, Canada. Hydrobiologia, 483: 147-60.  129  Lee, C.G., R.H. Devlin, and A.P. Farrell. 2003. Swimming performance, oxygen consumption and excess post-exercise oxygen consumption in adult transgenic and ocean-ranched coho salmon. Journal of Fish Biology, 62: 753-66. Pew Initiative.  2003.  Future Fish:  Issue in Science and Regulation of  Transgenic Fish. Pew Initiative on Food and Biotechnology. <http://www.pewtrusts.org/our_work_detail.aspx?id=442> Sundström, L.F., M. Lõhmus, J.I. Johnsson, and R.H. Devlin. 2007. Dispersal Potential is Affect by Growth-Hormone Transgenesis in Coho Salmon (Oncorhynchus kisutch). Ethology, 113: 403-10. USFDA. 2009. Guidance for Industry: Regulation of Genetically Engineered Animals Containing Heritable Recombinant DNA Constructs.  US Food  and  Medicine.  Drug  Administration  –  Center  for  Veterinary  <http://www.fda.gov/cvm/geanimals.htm>  130  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0067153/manifest

Comment

Related Items