Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Reproductive performance of growth-enhanced transgenic coho salmon (Oncorhynchus kisutch) Bessey, Cindy 2003

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

Item Metadata


831-ubc_2003-0444.pdf [ 3.05MB ]
JSON: 831-1.0091025.json
JSON-LD: 831-1.0091025-ld.json
RDF/XML (Pretty): 831-1.0091025-rdf.xml
RDF/JSON: 831-1.0091025-rdf.json
Turtle: 831-1.0091025-turtle.txt
N-Triples: 831-1.0091025-rdf-ntriples.txt
Original Record: 831-1.0091025-source.json
Full Text

Full Text

REPRODUCTIVE PERFORMANCE OF GROWTH-ENHANCED TRANSGENIC COHO SALMON (Oncorhynchus kisutch)  by CINDY BESSEY B.Sc., Simon Fraser University, 2000  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIRMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Zoology)  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA July 2003 © CINDY BESSEY, 2003  In  presenting  degree freely  at  this  the  available  copying  of  department publication  of  in  partial  fulfilment  of  the  University  of  British  Columbia,  I  agree  for  this or  thesis  reference  thesis by  this  for  his thesis  and  study.  scholarly  or for  her  financial  Department T h e U n i v e r s i t y o f British C o l u m b i a Vancouver, Canada ,  (2/88)  further  purposes  gain  shall  that  agree  may  representatives.  permission.  DE-6  I  requirements  It not  be is  that  the  Library  permission  granted  by  understood be  for  allowed  an  advanced  shall for  the that  without  make  it  extensive  head  of  my  copying  or  my  written  ABSTRACT t  The reproductive performance of growth-enhanced transgenic and nontransgenic coho salmon (Oncorhynchus kisutch) was examined to address concerns associated with using genetically modified fish in aquaculture. A major concern is the initial reproductive interaction between transgenic and wild fish that may occur if transgenic salmon escape into the natural environment. This thesis attempted to obtain an indication of reproductive allocation, effort and success for transgenic and nontransgenic salmon in a simulated natural environment. Several aspects of reproduction are examined for both transgenic and nontransgenic fish to provide an indication of gamete quantity and quality, courtship behaviour, spawning success, transgene transmission to offspring, and male competitive ability. Female and male body size, shape, and gonadal somatic index (GSI) provided a measure of phenotype and reproductive allocation. Female fecundity and egg diameter, as well as male sperm production, fertility and competition examined gamete quantity and quality. Female digging, probing, and covering, as well as male quivering examined courtship behaviour of ovulated females and ripe males paired together in spawning channels. Spawning success was recorded and fertilized eggs were collected and raised to alevin in order to examine offspring viability. Polymerase chain reactions were conducted on offspring blood samples to determine transgene transmission. Biting, chasing and spawning success of male pairs placed together with an ovulated female were analysed to determine male competitive interactions.  Results from these studies found several differences between transgenic and nontransgenic fish. At maturation, transgenic males lacked red coloration and had a less developed kype compared to nontransgenic fish, but no differences in male gamete quantity or quality were observed. Transgenic females were more fecund than nontransgenic females, but may have inferior quality gametes due to reduced egg size. Transgenic females spawned less frequently and displayed consistently low levels of courtship behaviour. No courtship behaviour differences between transgenic and nontrangenic males were observed, however, during competition, transgenic males were inferior; obtaining no spawnings, and displaying less courtship and competitive behaviour. These studies are thefirstto show that in a simulated natural environment, growth enhanced transgenic coho salmon display courtship behaviour and can spawn producing viable transgenic offspring.  iv T A B L E OF CONTENTS Abstract  ii  Table of Contents  iv  List of Figures  v  List of Tables  vii  Acknowledgments  viii  GENERAL INTRODUCTION  1  Chapter 1 ADULT PHENOTYPE, AND GAMETE QUANTITY AND QUALITY 1.1 1.2 1.3 1.4  Introduction Materials and Methods Results Discussion  8 10 16 24  Chapter 2 COURTSHIP AND SPAWNING BEHAVIOUR ACCOMPANIED WITH GENE TRANSMISSION 2.1 2.2 2.3 2.4  Introduction Materials and Methods Results Discussion  32 33 38 45  Chapter 3 MALE COMPETITIVE BEHAVIOUR 3.1 3.2 3.3 3.4  Introduction Materials and Methods Results Discussion  49 50 52 58  GENERAL SUMMARY  61  Bibliography  64  LIST OF FIGURES 1. Figure 1.1: Fecundity as a function offishweight for 2001 -N, 2002-N, T, and C females 2. Figure 1.2: Diameter of eggs (mm) as a function of fish weight for 2001-N, 2002-N, T, and C females 3. Figure 1.3: Male gonad weight (g) as a function of fish weight for 2001-N, 2002-N, T, and C males 4. Figure 2.1: Laboratory Spawning Channel: Solid square arrows represent water pumps and open arrows indicate the direction of water flow. Video camera and swim channel measurements are illustrated 5. Figure 2.2: Female Digging: Mean digs / 5 minutes (± standard error) conducted at 30, 20, and 10 minutes prior to spawning, during spawning, and 10 and 20 minutes following spawning (-30, -20, -10, 0, 10 and 20) 6. Figure 2.3: Female Probing: Mean probes / 5 minutes (± standard error) conducted at 30, 20, and 10 minutes prior to spawning, during spawning, and 10 and 20 minutes following spawning (-30, -20, -10, 0, 10 and 20)  7. Figure 2.4: Female Covering: Mean covers / 5 minutes (± standard error) conducted at 30, 20, and 10 minutes prior to spawning, during spawning, and 10 and 20 minutes following spawning (-30, -20, -10, 0, 10 and 20) 8. Figure 2.5: Male Quivering: Mean quivers / 5 minutes (± standard error) conducted at 30, 20, and 10 minutes prior to spawning, during spawning, and 10 and 20 minutes following spawning (-30, -20, -10, 0, 10 and 20) 9. Figure 3.1: Competitive Courtship Behaviour of Nontransgenic Versus Transgenic Male: Mean quivers / 5 minutes (+ standard error) conducted during competition between N and T males at 30, 20, and 10 minutes prior to spawning, during spawning, and 10 and 20 minutes following spawning (-30, -20,-10,0,10 and 20) 10. Figure 3.2: Competitive Courtship Behaviour of Nontransgenic Versus Cultured Male: Mean quivers / 5 minutes (±  standard error) conducted during competition between N and C males at 30, 20, and 10 minutes prior to spawning, during spawning, and 10 and 20 minutes following spawning (-30, -20,-10,0,10 and 20) 11. Figure 3.3: Male Bites During Competition: Number of bites per 30 minutes of observations over a period of one hour around the spawning event (± standard error). Biting behaviour was conducted in competition between N versus T, and N versus C males 12. Figure 3.4: Male Chases During Competition: Number of chases per 30 minutes of observations over a period of one hour around the spawning event (± standard error). Chasing behaviour was conducted in competition between N versus T, and N versus C males  LIST OF TABLES 1. Table 1.1: Summary of the rearing environments, age at maturity and presence of the transgene for each sample population 2. Table 1.2: Adult Phenotype: Sample size (N), body weight, standard length, weighfclength linear regression (r ) and equation, and condition factor (CF) for all fish sampled accompanied with one standard deviation (SD) 2  3. Table 1.3: Gamete Quantity: Sample size (N), body weight, gonadal somatic index (GSI), egg weight, fecundity, egg diameter and spermatocrit ( ± one standard deviation) 4. Table 1.4: Sperm Competition Trials: Trials, replicates, stock pair number, number of surviving alevin, and percent of offspring positive for the transgene accompanied by the 95% confidence interval for the proportion of transgenic offspring 5. Table 1.5: Sperm Fertilization Trials: Trials, replicates, stock and pair number, number of surviving alevin, and percent of offspring positive for the transgene accompanied by the 95% confidence interval for the proportion of transgenic offspring 6. Table 2.1: Number of Courtship Behaviour Pairs Observed: For each pair type, the number of pairs that spawned is shown in brackets, and followed by the calculated percentage 7. Table 2.2: Direct Visual Courtship Behaviours Regardless of Spawning Occurrence: Number of male and female courtship behaviours for 30 minutes of direct visual observation over a six hour period (+ standard error) 8. Table 2.3: Transmission of the Transgene: The percent of offspring that tested positive for the transgene from three pairs with at least one transgenic parent, the number of offspring tested, and p values from the binomial test 9. Table 3.1: Male Competition Pairs and Spawning Success: The total number of pairs utilized during competition trials accompanied by the mean body weight (kg) + one standard deviation (SD), and the genotype of the successful spawning male  Vlll  ACKNOWLEDGEMENTS First and foremost, I would like to thank Dr. Robert H. Devlin for his many years of guidance, support and encouragement, as well as, for the countless opportunities and scientific freedom that I was granted. I consider myself truly privileged to have had such an incredible mentor. Secondly, I would like to thank Carlo A. Biagi. Carlo's technical assistance and constant positive reinforcement was key in any success I may have had over my many years at the Department of Fisheries and Oceans. I would also like to thank Dr. Robin N. Liley for use of his facilities and his valuable editorial comments, and Dr. Rick Taylor and Dr. Mike Healey for statistical and scientific support. I would like to acknowledge the staff at the following facilities for all their assistance: Department of Fisheries and Oceans in West Vancouver, the University of British Columbia's Animal Care Unit, as well as, the Chehalis River Hatchery. Special thanks are extended to Dr. George Watters and Jefferson Hinke at the Pacific Fisheries Environmental Laboratory for their editorial comments and suggestions for my thesis. Finally, I would like to thank all my friends and family in both California and Vancouver for their invaluable moral support. Financial support for this work was provided by a grant to Dr. Devlin from the Canadian Biotechnology Strategy.  1 GENERAL INTRODUCTION  The production of growth-enhanced transgenic fish is being considered as a method to increase production efficiency for aquaculture products, but little is known of the potential effects these fish may have on natural populations if they escape from aquaculture facilities (Devlin 1998, Grainger 1998, Donaldson 1997). With increasing demand for aquatic food products and limited resources to support their production, increased production efficiency will provide an economic advantage for the aquaculture industry. Increasing the growth rate of cultured fish can be an important means of influencing production efficiency. Increased growth rate affects the economics of fish production by allowing more rapid returns on capital investment and by enhancing feed conversion efficiency (FCE) (McBride et al. 1982, McLean et al. 1996). Feed costs can amount to more than half the cost of raising fish in aquaculture, and consequently, even small improvements in FCE can dramatically affect the sustainability of aquaculture operations. There are, however, many issues associated with transgenic fish that must be addressed prior to the implementation of such biotechnology (Devlin and Donaldson 1992, Kapuscinski and Hallerman 1991). This thesis proposes to address some of the issues associated with the use of growth enhanced transgenic fish in aquaculture. Growth rates of fish can be enhanced in many ways, including increased nutrition, improved husbandry, genetic selection programs, and hormone treatment. For the latter approach, hormones can be injected directly (Sower et al. 1983), orfishcan be forced to increase endogenous hormone production by inserting gene constructs into their genome (Vielkind et al. 1971). Transgenicfishproduced by the introduction of such gene constructs can result in desirable phenotypes for a number of different fish species  2 (review by Devlin, 1997; i.e. rainbow trout, Chourrout et al., 1986, loach, Zhu et al., 1986, Atlantic salmon, Fletcher et al., 1988 and Pacific salmon, Devlin et al., 1994a, etc.). Substantial effort has been expended into producing transgenic fish, and these fish are known to have altered growth and behaviour patterns (Johnsson, 1996, Abrahams and Sutterlin, 1999, and Devlin et al. 1999). Both, Pacific and Atlantic transgenic salmon have been produced with dramatically increased growth performance (Devlin et al. 1995a, and Fletcher et al. 1988). These fish could potentially be used in the aquaculture industry to reduce production times and improve FCEs. It is also possible that production of transgenic salmon could alleviate some of the pressures that currently exist on natural stocks. There are, however, many economic, social and environmental issues regarding the commercial use of genetically modified fish in aquaculture that need to be addressed (Pandian and Marian, 1994, Devlin and Donaldson 1992, Kapuscinski and Hallerman 1991, Tiedje et al. 1989). Flesh quality, health, and behavioural performance are all factors that may determine if transgenic fish would offer an economic advantage to the aquaculture industry. An important social issue is the potential health risks that consumption of transgenic fish may present to humans. Major environmental concerns are the direct impact (i.e. via predation and competition for food resources and breeding grounds) of these fish on natural fish populations if transgenic fish are released into the natural environment, as well as the sustained genetic impact on the natural stocks if transgenic fish integrate and reproduce with wild stocks. Another environmental issue arises because transgenic fish are farmed or raised in culture. There is considerable concern over the impact of farmed fish on wild  3 populations, regardless of whether the farmed fish are transgenic. Farmed salmon may threaten wild populations both ecologically (i.e. by competition and disease introductions) and genetically (i.e. by genetic homogenisation and loss of local adaptation) (Fleming and Einum, 1997, Fleming et al. 1996, Fleming and Gross, 1993). The behaviour of cultured salmonids may differ from wild fish in a number of important fitness related traits (review by Jonsson, 1997). For example, cultured females make fewer nests, retain more unspawned eggs, and males display less combative and courtship behaviour and have more difficulty in acquiring access to mates. The culture environment, the transgene, as well as the combination of the two, may affect the reproductive ability of transgenic fish, but little is known of such interactions. The longterm multigenerational effects of such interactions is likely to depend on the fitness of transgenic fish, and this topic is largely unexplored. Since transgenic fish lack a history of natural selection pressures, it is difficult to accurately predict the fate of a transgene in nature (Devlin and Donaldson 1992, Muir and Howard, 1999, and Devlin 2000). Nevertheless, the potential effects of the transgene on fitness must be examined because any such effects will ultimately determine if transgenic fish have a sustained genetic impact on wild populations. Attempts have been made to address the possible threat of transgenic fish to natural populations by looking at the putative effect of transgenes on mating success (Muir and Howard, 1999, and Hedrick, 2001). Using deterministic models, these studies predict that transgenic fish will invade natural populations due to increased mating success, and local extinction of both the wild and transgenic populations may eventually occur (due to reduced viability of transgenic offspring). These studies, however, are based on the assumption that  4  transgenic fish would have a mating advantage due to their increased size. Although it is known that larger males adopt superior positions in the mating hierarchy during competition for females (Healey and Prince, 1998), mating success does not only depend on the size of the fish. Mating success also depends on body shape and coloration, intensity of courtship and competitive interactions, as well as an ability to migrate to and from the spawning grounds (Houde 1997, Jaervi, 1990). Therefore, it seems that existing models could be improved by obtaining an extensive experimental assessment of the behavioural and competitive abilities of transgenic fish. Measuring fitness is complex since there are many components that can influence the ability of an organism to contribute to the next generation. Fitness is determined by the ability of the fish to grow, survive, and reproduce compared with other individuals in the population. The fitness of transgenic fish depends not only on the effect of the transgene itself, but is also dependent upon the interacting effects of the culture environment with the transgene. Environmental conditions can also affect behaviour, and consequently fitness can be influenced by genotype-environment interactions (Devlin, 2000). The West Vancouver Laboratory of the Department of Fisheries and Oceans in British Columbia, Canada, has used microinjection techniques to produce growthenhanced transgenic coho salmon (Oncorhynchus kisutch) for the purpose of evaluating the potential risks that transgenic fish pose to humans, wild fish stocks, and the environment. These fish have also been produced to investigate gene function. During production, a gene construct is injected into the cytoplasm of a fertilized salmon egg (Chehalis River strain), that was developmentally arrested immediately after fertilization,  5 near the male and female pronuclei (Devlin et al. 1995b). The gene construct (OnMTGHl) is composed of a metallothionein-B promoter (MT) fused to a full-length, type-1 growth hormone gene (GH), both obtained from a sockeye salmon. Although it is not fully understood how the gene construct is incorporated into the genome, it seems to be randomly inserted into the host's chromosomes. Transgenic coho salmon display enhanced growth rates, and reach full size and maturity in two years. In nature, wild strain Chehalis coho mature in three years, and in the laboratory, thesefishcan take up to four years to mature. The gene construct can cause various degrees of growth enhancement, presumably depending on the different chromosomal site of insertion and number of gene copies. Research has shown that transgenic coho salmon differ from nontransgenic coho salmon in a number of ways. These include earlier smoltification, increased ability to compete for food in a controlled tank setting, inferior swimming ability, significantly lower GH (growth hormone) mRNA levels and smaller pituitary glands (Devlin et al. 1994a, Devlin et al. 1999, Farrell et al. 1997, and Mori and Devlin. 1999). Transgenic coho may also show morphological abnormalities that result in acromegaly syndromes (Ostenfeld et al. 1998). Given this list of effects and concern that transgenic fish might adversely affect wild populations, research has also been done on ways to ensure that transgenic fish are contained. Several methods of containment, both physical and biological (i.e. inducing triploidy, treatment with androgen, radiation treatment, autoimmune sterility chemosterilization) have been suggested (review by Devlin and Donaldson, 1992).  6  This thesis is focussed on the reproductive performance o f growth-enhanced transgenic salmon. The primary goal is to observe, under laboratory conditions, the initial reproductive interactions o f transgenic fish with fish found i n the wild. In particular, I investigate whether transgenic fish are capable o f courting and spawning with fish found i n the w i l d , and i f so, do transgenic fish transmit the transgene to viable offspring. A s well, I investigate whether transgenic males can successfully compete for access to females against males found in the wild. Such interactions may occur i f cultured transgenic salmon escape into the natural environment. A secondary goal is to determine i f differences in reproductive performance are a consequence o f the transgene or the culture environment. I examine the spawning behaviour and potential reproductive output o f growth-enhanced transgenic coho salmon derived from the Chehalis river population. Laboratory raised transgenic fish (T) are compared with anadromous nontransgenic (N) fish from the same population, and with non-transgenic fish cultured under laboratory conditions (C). Since the observations were made in the laboratory, this study provides only an approximation to the behaviour o f transgenic and cultured fish and their possible interaction with w i l d fish in the natural environment. Nevertheless, these observations provide a significant contribution to the knowledge of the reproductive success o f G H transgenic animals. The thesis is composed o f three chapters. In Chapter 11 compare adult phenotype and potential reproductive output o f T, N and C stocks by examining female gonadal somatic index (GSI), fecundity and egg diameter. I also consider male G S I , sperm production, fertility and competition. In Chapter 2 I examine male and female courtship and spawning behaviour of T, N and C stocks, as well as transgene transmission from T  fish to their offspring. Finally, in Chapter 3,1 examine male competitive behaviour between N and T, N and C, and T and C pairs.  8 1 ADULT PHENOTYPE, AND G A M E T E QUANTITY AND QUALITY 1.1 INTRODUCTION In order to contribute fertile offspring to the next generation, fish must possess gametes that can produce viable offspring. Both, gamete quantity and quality are key factors that contribute to fitness. Although measurements of fitness are difficult to obtain, the ratio of gonad weight to body weight, or GSI (gonadal somatic index), and fecundity data are typically used to assess reproductive allocation and provide a quantitative measure of reproductive effort (Begon et al. 1996, Helfman et al. 1997). Spermatocrit is correlated with number of sperm cells per unit volume of milt and also provides a measure of gamete quantity (Bouk and Jacobson, 1976). Since highly fecund organisms with low quality gametes may have low fitness, it is also important to consider gamete quality. For females, egg size may determine quality, and, for males, the ability of milt to fertilize and compete for access to eggs may determine quality. There are many studies examining the gamete quantity and quality of wild salmonids (Allen, 1958, Beacham, 1982, Stockley, 1997, Hoysak and Liley, 2001), but such studies have not been published for transgenic salmon. Phenotypic traits like body size may influence reproductive success by affecting the factors contributing to gamete quantity and quality. For example, there is a positive correlation between fecundity and adult body size in salmonids (Foerster and Pritchard, 1941, Drucker, 1972). Egg size is also positively correlated with fish size (Fleming and Gross, 1990) and offspring survival (Einum and Fleming, 2000). A trade-off exists between egg size and fecundity (Svardson, 1949); females must balance the number of  9 individuals that can be produced (by investing in fecundity) against their potential survival (by investing in egg size). Sources of variation in traits determining gamete quantity and quality are influenced by both genetic and environmental factors. The fecundity of salmonid fishes varies between populations of the same species, as well as, those from different localities (review by Crone and Bond, 1976, Fleming and Gross, 1990). It was previously assumed that the range in fecundity for each stock is genetically determined, but the possibility that variation in fecundity is due to environmental conditions has also been considered (Allen, 1958, Kinnison et al. 2001). Indeed, the biological significance and cause of these differences in fecundity between populations of the same species has been addressed for over 50 years (Rounsefell, 1957). It is difficult to partition the variance in gamete quantity and quality between genetic and environmental factors. This task requires fish of different strains to be reared under the same environmental conditions. The present chapter focuses on two goals. First, I examine differences in adult phenotype, gamete quantity, and gamete quality between nontransgenic and transgenic fish, to consider potential differences in reproductive output of these two genotypes. Second, I consider whether these differences are due to environmental factors or genetic differences. In order to achieve these goals, I studied fish from a similar genetic background and environment. I compare the gamete quantity of anadromous hatchery derived nontransgenic fish, cultured transgenic fish, and cultured nontransgenic fish, all of which were derived from the Chehalis River system. I measure gamete quantity and quality for female salmon by observing GSI, fecundity, and egg diameter. I measure gamete quantity and quality for male salmon by observing GSI, sperm production, sperm  10 fertility success, and sperm competition ability. Adult phenotype was also examined by measuring body size and noting body coloration and kype.  1.2 MATERIALS AND METHODS Study Site The study site for these experiments was South Campus Animal Care Unit, located at the University of British Columbia. The aquarium facility was inspected by the Department of Fisheries and Oceans' transplant committee and approved for use with transgenic fish.  Sample Populations The sample populations described below are used throughout this thesis and summarized in Table 1.1. i) Nontransgenic Fish Hatchery coho salmon (referred to in this thesis as Nontransgenic (N)) were obtained from the Chehalis River Hatchery, southwestern British Columbia, between January and March in both, 2001 (2001-N) and 2002 (2002-N). Mature, three-year-old fish returning to the river were collected during hatchery sampling or netted from troughs, and transported in a 1000-L tank to the study site, where females and males were separated upon arrival in 400-L holding tanks supplied with fresh water.  o Z  CO  0  >  . .  CD CD  CD • $  <D CO  OB  -§ *  .c — CO CD  LL -C  O  CD  LL  c  c  CO CD o  CO CD o  1  O o  c  CO  —• CD CD o >  O  CO  =  O  Q  O  . .  _ O  g IS  CD  8 CO >  o CD O  8 D _  LT  CO CO  . , CD  P  co 9v o CO  O W co  o £ O  g  a> £  co Z 13 ° g CO  x: O  x: O  LL  LL  Q  Q i_ CD  >  0  ir  "5  re  ter:  S Q  c  x: CO  § Q g CO x: CO 0 •a  a> x: CO O g o  sh  CD i_ "D LL 0 CO CO  0 0 k1  CO CL O C  0  co a CO o 0-  Z  O) CO C CO  lc  o z  0 •a 1 0 LL c  s0  i 0 LL c  s0  i_ l_  1_  < Q5 .  CO CL  o  'c 0 D) CO  c CO  FO  CD  -•—>  O-tg  sz  x: o as —  ter: FO  .c  0  —o  CO  •a O  3 O  12 ii) Transgenic Fish Mature transgenic (T) coho salmon, originating from a single clutch of fertilized eggs (derived from one male transgenic parent crossed with a Chehalis River nontransgenic female), were obtained from the West Vancouver Lab of the Department of Fisheries and Oceans (DFO), British Columbia, between January and March 2002. The transgenic parent was produced by microinjection of a fertilised egg at the 1-cell stage (egg and sperm originating from a Chehalis River coho) with a growth-hormone gene construct. The sequence consisted of a growth-hormone (GH) gene fused to a metalothionien-B (MT) promoter, both obtained from a sockeye salmon. The DNA sequence has been stably incorporated into the coho salmon genome and results in faster growing fish (Devlin et al. 1994a). Milt from the transgenic salmon was used to fertilise Chehalis river coho female eggs to produce a strain of MT-transgenic coho. These fish reached full size and maturity in two years. The specific cross identifications of transgenic fish used in these gamete quality experiments were Floy 77 MTGH1F2 Coho (male) x Chehalis Coho 981229-1 to 9 (females). Transgenic fish were incubated, and reared in various sized tanks throughout their lifecycle, at the West Vancouver Lab of DFO. Transgenic fish were transported in a 1000-L tank to the study site, and females and males were separated into 400-L tanks upon arrival, in the same fashion as nontransgenic fish.  iii) Cultured Fish Milt from Chehalis River males was used to fertilise a sample of eggs obtained from Chehalis river females. These eggs were incubated and reared in the hatchery and  13 aquarium facility at the West Vancouver Lab of DFO. These fish, referred to as cultured (C), took three or four years to reach maturity. Cultured fish were transported to the study site in the same fashion as the transgenic fish.  Data Collection and Experimental Design a) Adult Phenotype Fish were anaesthetised using 0.01% MS-222 : 0.01% sodium bicarbonate. Standard length and live weight were recorded and each fish was marked with a uniquely numbered modified floy tag. Condition factor was calculated as the live weight / standard length x 100. 3  b) Female Gonadal Somatic Index. Fecundity and Egg Diameter Reproductive investment of female fish was analyzed by measuring GSI, fecundity and egg diameter. Ripe females (defined as females with eggs free in the body cavity that could be easily exuded through the ovipositor with minimal pressure) were euthanised by a swift blow to the cranium and bled by slitting the gills. Fish were bled to prevent haemorrhaging and the release of blood into the internal cavity of the fish. Each fish was held vertically as an incision was made starting at the ovipositor and continuing along the belly until the pectoralfinarch was reached. Eggs fell freelyfromthe body cavity and were collected in a metal tray. The total mass of eggs per female was weighed and GSI was calculated. Fecundity was estimated by weighing 100 eggsfromeach female. A sample of unfertilised eggsfromeach female was placed in a small container of water at 4°C and the eggs were allowed to water harden. The diameter of 30 eggs  14 from each sample was measured using standard callipers (± 0.05mm). All samples were measured within two weeks of collection from the female and no signs of egg deterioration were noted.  c) Male Gonadal Somatic Index and Sperm Production Reproductive investment of malefishwas analyzed by measuring GSI and spermatocrit. The milt of a ripe male (free flowing milt when abdomen gently squeezed) was collected in a plastic cup by squeezing the abdomen of anaesthetised males. The male was then killed and the testis and vas deferens were removed and weighed. GSI was calculated by using the combined weight of the testis, ducts and milt as the total gonad weight. A mean spermatocrit per male was determined using three capillary tube readings. Capillary tubes were used to collect milt samples, critosealed, and centrifuged for five minutes in a micro-hematocrit centrifuge. The spermatocrit was read directly as a percent of packed sperm in the milt by using a micro-hematocrit capillary tube reader.  d) Sperm Competition and Fertilization Trials Milt samplesfromfivedifferent pairs of N and T males were used to conduct sperm competition and fertilization trials. The sperm competition trials used 7.5u,L of packed spermfromeach of a pair of N and T males (calculated on the basis of spermatocrit) that were placed together in a dry, plastic lOOOmL beaker containing 100 eggs from a N female. Milt samples and eggs did not come into contact. Fertilization was initiated by pouring 500mL of water into the beaker. The eggs, milt and water were swirled together for five seconds, let stand forfiveminutes, and then rinsed and  15 transferred to an incubator. This procedure was replicated for all five pairs. A similar procedure was used for fertilization trials, except 15uL of packed sperm from each N and T male was used separately (i.e. there was no competition between N and T sperm) to fertilize the eggs. Eggs from all competition and fertilization trials were incubated and the offspring were reared to the pre-ponded stage (yolk sac completely absorbed). Blood samples were obtained from the offspring and mixed with lOOuL of 0.0IN NaOH and stored at -40°C. 3pJL of each blood solution was used to conduct polymerase chain reactions (PCR), similar to that described in Mori and Devlin (1999), and used to determine the percentage of alevin that contained the transgene. Primers used during PCR were MT-1 (5'-CTGATTAAGTTTTGTATAGT-3') and GH-19 (5'GTTAAATTGTATTAAATGGT-3'). Reactions were run for 35 cycles with a 94°C denaturing cycle (1 min), 52°C annealing cycle (1 min), and a 72°C polymerization cycle (1.5 min). PCR products were run on 1% agarose gels containing ethidium bromide and photographed under ultra-violet lighting.  Statistical Analysis Standard statistical procedures were conducted using Sigma Stat (SPSS Inc, Chicago, IL) and Vassar Stats ( computational website. Statistical tests used are indicated with the results, and populations that were significantly different are indicated in data tables with different letters. In all circumstances, the null hypothesis stated that there was no difference between N, T and C fish, and the level of significance was cc=0.05. The null hypothesis that sperm from N and T males are equally capable of competing for eggs would result in  16 the expected percent of transgenic offspring produced to be either 25% or 50%. The former percentage would be expected if the T sire was heterozygous for the transgene, and the latter percentage would be expected if the T sire was homozygous for the transgene. Similarly, for non-competitive fertilization trials, 0%, 50%, and 100% of the offspring are anticipated to be transgenic if milt was from an N, a heterozygous T, or a homozygous T male, respectively. These predictions also assumed no difference in mortality of N and T offspring from the alevin to pre-ponding stage.  1.3 R E S U L T S  a) Adult Phenotype There were several phenotypic differences between N, T and C fish at maturation. C fish weighed significantly less (ANOVA on ranks pO.OOOl, Dunn's Test p<0.05) and were significantly shorter than N and T fish for both males and females (ANOVA pO.OOOl, Bonferroni Test pO.05) (Table 1.2). N and T females were not significantly different in weight, but 2001-N females were significantly longer than 2002-N females. 2001-N males were significantly heavier and longer than 2002-N males, but T males were not significantly differentfrom2001-N males with regard to weight, and 2002-N males with regard to length. The correlation betweenfishweight and length (linear regression) was high for N, T and C fish of both sexes. Transgenic males weighed more on average than nontransgenic males of the same length, and the slopes of the linear regression lines for 2001-N and T fish were significantly different (ANCOVA pO.01, Test for Homogeneity of Regression; p=0.100). The trend in condition factor was similar for both males and females, with N, C, and Tfishdisplaying  ro ro  — —  « I-  I-  CNJ  +i  co  +1 +1 + l  LL  s  q  +1  32 co s 2 co ^  O  .o  Q CO  O  •<-  q  s  q  CM  q  •H  tf  N  O)  W  CM CO  LO -tf CM LO  —  .*-«  to  w  CD  o  00 CO  1^ CO — .—.  CO 0 c  o  c " ra  O  X  LU Ico s  o iri ccd O  "  o  T—  •'-. +1  a>  o> i ^ £J i -  CO Is  CO •n '  £  X  co co  q +i  +1  s  . CO C '*= O CO  -—  CO  co CM I - CM q q cq ~  T-  CM  ro  co CO  s  o  CO LO i  CM ii  x  x  CO s  O)  CM  LO  CM  CO o  co  a>  ||  T-  T-  tf  II  II  II  cr,  tf CM O)  I-  co  i— tf i  Id t  >,  T 00  x r.  CO  "co o5  IS  CM co o)  CO  CM  ffl  CD  °°  o  o  d  d  °  t  r  n  Tt  co  CD  ^  tf"  tf  ico ai  CO o) oo  CM co oo  o  o  N  tf  T-  SS  d  CO CD CD H -  § CO. HE c  §•£ 0  CO _ 0  5 >  iri  II co to  -H  +1 +1 +1 +1 Ol CO rto s ^ ID w in *  0  ® -g  iri  o>  O  in  tf  oo  LO  CD LO  CO  d CO  in  CO'^j +1 +1 -H  to  CO  c CO  c o  •4-*  CO  0 •D  o  •a ^ ~~ Z  S ^ 9E co CO . .  ro ro ro .o 00 tf CO I I - c o CD CM  "D  ro £1 ro co o> co tf tf d d d  s  Q CO  I9-  s  +1  -*—'  d  d  +i  +l  d  d  +1 +1  tf I-  E  x:  LO  O)  LO co  o co CO cz  0  CM  <N  oo cvi d  —^  T—  O  tf  o o  -o  .2  E E. co o co o . x: c co 0 H= 0  LO  Is  tf  CM  CM  d  •H  -H  -H  co  O) CD  C) •tf  O) LO  *r-  CM  d  O) CM  CM CM  in  in  i iO  i CM O  CM  CM h  tf  co LO  co  •H  tf CM  s  u  0 = 5 o  • c  o  * " co  a o £  ~ o ~ < ..  CO "Cl ^  ^ <= = - .2 8  111 ( B o . ? O CO  I—  CO  0 CO E 0  CO — O O CM  CM O O CM  0  m  O  O  18 the lowest, intermediate and the highest condition factors respectively (ANOVA on ranks; p<0.0001, Dunn's Test; p<0.05). These data support the visual observations that both T and C fish have a rounder body shape then N fish. Body coloration of both male and female N, T and C fish was also different. Nontransgenic fish possessed a reddish-brown coloration along their side below the lateral line, whereas, the cultured and transgenic fish did not. However, the cultured and transgenic fish did darken at maturation from their silver coloration and possessed a wide variation of shadesfromnear silver to dark brown. Male nontransgenic fish possessed a well developed kype and clearly visible teeth. Neither the transgenic, nor the cultured males developed a kype, and the teeth could not be easily seen. Transgenic fish developed cartilaginous growth around the mouth that obscured the teeth.  b) Female Gonadal Somatic Index, Fecundity and Egg Diameter Differences in GSI, fecundity and egg size were detected between N, T, and C females (Table 1.3). The average GSI of 2002-N females was significantly less than those of T and C females (ANOVA pO.001, Bonferroni t-test p<0.05). However, the average GSI of 2001-N, T and C females were not significantly different. The average weight per egg of 2001-N females was significantly greater than egg weights of T and C females but was not different from 2002-N females (ANOVA on ranks p<0.0001, Dunn's Test p<0.05). There was no significant difference in fecundity between 2001-N and 2002-N females, but T fish contained significantly more eggs whereas, C fish contained significantly fewer eggs than N females (ANOVA on ranks pO.OOOl, Dunn's Test p<0.05). These data indicate that transgenic females, on average, have more, smaller  CD  I  Q  o o CO LO  as  a)  El El  <M O  d d +1 +1 +1 +1  O CO -r- CM CO LO LO LO  CD CO LU  CM ^ CM LO CO CM CM CO  c o  CO O O O  CD U_  CO CM i 00 N I D  T -  CO  d o d o  +i +i +i +i  LO O CM CM CM CO T-  1  CO  CO ^ \ LU  to  q S P o d ° d 0  +i +i + +i 1  Q_ CD  8 CM d d  co CO  +i „  •^f O  £  LO  d  d  O O CO  E i_  <». aj  CO +1  CD LO CJ) r-~ C D CO +i +i +1  CD co co co CM CM CM CO  CO  CO CM O O  o d d +i +i +i CO o o  co  CO  CD  o o o o d d d d  +i +i +i +i  °2 CM CM ^ d d d  o o o o d d  CO O ) CM i -  CO CO N CM i r -  i -  N T -  CD CM  co _CD CO  E cu u_  z z •A  CM O O O O , CM CM I—  z  I  z  1 CM O O O O , CM CM i — T-  CO CO  20 eggs, but the same total weight of eggs as a nontransgenic female Fecundity was correlated with total weight for 2001-N, 2002-N, T and C females (Figure 1.1). However, the slopes of regression lines relating fecundity to body mass did not differ significantly (ANCOVA p=5.00, Test for Homogeneity of Regression p=0.09). The mean egg diameter of 2001-N fish was significantly larger than 2002-N, T and C fish (ANOVA on ranks pO.OOOl, Dunn's Test pO.05) (Table 1.3). 2002-N eggs had a significantly larger diameter than T eggs, but there was no difference between T and C egg diameter. The diameter of a female's eggs increased significantly withfishweight in 2001-N (ANOVA, pO.001) and 2002-N (pO.001) females but was not correlated with weight in T (p=0.9933, r =0.000004), and C (p=0.0525, r =0.1092) females (Figure 1.2). These data 2  2  indicate that the egg diameter in C fish is more variable than 2002-N, T and 2001-N fish (coefficient of variation = 10.3%, 6.33%, 5.36% and 3.53%, respectively).  c) Male Gonadal Somatic Index and Milt Production There were no significant differences in GSI or spermatocrit of 2001-N, 2002-N or T males used for these experiments (ANOVA pO.OOOl, Bonferroni t-test p<0.05) (Table 1.3). However, C males had a significantly higher GSI and spermatocrit than either 2001-N, 2002-N or T males. Gonad weight increased significantly with fish weight for N, T and C males (Figure 1.3). As well, there was a significant difference between the slopes of the linear regression lines of 2001-N and C fish (ANCOVA pO.01, homogeneity of regression p=0.3035). It was noted in two transgenic fish and one culturedfish,that only one testis was fully developed (identified on figure by a hollowed shape).  o o o  CO <D CO  CO  E  CM o o o o CM CM U U T-  •CP -*— o  c<D 0)c  c  CO  LO  CD CO O •—  g i  "a  o o o  § -o  CM  o o  CM  o o  z  I  o o O)  CM  CD 1— =3  CD  § CO  fl) >  CO  o o o  (N  CD  5 CO  c^  O c o o c =3  T3  CD -*—* CO  o  C  CD CO CO CD 3  CO CO CO  o o o  CO  >  T3  ~—  C  c O  1I CD  . . i_  T-  CD  2 co CD c  o o o  CD  o o o  m  o o o  o o o  CO  o o o  CN  o o o  8|BUJ9j / s 6 6 g jo j a q i u n N  o o o  sz a  co CD  z  Lf)  i—  O  o o  "O  OJ  1 1— 1 CO d 1 1 li 1  1  o o o  _  o o  CM  I I  k.  •  LU rill  •  o o o  O)  o o o  CM  CM , O O ' o  o  Ui •—  S 8 Si •  «  i  o  I LO  5  CO  CC CO CO  ^  ^—s  E E  h-  •  "CD  =5  cn •  CO  s66g  0 =3  CO  >  CM  '  1  c  o CO CO CD  i_  CO CD  •  c; <: o  sz CO CO  v_ CO  _Q o  V—  CD "O  i—  CO CC CD " O  c  o i_ "O o  CM  I  CO  c  CD  II  lO  i_ CO  T—  o  LL  Lf)  Lf)  Lf)  •  CD  (LULU)  O  •  3  CD  H  LUBj  §  0  C  o c 13  M—  C C  o  T3  o  CD  O  CO  '•4—'  O)  CM CM  •  CO  CO CJ)  T-  o o  JC  c  m  Ui  CD  o  D)  <u o  C T3  CD  CO  O  -4—*  iear  II  co o  igh'  I \  o  CM  o  CD  UP  g  0.51  •  CC  femal e sta  CM  c  z~ £  LO  JO  J3}8LUB!a  c  CO  c o  CL  23  O  §  cn  - sz Z **  LO  CM  ^  O  "O  •  CD  CO  O  0)  •-  CD  w  O  7  o o o  "  s  ' Ei  o o  CM  0  D_  o « > t:  CD  CD  . N T3  O) <"  "5 ® 3 SO. *1  3  cw E ^  o o o  o •a >• cCO cO  CO  U)  "3  co co CM  T -  II  II  c i-  o o o  c  CM  CN I s  O O O O II CM CM  u "ft  CO  C/>  • mum  LL  CO CO  O  "CD >  c c a> o)  o  U)  —  O)  fills  o o o  Z if  c  LO CM  CD  O  _l CO  CO  C  <D ® o cn _=i S«  •i J » ®£ ^  CO  I-  1  CD  -a  :| -1  &• o o CM  .Q  co E o ^ O  r"  o  "co  cn E  W  co o c E *  Z •  ^ CO " — ' _CD  Big  c  U  (0 T3 CD CO  o  LO T—  ©  o T—  ©  LO  (6) iMBiaM peuoo  24 d) Sperm Competition and Fertilization Trials In each replicate of all sperm competition and fertilization trials, the proportion of transgenic offspring fell within the proportion predicted if both N and T sperm contributed equally (Table 1.4, and Table 1.5). None of the offspring tested positive for the transgene when milt was obtained from a nontransgenic male (Table 1.5). In trials 1 to 4, the percent of transgenic offspring fell within the confidence interval expected for a male heterozygous for the transgene. In trial 5 100% of offspring were transgenic, indicating that the parental male was homozygous for the transgene.  1.4 DISCUSSION Exploring differences in fitness related traits facilitates evaluation of the potential risks that transgenic fish may pose to the natural salmon stocks. There are significant differences in adult phenotype, gamete quantity and gamete quality between N, T and C fish. These qualities are important factors in determining the fitness of transgenic, nontransgenic and cultured fish, a) Adult Phenotype Adult phenotype influences reproductive success in nature. For male salmonids, body size and other secondary sexual characteristics (i.e. kype and coloration) can determine their proximity to the ovipositing female which, in turn, is positively correlated with fertilization success (Fleming and Gross, 1994, Gross, 1991, Chebanov 1990). It is anticipated that such phenotypic characteristics could enhance or diminish the reproductive ability of growth-enhanced transgenic salmon. In this study, the average weight of N and T males and females were similar to the average weight of 3 year old  25 Table 1.4: Sperm Competition Trials: Trials, replicates, stock and pair number, number of surviving alevin, and percent of offspring positive for the transgene accompanied by the 95% confidence interval for the proportion of transgenic offspring.  Trial  Replicate  Stock and Surviving Pair Number Alevin  % Transgenie  95% Conf. Inter. Lower Upper  1  1 2 3  T-1 vs. N-1  78 83 91  28.21 21.69 31.87  18.88 13.69 22.71  39.7 32.35 42.58  2  1 2  T-2 vs. N-2  90 78  15.56 20.51  9.07 12.53  25.06 31.46  3  1 2  T-3 vs. N-3  82 71  20.73 23.94  12.87 14.95  31.39 35.8  4  1 2  T-4 vs. N-4  18 26  22.22 26.92  7.37 12.35  48.08 48.05  5  1 2  T-5 vs. N-5  72 72  40.28 47.22  29.09 35.48  52.51 59.27  "S I  o o  O  O  o o  o o  o o  o o  8 °>  CM 00  LU Qi  c « c cn £ c >>  c Q.  „  c «> 8 o 55 .a  o o  o o  LO 00 iN 00 O)  a> oo co -tf  CO 00 CJ)  T -  r~-  oo  CO  "I  Q. C TI <°  «S aj o re c cn .2  1 1£  10  CO  re E o o ^  I  CO 're  CL  8-  •4-  O  «>  £  II L_  CD  = T  o  -9 °-  c t o  T -  C  O  -»—»  CD tf 05 in  tf  CM CO  tf  CM CM  •tf  ci  CO  tf CO  CM CO CO c o 00 LO • t f  CO CO  CO CM  r-  05  -,—  •tf  CM  00 a> CO CO  CO CO CO  CO CO 00 00 CM 00 •tf i n  tf> CD C CO  CO  c o  3 C 05 C . CD <D • ^ "D S 're  ?5  ^  ^ Si  CD  CD  o re re  •tf  LO i n  tf  o o o o  Q- g O T3 O !C  re re o !5 o» « O w  05  CD Q  £ E >> £  '-15  MS  r e d ) re . a c  re x: c  CD U-  00 CD  CM  co • I-  <  re  i  o o _i_  co  I-  're o.  T3  > re  CM CO  T-  CM i - CM  O 05  a. CL CO LO  CO T 00 CM  S8  £ S g  3 CO  00 CM CM CD c n OS  Q-  •o5 OH <D"D S CD to o N "5. 5 CD 2 CD >< l03 - D- Q) S ™  N  > >  a)  CD Q. 05" ">  CM  co  - C M  o  CO CO h -  27 mature fish returning to the Chehalis River (pers. com Larry Kahl, actual data based on 1995 Chehalis River Hatchery data), even though T fish mature at two years and are held in culture conditions. However, culturedfishhad a much lower average weight. As the culturedfishwere three or even four years old, these data indicate that the culture environment used in these studies (i.e. tanks) has a large negative effect on growth rate that can not be overcome even with the additional year of growth. However, growth rate is enhanced by the GH transgene under tank cultured conditions. Indeed, the presence of the transgene enables the transgenic fish to overcome the negative effects of the culture environment (Devlin, 1994a). Previous studies have indicated that culture regimes result in phenotypic differences in salmon when compared to wild fish. For Atlantic salmon, Fleming et al (1994) found that cultured male parr had smaller heads andfins,and narrower caudal peduncles in comparison to wild fish. However, short term, temporary culture regimes, such as net pens, before smolts are released into the natural environment, have been reported to have significant positive effects on rate and size of returning adults (Linley, 2001). These data have several possible implications for the fitness of T and Cfishif they are released into the natural environment, and successfully return to a river to spawn. It is known that the larger males become the alpha male in the spawning hierarchy for access to females, and these larger males obtain primary access to the spawning female (Gross, 1991). Therefore, male fish cultured in a tank environment may have reduced access to females during spawning as a result of a reduction in size. However, alternate mating tactics, such as 'sneaking', may be utilized which would reduce the size  28 disadvantage because smaller fish can dart in and out during the spawning event more effectively than a larger male (discussed further in Chapter 2). The transgenic fish used in this study, however, were not significantly larger than hatchery fish. Therefore, transgenic fish may not have a mating advantage due to size. Furthermore, cultured and transgenic fish lack the characteristic red coloration and kype of wild salmon which are important in obtaining mates and during fighting bouts (Gross, 1984, Jaervi, 1990). Due to these phenotypic differences, transgenic and cultured males could be at a disadvantage in obtaining mates, and this would ultimately result in reduced fitness, particularly if a fighting mating tactic is used to obtain access to females (explored further in Chapter 3). There are also interesting biological implications that these data present. Transgenic fish may not ultimately achieve a greater size than hatchery fish because the intentional, or unintentional, selective breeding at the hatchery over time may have resulted in the maximization of fish size. It has been previously demonstrated in trout that the transgene could increase the size of wild but not domesticated fish (Devlin et al. 2001). Alternatively, it is possible that maturation is size dependent rather than age dependent which could account for the various different age classes at maturity for salmonids. The body shape of salmon is both genetically and environmentally determined (Fleming et al. 1994, Taylor and McPhail, 1985), and this study provides an indication as to the extent to which each factor can contribute to phenotypic differences. Condition factors of nontransgenic, transgenic and cultured males arid females were significantly different. Nontransgenic fish have a sleek body shape, transgenic fish are more rounded, and C fish are intermediate. Culture in an artificial rearing environment may cause  29 divergence from natural shape, and it is possible that the transgenic condition intensifies this divergence. The difference between the mean condition factor of T and N fish is approximately three times the difference between C and N fish (for both males and females), indicating that genetics can potentially play a greater role than environment in determining phenotypic differences.  b) Female Gonadal Somatic Index, Fecundity and Egg Diameter The fecundity and quality of eggs have important implications for female fitness. A positive correlation exists between fecundity and fish weight (Foerster and Pritchard, 1941), and between egg size and early survival of offspring (Heath et al. 1999). Nevertheless, fitness advantages may also be obtained by investing in egg production (fecundity) over ova size (Einum and Fleming, 2000). The average weight of nontransgenic and transgenic fish used for the fecundity measurements did not differ significantly, nor did the average weight of all eggs collected. However, the culture environment did reduce the total egg mass and fecundity in cultured nontransgenic salmon. Despite this, GSI was not different between 2001-N, T or C females, indicating that total reproductive investment is closely coupled with body mass over a wide range of growth conditions. However, the significant difference between fecundity and egg size of N and T fish may have important fitness implications. Transgenic fish had more, smaller eggs than nontransgenic females of the same body weight. Apparently, this is a direct result of the transgene and may occur because of either the increased growth rate of the transgenic fish, the reduced duration of life, or a combination of the two. Grachev (1971) found that  30 the number of oocytes in pink and sockeye salmon ovaries are at a maximum during downstream migration and gradually decreases with time spent at sea, and rate of maturation. It is possible that transgenic fish have more oocytes then nontransgenic salmon because they spent less time in salt water and mature at an earlier age. McPhail and Lindsey (1970) estimated the average egg diameter of coho salmon to be between 4.5-6.0mm for most stocks, and all of the fish tested in these experiments fell within this range. Egg diameter was significantly reduced in transgenic fish. If reduced egg diameter is due to the action of the transgene, it is possible that increased fecundity in transgenic fish may be coupled with reduced egg size which may result in a reduction in offspring survival. Alternatively, transgenic females may have reduced offspring survival but this may result in increase fitness due to the maximization of fecundity.  c) Male Gonadal Somatic Index and Milt Production Sperm quality may be affected byfishsize. GSI and spermatocrit were significantly higher in C males than N or T males. However, C males were also significantly smaller than N and T males. Perhaps C fish allocate more energy into producing larger and more sperm concentrated gonads due to their smaller size. There are two known alternative life history strategies for male coho salmon attempting to gain access to females: fighting (size usually determines dominance) and sneaking (male darts in to release milt during the spawning of a female with an alpha male) (Healey and Prince, 1998). Assuming smaller males adopt a sneaker mating tactic (since a small male can not successfully compete for a female by fighting), the production of milt with a higher concentration of sperm could increase the probability of egg fertilization during a  31 sneak mating with a female. Vladic and Jarvi (2001) found that, for male Atlantic salmon, mature parr invested more in gonads, had a greater number of motile spermatozoa, and had a greater sperm ATP content than anadromous males with a larger body size (50.14g versus 7695g body weight, respectively). Vladic and Jarvi (2001) used their observations to suggest that behaviourally subordinate males may have physiologically superior spermatozoa. Since T males possess similar sperm volumes and spermatocrits as N salmon of the same size, these data do not indicate a fitness advantage for transgenic males.  d) Sperm Competition and Fertilization Trials N and T milt were equivalent in their ability to successfully fertilize eggs during sperm competition. There is no indication, based on these data, that the culture environment or the transgene have any affect on ferilitiy of the milt of transgenic salmon. Since transgenic males mature one year earlier than nontransgenic males of a similar size, these data also indicate that age is not a factor in determining the ability of sperm to fertilize eggs. Hoysak and Liley (2001) also found that there is no difference in the fertility of sperm of different age classes of male sockeye salmon.  32 2 COURTSHIP AND SPAWNING BEHAVIOUR ACCOMPANIED WITH GENE TRANSMISSION  2.1 INTRODUCTION The mating system has a direct impact on an individual's reproductive success, and therefore, its ability to contribute to succeeding generations. Consequently, any effect of the growth hormone gene over-expression on mating ability may have important implications regarding the reproductive success and fitness of transgenic salmonids. All salmonids display similar mating tactics that include courtship and spawning behaviours (Groot and Margolis 1991). Typically, females spawn in several nests that they dig in a single redd (Healey, 1991). Digging behaviour can be identified as the lateralflexingof the females body while constructing a nest. Males court females by continuously maintaining a position along side the female (termed attending) and occasionally quivering. Quivering is an undulation of the body of the male beside the female. The female extends her anal fin down into the nest depression (termed probing), and during spawning, both the male and female simultaneously open their mouths wide (termed gaping) while releasing their gametes. Following this, the female covers the nest. Covering is identified as light, forward lateralflexesby the female that result in the displacement of rocks over the nest (Liley et al. 1986). Typically, quivering and probing increase as the spawning event nears (Tautz and Groot 1975, Berejikian et al. 1997). Male quivering frequencies have been correlated with female digging and probing frequencies (Berejikian et al. 2000), however, female courtship behaviour was not effected by the size of the courting male.  33 It has been previously demonstrated that cultured salmon achieve less than one third the breeding success of wild fish because the former are competitively inferior and less aggressive (Fleming et al. 2000, Fleming and Gross 1993). Nevertheless, it has been suggested that transgenic fish, despite being culture raised, may have a reproductive advantage due to their increased size (Muir and Howard, 1999) because it is known that larger males adopt superior positions in the mating hierarchy for females (Healey and Prince, 1998). This suggestion has not been tested, and nothing is known about the effects of genetic modification on courtship and spawning behaviour of salmonids. Empirical data on courtship and spawning interactions of transgenic salmon with wild fish would assist the determination of whether transgenic fish, if accidentally released into the wild, would have a fitness advantage over wild fish. This chapter has three main goals. My first objective is to explore whether transgenic salmon are capable of courting, spawning and producing viable offspring in matings with salmon found in the natural environment. My second objective is to obtain an indication of the effects that growth hormone transgenesis has on salmonid courtship and spawning behaviour. My final objective is to determine the extent to which any behavioural differences observed in transgenic fish are due to the transgene or to the research culture environment in which they are necessarily raised.  2.2 MATERIALS AND METHODS Laboratory Spawning Channels All behaviour trials were conducted in artificial stream channels located at the South Campus Animal Care Unit of the University of British Columbia. Three fiberglass  34 tanks (468cm long x 152cm wide; 0% gradient) were separated into two stream channels each (300cm long x 74cm wide) with plastic screens (Figure 2.1). The sides of the tanks were made of clear Plexiglas; providing a window to observe fish behaviour. A constant supply of dechlorinated fresh water was delivered to the tanks and circulated at an average rate of 0.14m/s by several water pumps. Gravel (1.5cm - 6.0cm) was placed in the stream channels, providing a total available spawning area of 2.22m per stream channel, with a gravel depth of approximately 7cm. The typical spawning area utilized by female coho salmon is 2.5m (Burner 1951), and this redd size accommodates 2  multiple nests. Nevertheless, the spawning area provided here seemed adequate since data compiled in this chapter were based on thefirstspawning event. The water depth in each tank was maintained at 38.8cm. A 12-hour lightdark regime was maintained by overheadfluorescentlighting and supplemented by natural light. Although the tank dividers prevented the passage offishand eggs from each side of the stream channel, they did not provide a barrier for pheromones or other chemical emissions that may have been released by the fish. It was noted by Scott and Liley (1994) that these excretions can affect hormone levels of courting males and possibly of other nearby males. Thus, fish examined on opposite sides of a single tank may not have been behaving independently. Nevertheless, the simultaneous occurrence of spawning in each side of the stream channel was not significantly different than the occurrence of  Figure 2.1: Laboratory Spawning Channel: Solid square arrows represent water pumps and open arrows indicate the direction of water flow. Video camera and swim channel measurements are illustrated.  36 spawning on only one side of the stream channel for any of the three tanks (Fisher Exact, p=0.07, p-0.2, and p=l; Tanks 2 vs.3, 1 vs. 2, and 1 vs. 3, respectively).  Courtship and Spawning Behaviour Nontransgenic (2001-N and 2002-N), transgenic (T), and cultured (C) coho salmon were used in these experiments (refer to Chaper 1.2 - Sample Populations) (note: all Tfishused in this chapter were heterozygous for the transgene). Pairs of ripe females and males were placed into the spawning channels and observed by both periodic direct observation and by continuous video surveillance for six hours. Table 2.1 indicates the number of pairs, and their respective genotypes, for which data were obtained. Direct visual observations of courtship and spawning behaviour were recorded on a check sheet for five minutes every hour during six consecutive hours, for one observation day. These five minute observation periods were combined to provide a total of 30 minutes of direct visual observations for analysis. In addition, the continuous video record was analyzed in a series of five minute periods at 30, 20 and 10 minutes prior to spawning, during spawning, and 10 and 20 minutes after spawning (i.e. -30, -20, -10, 0, 10 and 20, respectively). I recorded digging, probing, and covering behaviours by females, and quivering by males.  Transmission of the Transgene Eggs from spawning events were collected from the gravel and incubated (7°C) until the pre-ponded, alevin stage. The offspring were then tested for the presence of the  37  CO o ' ' cn  3  CD CL >*  O c  a> co  c  CO CJ! CO  c co  ^- o o) in, co LO LO  o c CO CT CO  c co c o  ^ o to ^ i n c\i  " d  c  c  CO CO  CD CO (0  co £=  2 +-»  c  z  i-  c o  co 0. I— D o  CD  CO  CD CL  2?  o c CD  (5 ® CO  co •c 3 o O  E CD LL  CD  o  38 transgene using polymerase chain reaction (PCR) techniques (refer to Chapter 1.2 Sperm Fertilization and Competition Trials).  Statistical Analysis Courtship and Spawning Behaviour Three types of statistical tests were used to analyze the behavioural observations. Kruskal-Wallis (ANOVA on ranks) analyses were used to analyze all behaviour data, and if the null hypothesis of no difference between treatments was rejected, a Dunn's Test was used to determine which treatments were different. A Fisher's Exact test was used to determine if the genotypes of the parents were related to differences in the proportions of the fish that spawned. Sigma Stat (SPSS Inc, Chicago, IL) was used to perform all statistical analyzes.  Transmission of the Transgene The gene transmission data were analyzed with both a Binomial Test and an Exact Binomial Test to determine if significant differences existed between the actual and predicted proportions of offspring containing the transgene.  2.3 RESULTS Courtship and Spawning Behaviour Treating nontransgenic parents as a control group, there was a significant reduction in spawning frequency whenever the female was either transgenic or cultured (Table 2.1; NxN vs. TxN (p<0.002), TxT (p<0.02) and CxC (p<0.03).  39 There was considerable variability in nesting activity from one observation period to another. To some extent this reflected the impact of the occurrence of spawning. When the data for each observation period were combined to give a 30 minute total for each pair, T females conducted significantly less digging and covering than N females when paired with an N male (Table 2.2). There were no significant differences in male quivering for any of the pairs. The video record allowed a more detailed analysis of courtship activity immediately before and after each spawning event. There were no significant differences among genotypes in the frequencies of digging, probing, or covering from 30 minutes before, to 20 minutes following spawning (Figures 2.2, 2.3, and 2.4). No digging occurred for 20 minutes following the spawning event, and covering only occurred after the spawning event. Although only one T x T and one T x N pair spawned, T females appeared to perform less digging but probed and covered as actively as N females. There were also no differences in the amount of male quivering prior to the spawning event, but C males quivered significantly more than N males after the spawning event (Figure 2.5).  BJ  o  X  o  CT)  cvi  +1CM  CD  CM  H  co"  o  X  z O  >- -o  0) CO  CD O  l \ t Z 3 .2 .. o .TS  O O  CD CD > CD  O o 5  'i | 8  CM  +1  CM  00  « LO  X  CO  Q  +1  CO  CD ^ o-g § CO o o- c o3 c>  CD  E  LO  CO  CD  Li-  CC  0_  c o L_ 3  +1oo CO  cd  O  *- o ® 0 _  t  SJ €  1 > s n o ? Ol (D ® -  7 O)  O  •4—  O  CD Q. >>  O C CD  CM  +1 °>  CD CO  * ^ "l?  co  CD  +1 cn  i2 o > O  g  > 3 £ CO c w -E ~  LO  o E co CD o "55 Q.  |  CO CD  °  s  i l l >  ^ Q  -Q  O  c  co  (0 CD  t: e  3 CD  <N 8  i  <M CD §  3  CD > O  o >  O  si  CD CD g> N Q CO  CO CD  OQ Q. !c CO  •c  • 3  o  O  ~cb c •  0 CO  :  CD Q.  E E CD CO LL CO  CJ)  C 'c  o  o  OJ  CM  o co  5  CO CL CO  co —  "D II  G  if) II c  II  c  co  in  c  c  II  CD  II  o t  •  CD  2 S  c o |  CD  Z Z D  iII c  H  X  X  X  Z D  IB  II  O  x Z I  U  x O •  E T3 o S CD CM CO -a .9 T3 i_ C T 3 CO  CO >  T3  C CO  co r-  c  CO  E «  0  co Q.  +•»  £  co  O  a  - .c O) o C CO  0 s c »  ill  .i  I—  i — — i  I w  CO  D)  T3 C CO CD  o -  I °  C L CD CO p. C O 5" c o •c c 3 CD ~o C O  -  CD  EG  c co o~  CO  CO C C L CM CO O  o CO  b o CO E  * - co o o  CD LL 3 CM  C  y1  CM E o ~- CM i CD O  i_  CD  sajnujiAi 9 / s6|Q e i e u i e j j o jeqiunisj  •o c co i co »_»  o  CM  c c  o 03Q. CO  O CN  CO  CO  • a ^  in  -r-  co  co  1  1  TD  o CJ  g  O  c:  o x  C 0)"Si£ O O "5 oCD c >i±:  TD  0 3  £  "D  CM CO  roT 3 TJ  C  CO roco C  >  ^ T3  c  GO  -H  s_ en  E «  - x: c co C c a> CD E I o E in C L 03 Q. CO CO 0 3 2? C O 0 -Q i3Z 03C O "o cn CD  +-  CD  •4—I cn o ZJ C  ro Et cz  m  Cn  c o 0_CD -  ro E CD  LL CO  C13  co o~ Q . CM <" " D O C ' CO  oi r o CL  -  CO  °  £  ZJ C  d  T~ 1  £ d c\i o * CD Z3  cn  sainuiiAi g / saqojd 3|eiu8j io jaqiunN  c 9 co  CO  o c CAJ 'c o co  o CM  CL 03  £ CO o ZS  C ^ 0 co  c c5 E o o  1 II  e  CD  "E o 03  >  C  to  o  +i  <D  10  C  co  Q>  o II  II  IO 11  c  c  Ti c  CD  II  C  z X  H X  z  z  • •  z X i B  LO  in cn  o  c  1X  o X  o X  1-  z  o  CM  CO  I  I  I  o  co  co  S  ro >4_ CL  O  CO  CL  co  0 5  C  CD  1H  CD >  o i- C O g ro _CD  CO  E CD  CM  'ro  "i—  • • • CO  CO CO  c  c ro  o  o  CD  3 ro =g  CO  E  CM  ^ ro 5 CJ  C\i CM  CM  sajnui|/\| 9 / SJGAOQ aieuiej jo jaqwriN  CO CD  o  .  io  o o l - CVJ a c o ro co CD L.  44  O  CD  X  co c CO  CO  E  .2 £ — g CO -o ffl CD O  o  o  t  CM  n=10  CD  in M c  IT c  CD  II C  Ti c  in II c  Z  I-  Z  X z  X  |—  X  X  o X  o X  z  1-  1-  z  o  • • a  co  C  CD O  c  CO  -a 2 ^ 1 6 c co n si c  - S° 3  f . o ^ C >> CO — g> CD e.  • •  c CJ  o . i  CO CD  +-*  CO  *  LO  CD  CD  c  CO  CM  00 1  ••— CO  3  CO  CD X !  2>  3  o  CO  E  3  C  §  CO CD  >  CM C  B  C  g>S o  g  ™  1 i ? 2 .> Q. ' O o= co CD C M  o CO  sajnujiAi Q /  sjeAm©  ©lew  J O  jaquiriN  CO  CD  3  CO  . £ e o  CD  g  co  C i lis ° C™^  CD  . Tc— cC J ) •Q  3  <°  J-  CM  co E  45  Transmission of the Transgene I recovered sufficient numbers of offspring to test for the presence of the transgene from three out of seven pairs in which at least one parent was transgenic (Table 2.3). In two of the three crosses, the one transgenic parent was heterozygous for the transgene, and the number of offspring testing positive for the transgene did not differ significantly from the expected 50%. In the cross in which both parents were heterozygous for the transgene, the number of offspring testing positive for the transgene did not differ significantly from the expected 75%. The transgene was not detected in any offspring that were a product of NxN parents.  2.4 DISCUSSION  These data indicate that, although fewer transgenic and culturedfishspawned, females display the full range of courtship behaviours and are fully capable of participating in a spawning event. Out of 32 N females paired with males, 21 spawned regardless of the male genotype, whereas only 2 of 13 T females spawned and 5 of 12 C females spawned. T females did display overall less digging and covering behaviour than N females when spawning occurrence was not taken into account. This difference may reflect the fact that significantly fewer T females participated in spawning. Indeed, when only spawning females were compared, there was significant difference between courtship behaviour of T, N and C females, although, T females showed very low levels of digging. It is possible that differences in female  46  Table 2.3: Transmission of the Transgene - The percent of offspring that tested positive for the transgene from three pairs with at least one transgenic parent, the number of offspring tested, and p values from the binomial test. % Transgenic  % Expected  Sample Size (N)  p-value  Pair 1 - N x T  51.4  50  70  0.905  Pair 2 - N x T  49.1  50  106  0.92  Pair 3 - T x T  55.6  75  18  0.981  47 behaviour may become more significant if fish are placed in competition for spawning grounds. Berejikian et al. (1997) found that captively reared coho salmon display the full range of behaviours shown by wild coho but wild salmon have a competitive advantage due to their ability to establish nesting territories earlier and construct more nests. It is also possible that differences in female behaviour may be more significant when wild fish are used to compare behaviours. Note that these experiments used N females that were reared in a culture environment prior to their life in the ocean. Fleming and Gross (1993) found that hatchery females delayed the onset of breeding. In addition, Jonsson (1997) demonstrated that cultured Atlantic salmon females were less active in nest construction than wild fish. Similar to female behaviour data, the male behaviour data indicates that although transgenic males display less courtship behaviour and reduced spawning occurs, male transgenic fish are fully capable of displaying the full range of behaviours and spawning successfully. Spawning occurrence was reduced equally when one of the male parents was transgenic or cultured. Fleming and Gross (1993) found that hatchery coho males partook in fewer spawnings and were less aggressive which consequently resulted in less access to ovipositing females. DatafromBerejikian et al. (1997) are similar to those presented here; culture-reared coho displayed all courtship behaviours and were capable of spawning in a natural setting. The gene transmission data also indicates that the transgene is successfully passed on to viable offspring. These studies suggest that transgenic coho salmon are capable of courting, spawning and transmitting the transgene to viable offspring in a simulated natural environment. However, there is an indication that spawning ability is impeded by culture  48 and further impeded by the presence of the transgene. Within the time constraints of the experimental design, spawning occurred less frequently in C and T fish than among N fish. Nevertheless, if T and C fish take longer to achieve the motivational condition necessary for spawning to occur, it is possible that a longer period of observation would result in increased spawning in C and T females. These findings contradict the claim that transgenic fish will have a spawning advantage (Muir and Howard, 1999). However, it is important to remember that transgenic fish that are released accidentally into the natural environment at an earlier age will be subjected to different environmental conditions. A transgene-environment interaction could result in different courtship and spawning behaviour than those observed here. Also, there are other aspects of reproduction that are key factors in determining whether fish will be successful spawners. Such factors include competition between males and females during mating, as well as, the ability to 'sneak' a mating (Healey and Prince, 1998).  49 3 M A L E COMPETITIVE BEHAVIOUR 3.1 INTRODUCTION The effect that hatchery and aquaculture practices have on male breeding success has become an increasingly important issue, primarily due to the possibility of negative impacts that introducedfishmay have on the sustainability of wild salmon stocks. Fleming and Gross (1993) found that hatchery coho males were less aggressive than wild males, had less access to spawning females, and a reduced breeding success. Likewise, Berejikian et al. (1997) concluded that wild coho were superior to captively reared coho because the former obtained a large majority of the spawning events. Although cultured salmonid males appear to be less successful spawners than wild males, it is not known whether transgenic males (that have altered phenotypes and are reared artificially) can be successful spawners in competition. Transgenicfishthat are purposefully or accidentally released into the natural environment may return to the spawning ground, encounter wild males, and compete with these wild males for access to females. Breeding success by male salmonids is determined by social status and the particular mating tactic that is utilized (Mjolnerod et al, 1998). Social status and mating tactics are associated with the age and size of the fish atfirstreproduction. Male coho have been observed to establish a hierarchy based on size: alpha males (large, three-year olds) maintain a primary position with a courted female, satellite males (small three-year olds) maintain secondary positions behind the alpha male, and 'jack' males (small twoyear olds) occupy peripheral positions (Healey and Prince, 1998). Gross (1985) proposed that alpha males compete for access to spawning females by displaying combative behaviours such as chasing and biting, while 'jack' males obtain matings by sneaking  )  V  50 (darting into the nest depression to release milt during spawning between more dominant males and the female). Gross (1991) suggested that these two different behavioural tactics are conditional and may respond to environmental changes. Competition between males is common in many species of salmonids (e.g. sockeye salmon (Hanson and Smith, 1967), pink salmon (Keenleyside and Dupuis, 1988), and Atlantic salmon (Fleming, 1998)). Secondary sexual characteristics also vary between male types; alpha males develop bright red coloration and a large kype, while jack males retain cryptic coloration. The goal of this chapter is to compare the competitive reproductive success of transgenic, cultured and nontransgenic males, by placing the fish in competition for access to sexually active females. This study analyzes male mating success, courtship and aggressive behaviour in the presence of a male competitor.  3.2 MATERIALS AND METHODS Nontransgenic (N), transgenic (T), and cultured (C) males were paired (i.e. N & T, N & C and T & C)and placed in stream channels with a ripe nontransgenic female for one observation day (refer to Chapter 1.2, Sample Populations, Table 1.1, and Chapter 2.2, Laboratory Spawning Channels, Figure 2.1). N and T males were paired by weight, but cultured males were significantly smaller than N and T males (Table 3.1)  co o d  Ol  +1 d  co cn d  +l  t-l  CO CO  CO d  +l  Ol  cn c 'c  CO d  z:  LO d  +l  03  CL  CO  CO o c o it;  CO d  0  E o O c oo 0  +l  I—I  CL  CM  h-  >  00  OS  LO d  **— o  +1  CO  LO  CM  "cc 0_  W  CO  .!= — CO — ' 0L Q CD  E  +1  sz cn  ZJ  c  CO 0  T3  — o o  O  cn c 03 CL  CO  52 Cultured males never attain the size of N and T males (refer to Chapter 1.3, Table 1.2). Behavioural data were obtained from continuous video records of each trial (refer to Chapter 2.2). I recorded courtship behaviour by males (quivering), and competitive behaviour (chasing and biting). Quivering is an undulation of the body of the male usually in front of or beside the female but may also be directed towards another male as a threat display. A chase is the active pursuit by one male towards another that is retreating. A bite occurs when one male attempts to, or successfully, bites another male. Data was only recorded from trials in which spawning occurred. Quivering, chasing, and biting behaviours were recorded from five minutes of video footage analyzed at 30, 20, and 10 minutes prior to spawning, during spawning, and 10 and 20 minutes after spawning (i.e. -30, -20, -10, 0, 10, and 20). Chasing and biting observations were summed across these observation periods to obtain a total of 30 minutes of observations. Quivering observations were not totaled. All courtship data were analyzed with Sigma Stat (SPSS Inc, Chicago, IL) using a Mann-Whitney rank-sum test or a t-test (determined by normality of the distribution of the data), and spawning success was analyzed using a Fisher Exact test. In all circumstances the null hypothesis was that no difference existed between N, T or C populations, the level of significance was set at a=0.05, and different letters are used to represent statistical differences occurring within each time interval.  3.3 RESULTS Dominance hierarchies among males were established relatively early in all the competitive trials. The male that most actively courts the female and displays the most aggressive behaviour (i.e. biting and chasing) towards the competitive male is identified  53 as the male dominant. The dominant male also obtains primary access to the female during spawning. Nontransgenic males were always dominant when they were included in the trial. In all trials where N and T males competed and spawning occurred, the N male was the successful spawner (Table 3.1). When N and C males competed, all spawning events were won by the N male, except for one spawning in which both males were participants. Insufficient data were available from T versus C trials to accurately identify which male was dominant; both males displayed quivering behaviour towards the female and during only one observation period was a T male seen chasing a C male. During competition between the T and C males, only one spawning occurred and both males participated. Nontransgenic males always displayed more courtship and aggressive behaviour than transgenic and cultured males. Prior to spawning, N males quivered significantly more than T males during N and T male competition (Figure 3.1). During competition between N and C males, the N male quivered significantly more than the C male during all time intervals except during spawning (Figure 3.2), however; during spawning, C fish did not display any quivering behaviour. In all trials, N males displayed significantly more biting and chasing than either the T or C males with the exception of biting during competition between N and T males (Figures 3.3 and 3.4). Although biting during competition between N and T males is not statistically significant, T males did not display any biting towards N males.  s  c ™ o cu o  T 3 C C O o) 5 X J C O w iC -A 2 £ di 0) CO +0-») (3 A a > (> A- CL CO _c C > 5 (A «> E o C "E 3O T 3 CD M > o (A T 3 C 3C W CO£ C T c CO CO o _c CD Z 1* o" 3 O •o .E § 8 o (A § o .2 v C UD o ? J o ^ C D (A CMi cCD ^ m co 0) o kD CO < T3 3 C C cn c (A COoE c CO 'E •c +1 5 o 3 O (A T3 co o i © Q. 'E CA CO > CD. = CM CO"55 Q. P > E LO CO "co o O c .2o AC o (a, « D o ca > C 3 | (A CO C) C H CD CO "o 0 (C A 3 D Dc Ui C iZ CO a P 4-*  r  LO  f  00  CM  UI1AJ 9 / SJ3AJI10  55  ii- S ro ro  UIIAI g / sjQAmo  56  ;ant  en t(± ver  pei  0) Q. E o  ns  o  ipu  3  c o  o ffl C 0)O CO (A (A E cn o z (A CO > c >. CU (0 O) d) o (A c .52 CO 'E +•» JD o c  nuiiu  •o o  •a  an  "S  ds  iber aw the peti tio nb ese sta  o o  to >  c 3 Z c o  t lett  £o  £  CO o CO (0  m  CJ CO (0 o 3 c  igure bserv rror). 1 versi  AO  Male ns be ale  c o  E i— o a. cu u c  c CO < 4> sz CO a. 3 (A E co TJ i_ o c C V o o o o O) o (A c— T3 CO c 5 2! 3 'O Z Q L 0) k. 0) (A Q. O 3 o CJ (0 "> i _ m CU n (A 1 -  no  o 'E cu cn <o c CO  T3 C 3 O (0  LL o  "UI|/M oe / sang  z  57  ujifl 0£ /  SGSBLIO  58 3.4 DISCUSSION The results of this chapter indicate that, in a laboratory setting, both transgenic and cultured males are inferior to nontransgenic males during competition for sexually mature females. Transgenic and cultured males displayed less courtship and aggressive behaviour, and access to ripe females was clearly dominated by nontransgenic males. The inferiority of T and C males may stem from different sources. Both body size (Fleming and Gross, 1992, 1993 and 1994) and secondary sexual characteristics (Fleming and Gross, 1994, Quinn and Foote, 1994) influence a male's competitive spawning ability. In this study, T males were size-matched to N-males, but T males did not develop a kype and red body coloration. Thus, T males are apparently inferior to N males because the former do not develop secondary sexual characteristics that are appropriate for their size if a fighting mating tactic is used. Fleming and Gross (1994) suggested that coho salmon with longer snouts obtain greater access to sexually mature females. Nor do T males show any indication that during competition with an N male a sneaking tactic may be utilized. Sneaking may be difficult for T males due to their rounder body shape and inferior swimming ability as compared to N males (refer to Chapter 1.3, Table 1.2. and Farrell et al. 1997), and perhaps the mobility of a male is critical in obtaining access to females during matings. In contrast, C males were always smaller than their competitors, and the inferiority of C males relative to N males was related to this size difference. During competition between C and N males there was some indication that cultured males can mate by adopting a sneaking tactic. This was indicated by the trial in which both males from an N and C pair participated in spawning. It is not possible to say whether the secondary sexual characteristics of C fish influenced  59 their ability to spawn with ripe females because the tanks used in this study did not provide the types of habitat (i.e hiding places) in which characters like cryptic coloration might be beneficial. It is reasonable to suggest that lack of kype and red coloration in C and T fish are a consequence of the rearing environment and are not affected by the presence of the transgene. However, size alone can not be an explanation for increased mating success between males. During competition between a T and C male, Jboth males participated in spawning, both actively courted the female, and the dominant male could not be determined. It is important to recognize that mating tactics and secondary sexual characteristics act together in determining a male's ability to spawn successfully during competition, and some secondary sexual characteristics have probably evolved to enhance the success of particular tactics (Gross, 1985). The sneaking tactic can result in fertilization success that is not significantly different from that of the alpha male, even though the sneaker is smaller and farther away from the female (Foot et al. 1997). These experiments do not allow us to determine whether the reduction in courtship and aggressive behaviour in transgenic salmon is caused by genetic or environmental factors, or a combination of both. However, these data do provide a clear indication that the culture environment can have negative effects on male breeding success and the presence of the transgene does not increase breeding success in a laboratory setting. Caution should be exercised in interpreting these data because different phenotypes may result if transgenic salmon are reared in different environments. It is possible that escaped transgenic salmon that complete their life cycle in nature may have very different phenotypes from the transgenic fish used in these experiments. Perhaps the only way to further explore the risk that transgenic salmon may have on wild  60 populations, without releasing genetically modified fish onto natural spawning grounds, is to conduct studies, such as the ones presented here, and then incorporate the results into simulation models.  61 GENERAL SUMMARY  The goal of this thesis was to examine the initial reproductive performance of transgenic fish with fish found in the wild. Major environmental concerns associated with using genetically modified fish in aquaculture are the direct impact, and sustained genetic impact, that transgenic fish may have on naturalfishpopulations if they are released into the natural environment. One way to address such concerns is to obtain an understanding of the possible reproductive abilities of transgenic fish. This thesis focused on three aspects related to the reproductive success of transgenic fish. First, adult phenotype and female and male gamete quantity and quality were examined. Secondly, female and male courtship and spawning behaviour were examined, as well as transgene transmission from transgenic fish to their offspring. Third and finally, male competitive behaviour was examined. Results from these studies found several differences between nontransgenic and transgenic fish in regards to reproductive effort. Although, no differences in body weight between N and Tfishwere observed, transgenic fish were much rounder in body shape than nontransgenic fish. Transgenic males had a less developed kype and lacked the red body coloration that nontransgenic fish developed, but no differences in male gamete quantity or quality were observed. In contrast, transgenic females were more fecund than nontransgenic females, but may have inferior quality gametes due to reduced egg size. Transgenic females spawned less frequently then nontransgenic females and displayed consistently low levels of courtship behaviour. No courtship behaviour differences between transgenic and nontransgenic males were observed. However, during  62 competition for access to an ovulated female, transgenic males were inferior to nontransgenic males. Transgenic males obtained no spawnings, displayed significantly less courtship behaviour toward the female, and displayed significantly less aggressive behaviour towards the competitive male. However, these studies are the first to have shown that in a simulated natural environment, transgenic fish possess gametes that can result in viable offspring, display courtship behaviour, and are capable of spawnings that result in viable transgenic offspring. The second goal of this thesis was to consider whether reproductive differences found between transgenic and nontransgenic fish were a result of environmental or genetic differences between the two groups of fish. In order to achieve this goal, fish from a similar genetic background and environment needed to be examined. Therefore, . nontransgenic fish raised in the same culture environment were utilized to interpret the effect of the culture environment. It was difficult to decipher whether environmental or genetic differences were the cause of the observed differences in reproductive effort between transgenic and nontrangenic fish. Indeed, the culture environment results in both reduced growth rates and body weight of fish, and these effects can be compensated for by the presence of the transgene. As well, body shape was directly affected due to the culture environment, and this divergence from wild body shape seemed to be intensified by the presence of the transgene. However, it was not clear whether increased fecundity in transgenic fish was a result of the transgene, the culture environment, or the fact that transgenic fish matured early. Likewise, the lack of red coloration and well developed kype in males occurred in both cultured and transgenic males making it difficult to determine the cause of these differences. Although cultured nontransgenic fish invested  63 more in gonadal development, this was not the case with transgenic fish, and therefore, it is difficult to conclude whether environment or genetics is the determining factor in reproductive investment of males. The data provide some indication that GSI and sperm production may be closely related to body size. Finally, courtship and spawning ability were negatively affected by both the culture environment and the presense of the transgene. The ultimate purpose of this thesis is to address the effects that genetically modified fish may have on natural populations. Since these observations were made in a laboratory setting, this study provides only an approximation to the behaviour of transgenic fish and their possible interaction with wildfishin the natural environment. Although these studies provide a significant contribution to the knowledge of the reproductive success of growth enhanced transgenic animals, genotype-environment interactions can strongly influence the fitness of fish. If transgenic fish escaped into the natural environment, the resulting phenotypes and behaviours may be different from those observed in this study. Perhaps another way to further explore the risk that transgenic animals may have on wild populations without releasing genetically modified organisms into natural environments, is to conduct studies like the ones presented in this thesis, and then incorporate the results into simulation models.  64 BIBLIOGRAPHY  Abrahams, M.V. and Sutterlin, A. 1999. The foraging and anti-predator behaviour of growth-enhanced transgenic Atlantic salmon. Animal Behaviour. 58:933-942. Allen, G.H. 1958. Notes on the fecundity of silver salmon (Oncorhynchus kisutch). The Progressive Fish-Culturist. 20: 163-169. Baker, D.M., Davies, B., Peirce, A.L., Dickhpff, W.W. and Swanson, P. 2000. Effects of fasting and metabolic hormones on the reproductive axis of coho salmon, Oncorhynchus kisutch. Reproductive Physiology of Fish. 478-480. Beacham, T.D. 1982. Fecundity of coho salmon (Oncorhynchus kisutch) and chum salmon (O. keta) in the northeast Pacific Ocean. Canadian Journal of Zoology.. 60:1463-1469. Begon, M., Harper, J.L., and Townsend, CR. 1996. Life-history variation. In Ecology, Individual, Populations and Communities (3 ed.). pp.526-566. Oxford: Blackwell Science, Ltd. rd  Berejikian, B.A., Tezak, E.P., Schroder, S.L., Knudsen, C M . and Hard, J.J. 1997. Reproductive behavioral interactions between wild and captively reared coho salmon (Oncorhynchus kisutch). ICES Journal of Marine Science. 54:10401050. Berejikian, B.A., Tezak, E.P., and LaRae, A.L. 2000. Female mate choice and spawning behaviour of chinook salmon under experimental conditions. Journal of Fish Biology. 57:647-661. Bouk, G.R. and Jacobson, J. 1976. Estimation of salmonids sperm concentration by microhematocrit technique. Transactions of the American Fisheries Society. 105:534-535. Burner, C J . 1951. Characteristics of spawning nests of Columbia River salmon. Fish. Bull. Fish. Wildl. Serv. 61:97-110. Chebanov, N.A. 1990. Spawning behavior, assortative mating, and spawning success of coho salmon, Oncorhynchus kisutch, under natural and experimental condition. Journal of Ichthyology. 30(6): 1 -12. Chourrout, D., Guyomard, R. and Houdebine, L.M. 1986. High efficiency gene transfer in rainbow trout (Salmo gairdneri Rich) by microinjection into egg cytoplasm. Aquaculture. 51:143-150.  65 Crone, R.A., and Bond, C E . 1976. Life history of coho salmon, Oncorhynchus kisutch, in Sashin Creek, southeastern Alaska. Fishery Bulletin. 74: 897-923. Devlin, R.H. 1997. Transgenic salmonids. In: Houdebine, L.M. (ed.) Transgenic Animals Generations and Use. Harwood Academic Publishers, France. ppl05117. —. 1998. Benefits, Limitations, and Risks of Transgenic Fish in Aquaculture. ICES Conference, Caseous, Portugal. —. 2000. Difficulties in Ecological Risk Assessment of Transgenic and Domesticated Fish. SFU Workshop. March 2-3. Devlin, R.H., Biagi, C.A., Yesaki, T.Y., Smailus, D.E., and Byatt, J.C. 2001. Growth of domesticated transgenic fish, a growth-hormone transgene boosts the size of wild but not domesticated trout. Nature. 409:781-782. Devlin, R.H. and Donaldson, E.M. 1992. Containment of genetically altered fish with emphasis on salmonids. In: Hew, C L . and Fletcher, G.L. (ed.) Transgenic Fish. World Scientific, New Jersey. Devlin, R.H., Johnsson, J.L, Smailus, D.E., Biagi, C.A., Joensson, E., and Bjornsson, B. Th. 1999. Increased ability to compete for food by growth hormone-transgenic coho salmon Oncorhynchus kisutch (Walbaum). Aquaculture Research. 30:479482. Devlin, R.H., Yesaki, T.Y., Biagi, C.A., Donaldson, E.M., Swanson, P. and Chan, W.K. 1994a. Extraordinary salmon growth. Nature. 371:209-210. Devlin, R.H., Yesaki, T.Y., Donaldson, E.M., Du, S.-J. and Hew, C L . 1995a. Production of germline transgenic Pacific salmonids with dramatically increased growth performance. Canadian Journal of Fisheries and Aquatic Sciences. 52:1376-1384. Devlin, R.H., Yesaki, T.Y., Donaldson, E.M., and Hew, C L . 1995b. Transmission and phenotypic effects of an antifreeze/GH gene construct in coho salmon (Oncorhynchus kisutch). Aquaculture. 137:161-169. Donaldson, E.M. 1997. The role of biotechnology in sustainable aquaculture. In: Bardach, J.E. (ed.) Sustainable Aquaculture. John Wiley and Sons, Inc., New York, NY. Drucker, B. 1972. Some life history characteristics of coho salmon of Karluk River system, Kodiak Island, Alaska Fisheries Bulletin (U.S.). 70:79-94.  66 Einum, S., and Fleming, LA. 2000. Highly fecund mothers sacrifice offspring survival to maximize fitness. Nature. 405: 565-567. Farrell, A.P., Bennett, W. and Devlin, R.H. 1997. Growth-enhanced transgenic salmon can be inferior swimmers. Canadian Journal of Zoology. 75:335-337. Fleming, LA. 1998. Pattern and variability in the breeding system of Atlantic salmon (Salmo salar), with comparisons to other salmonids. Candian Journal of Fisheries and Aquatic Sciences. 55(Suppl. l):59-76. Fleming LA., and Einum, S. 1997. Experimental tests of genetic divergence of farmed from wild Atlantic salmon due to domestication. ICES Journal of Marine Science. 54:1051-1063. Fleming, LA., and Gross, M.R. 1990. Latitudianl clines: a trade-off between egg number and size in pacific salmon. Ecology. 71: 1-11. —. 1992. Reproductive behavior of hatchery and wild coho salmon: does it differ? Aquaculture. 103:1-21. —. 1993. Breeding success of hatchery and wild coho salmon (Oncorhynchus kisutch) in competition. Ecological Applications. 3(2) :230-245. —. 1994. Breeding competition in a Pacific salmon (Coho: Oncorhynchus kisutch): measures of natural and sexual selection. Evolution. 48:637-657. Fleming, LA., Hinder, K., Mjoelneroed, LB., Jonsson, B., Balstad, T., and Lamberg, A. 2000. Lifetime success and interactions of farm salmon invading a native population. Proceedings of the Royal Society of London, Series B: Biological Sciences. 267(1452):1517-1523. Fleming, LA., Jonsson, B., and Gross, M.R. 1994. Phenotypic divergence of searanched, farmed and wild salmon. Can. J. Fish. Aquat. Sci. 51: 2808-2824. Fleming, LA., Jonsson, B., Gross, M.R. and Lamberg, A. 1996. An experimental study of the reproductive behaviour and success of fanned and wild Atlantic salmon. Journal ofApplied Ecology. 33:893-905. Fletcher, G.L., Shears, M.A., King, M.J., Davies, P.L. and Hew, C.L. 1988. Evidence for antifreeze protein gene transfer in Atlantic salmon. Canadian Journal of Fisheries and Aquatic Sciences. 45:352-357. Foerster, R.E., and Pritchard, A.L. 1941. Observations on the relation of egg content to total length and weight in the sockeye salmon (Oncorhynchus nerka) and the pink salmon (O. gorbushcha). Transactions of the. Royal Society of Cananda. Ser. 3, Sec. 5. 35:51-60.  67  Grachev, L. Ye. 1971. Alteration in the number of oocytes in the sockeye [Oncorhynchus nerka (Walb.)] during the marine period of its existence. Journal of Ichthyology. ll(6):897-906. Grainger, R. 1998. Recent Trends in Global Fishery Production. Fisheries Department, FAO, Rome. Groot, C., and Margolis L. 1991. Pacific Salmon Life Histories. Vancouver: UBC Press. Gross, M. R. 1984. Sunfish, salmon, and the evolution of alternative reproductive strategies and tactics in fishes. Pp.55-75. In R. Wooton and G. Potts, eds. Fish Reproduction: strategies and tactics. Academic Press, London. —. 1985. Disruptive selection for alternative life history in salmon. Nature. 313:47-48. —. 1991. Salmon breeding behaviour and life history evolution in changing environments. Ecology. 72: 1180-1186. Hanson, A.J., and Smith, H.D. 1967. Male selection in a population of sockeye salmon (Oncorhynchus nerka) of mixed age groups. Journal of Fisheries Research Board of Canada. 24:1955-1977. Healey, M.C. and Prince, A. 1998. Alternative tactics in the breeding behaviour of male coho salmon. Behaviour. 135:1099-1124. Heath, D.D., Fox, C.W., and Heath, J.W. 1999. Maternal effects on offspring size: variation through early development of chinook salmon. Evolution. 53(5): 16051611. Hedrick, P.W. 2001. Invasion of transgenesfromsalmon or other genetically modified organisms into natural populations. Canadian Journal of Fisheries and Aquatic Sciences. 58:841-844. Helfman, G.S., Collette, B.B., and Facey, D.E. 1997. Early life history, la. The Diversity of Fishes, pp.117-134. Oxon, Blackwell Science, Ltd. Houde, A.E. 1997. Sex, color and mate choice in guppies. Princeton University Press. Hoysak, D.J., and Liley, N.R. 2001. Fertilization dynamics in sockeye salmon and a comparison of sperm from alternative male phenotypes. Journal of Fish Biology. 58:1286-1300. Jaervi, T. 1990. The effects of male dominance, secondary sexual characteristics and female mate choice on the mating success of male Atlantic salmon Salmo salar. Ethology. 84(2): 123-132.  68  Johnsson, B. 1997. A review of ecological and behavioural interactions between cultured and wild Atlantic salmon. ICES Journal of Marine Science. 54:10311039. Johnsson, J.L, Petersson, E., Jonsson, E., Bjornsson, B.Th. and Jarvi, T. 1996. Domestication and growth hormone alter antipredator behaviour and growth patterns in juvenile brown trout, Salmo trutta. Canadian Journal of Fisheries and Aquatic Sciences. 53:1546-1554. Jonsson, B. 1997. A review of ecological and behavioural interactions between cultured and wild Atlantic salmon. ICES Journal of Marine Science. 54:1031-1039. Kapuscinski, A.R. and Hallerman, E.M. 1991. Implications of introduction of transgenic fish into natural ecosystems. Canadian Journal of Fisheries and Aquatic Sciences. 48:99-107. Keenleyside, M.H.A., and Dupuis, H.M.C. 1988. Courtship and spawning competition in pink salmon (Oncorhynchus gorbuscha). Canadian Journal of Zoology. 66:262-265. Kinnison, M.T., Unwin, M.J., Hendry, A.P., and Quinn, T.P. 2001. Migratory costs and the evolution of egg size and number in introduced and indigenous salmon populations. Evolution. 55(8):1656-1667. Liley, N.R., Fostier, A., Breton, B., and Tan, E.S.P. 1986. Endocrine changes associated with spawning behavior and social stimuli in a wild population of rainbow trout (Salmo gairdneri). General and Comparative Endocrinology. 62(1):157-167. Linley, T.J. 2001. Influence of short-term estuarine rearing on the ocean survival and size at return of coh salmon in southeastern Alaska. North American Journal of Aquaculture. 63(4):306-311. McBride, J.R., Higgs, D.A., Fagerlund, U.H.M. and Buckley, J.T. 1982. Thyroid and steroid hormones: potential for control of growth and smoltification of salmonids. Aquaculture. 28:201-209. McLean, E, Devlin, R.H., Donaldson, E.M. and Byatt, J.C. 1996. Bovine placental lactogen invokes rapid growth in chinook salmon. Aquaculture Nutrition. 2:243248. McPhail, J.D., and Lindsey, C C . 1970. Freshwater fishes of northwestern Canada and Alaska. Bulletin Fisheries Research Board of Cananda. pp: 173:3 81.  69 Mjolnerod, I.B., Fleming, I.A., Refseth, U.H., and Hindar, K. 1998. Mate and sperm competition during multiple-male spawnings of Atlantic salmon. Canadian Journal of Zoology. 76:70-75. Mori, T. and Devlin, R.H. 1999. Transgene and host GH gene expression in pituitary and nonpituitary tissues of normal and GH transgenic salmon. Molecular and Cellular Endocrinology. 149:129-139. Muir, W.M. and Howard, R.D. 1999. Possible ecological risks of transgenic organism release when transgenes affect mating success: sexual selection and the Trojan gene hypothesis. Ecology. 96(24):13853-13856. Ostenfeld, T.H., Devlin, R.H. and McLean, E. 1998. Transgenesis changes body and head shape in Pacific salmon. Journal of Fisheries Biology. 52:850-854. Pandian, T.J. and Marian, L. A. 1994. Problems and prospects of transgenic fish production. Current Science. 66(9):635-649. Quinn, T.P., and Foote, C.J. 1994. The effects of body size and sexual dimorphism on the reproductive behaviour of sockeye salmon, Oncorhynchus nerka. Animal Behaviour. 48:751-761. Rounsefell, G.A. 1957. Fecundity of North American salmonidae. Zn Fisheries Bulletin. U.S. Department of the Interior, Edited by Interior, U.S.D.o.t. pp. 451-468. Scott, A.P., and Liley, N.R. 1994. Dynamics of excretion of 17 alpha, 20 beta dihydroxy-4-pregnen-3-one 20-sulphate, and of the glucuronides of testosterone and 17 beta -oestradiol, by urine of reproductively mature male and female rainbow trout (Oncorhynchus mykiss). Journal of Fish Biology. 44(1): 117-129. Sower, S.A., Iwamoto, R.N., Dickhoff, W.W. and Gorbman, A. 1983. Ovulatory and steroidal responses in coho salmon and steelhead trout following administration of salmon gonadotropin and D-Ala super(6), Des-Gly super(lO) ethylamide (GnRHa). At: Conference: Salmonid Reproduction: An Inernaltional Symposium., (198, P.6, Publ. Washington Sea Grant Program. Stockley, P., Gage, M.J.G., Parker, G.A., and Moller, A.P. 1997. Sperm competition if fishes: the evolution of testis size and ejaculate characteristics. 772e American Naturalist. 149(5):933-954. Svardson, G. 1949. Natural selection and egg number in fish. Institute of Freshwater Research, Drottningholm Report Number 29:115-122. Tautz, A.F. and Groot, C. 1975. Spawning behavior of chum salmon (Oncorhynchus keta) and rainbow trout (Salmo gairdneri). J. Fish Res. Board Can. 32(5):63342.  70  Taylor, E.B., and McPhail, J.D. 1985. Variation in body morphology among British Columbia populations of coho salmon, Oncorhynchus kisutch. Canadian Journal of Fisheries and Aquatic Sciences. 42:2020-2028. Tiedje, J.M., Colwell, R.K., Grossman, Y.L., Hodson, R.E., Lenski, R.E., Mack, R.N. and Regal, P.J. 1989. The planned introduction of genetically engineered organisms: ecological considerations and recommendations. Ecology. 70(2):298315. Vielkind, J., Vielkind, U., von Grotthus, E. and Anders, F. 1971. Uptake of bacterial H DNA into fish embryos. Experientia 27:347-348. 3  Vladic, T.V., and Jarvi, T. 2001. Sperm quality in the alternative reproductive tactics of Atlantic salmon: the importance of the loaded raffle mechanism. Proceedings of the Royal Society of London B. 268:2375-2381. Zhu, Z., Xu, K., Li, G., Xie, Y. and He,L. 1986. Biological effects of human growth hormone gene injected into the fertilized eggs of loach Misgurnus anguillicaudatus (Cantor). Kexue Tongbao. 31:988-990.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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


Related Items