USE OF MULTIGENERATIONAL STUDIES TO ASSESS GENETIC STABILITY, FITNESS, AND COMPETITIVE ABILITY OF TRANSGENIC JAPANESE MEDAKA: II. PRODUCTION OF TRANSGENIC FISH AND PRELIMINARY STUDIES

William M. Muir^a, Richard Howard^b, Robert Martens^b, and Christopher A. Bidwell^a

^aDepartment of Animal Sciences, and ^bDepartment of Biological Sciences; Purdue University, West Lafayette, IN 47907

SUMMARY

Risk assessment of fish, transformed with growth hormone, to natural aquatic species needs to be established before industry can take advantage of this technology. Research to date has concentrated on production of transgenic medaka and predicting their effects when introduced into a wild population. One cell medaka zygotes were microinjected with one of two constructs containing the human growth hormone gene (hGH) driven either by the chicken -actin promoter (cA) or the Atlantic salmon growth hormone (sGH) promoter. Medaka eggs injected with the cA/hGH construct had greater than 60% mortality by first feeding and the sGH/hGH injected eggs had approximately 15% mortality by first feeding. Seven cA/hGH transgenic and ten sGH/hGH transgenic founders were identified by PCR amplification of a portion of the hGH gene. Of these 17 founders, two sGH/hGH and one cA/hGH were found to transmit the transgene to progeny. Eggs were collected from matings of the three founders for six days and the resulting full and half siblings from each founder were reared together in a 40 l aquaria and fed four time daily for 8 weeks. The fry were weighed individually and assayed for the transgene by PCR. Transgenic progeny were only identified from one sGH/hGH founder, which transmitted the transgene at a frequency of 4.3%. The transgenic progeny had an average weight of 67 mg which was 22.9% larger than their non-transgenic full and half siblings whose average weight was 54 mg. Three generations of introgression of the transgene from the founder into the base population has since occurred and direct examination of competitive fitness is currently being conducted. Potential effects of growth hormone on overall fitness of fish include changes in mating success, fecundity, or viability (competitive ability) because body size affects all life history characteristics in such species.

In preliminary studies, experiments were conducted using natural body size variation in a cohort of medaka to examine size effects on mating success. Results showed that large males obtained about 80% of the matings relative to the 20% obtained by small males. This four-fold mating advantage resulted because large males were preferred as mates by females and because large males could control access to sexually receptive females better than small males could. The size difference between large and small males in these experiments is similar to that expected to occur between wild type and transgenic males. Thus, based on these experiments using natural body size variation, we estimated the rate of spread of a growth hormone transgene in a population of wild type fish. We assumed that the only beneficial effect of the transgene was to increase the mating success of the larger transgenic males relative to their smaller wild type counterparts, and that there was no detrimental physiological effects associated with the transgene. Our results showed that an initially rare transgene could spread quickly in a population. If a transgenic male had four times the mating success of a wild type male and the initial frequency of transgenic males in the population was one in 100,000, then the transgene would comprise more than 50% of the population in 16 generations; if the initial frequency of transgenic males was higher (say, one in a thousand), 50% of the population would be transgenic in 11 generations. Because transgenic effects on other life history attributes (e.g., female fecundity, age at sexual maturation of both sexes) are also expected to occur due to enhanced growth rates, but were not included in these calculations, we consider these projections to be conservative estimates of how rapidly a transgenic line could spread in a wild type population

Key words: Performance testing, fish, transgenic, risk, fitness components

INTRODUCTION

Transgenic technology has attracted commercial and scientific interest because fast growing fish are desirable in both aquaculture and sport fishery and DNA sequences for growth hormone (GH) genes and cDNA's are well characterized. GH sequences have already been transferred to several game and commercial fish species (See Muir et al., 1994 for review). Because larger body size often confers higher fitness in fish due to enhanced competitive ability and fecundity, transgenic individuals could eliminate their wild-type conspecifics in nature. Thus, assessing potential risks of transgenic fish to the environment is imperative.

Because transgenic organisms cannot be released into the wild, the potential environmental impact of such organisms must first be inferred from inspection of causal components in a laboratory environment. If a transgenic organism exhibits superior fitness in an optimal laboratory environment in terms of growth, fecundity, competitive ability etc., the organism could be a hazard in the wild despite nature's less than optimal conditions and greater complexity. On the other hand, if a transgenic organism has low fitness in an optimum laboratory environment, the common conclusion is that it will not thrive or compete effectively in the wild where conditions are usually suboptimal. However, this argument has one flaw: evolution. Even if a transgenic organism initially displays low fitness, the cumulative action of natural selection over several generations could modify the expression of the transgene and make the organism more successful than native species. The net result could be an ecological catastrophe. Thus, multigenerational studies are needed to assess risk of transgenic release.

The overall objectives and methods for this research were presented by Muir et al. (1994). Here we discuss production of transgenic fish and preliminary experiments on sexual selection as related to body size in fish. Our sexual selection experiments concentrated on the mating success fitness component of males and exploited the natural size variation that exists in a cohort of medaka. Sexual selection occurs when individuals with different trait expressions obtain differential mating success and is a well documented phenomenon in all metazoan taxa (see Andersson, 1994 for a recent review). Sexual selection consists of two components: mate competition (interactions among members of the same sex for mating opportunities) and mate choice (proclivities to mate with individuals of the opposite sex that possess particular characteristics). Mate competition is typically more intense in the sex with less parental investment in young (typically males); mate choice is more developed in the sex with more parental investment (typically females) (Trivers, 1972). The common result of sexual selection is the evolution of trait elaborations limited chiefly to one sex, usually males. These sexual dimorphisms may involve morphology, behavior, physiology, and life history. The significance of sexual selection for risks associated with transgenic release is that when transgenic traits are also sexually selected ones, transgenes may increase in wild type populations at a surprisingly high rate.

Although the magnitude of size difference between the small and large male medaka used in our experiments closely corresponds to that expected to occur between wild type males and transgenic males, our results must be considered preliminary until confirmed by similar experimentation using transgenic individuals. However, the impact of sexual selection on transgenic animals should always be considered a potential risk because body size enhancement is a common goal in the production of transgenic animals and body size is a sexually selected trait in many animal species including medaka (as we show below). In a recent review, Andersson (1994) found that body size was affected by sexual selection in 40% (74/186) of animal species investigated and in 18 of the 36 (50%) fish species reviewed.

MATERIALS AND METHODS

The study organism is a non-commercial fish, Japanese medaka (Oryzias latipes). Medaka are small (< 4 cm), have a short generation interval (10 weeks), and are highly conducive to laboratory experimentation. Because evolutionary mechanisms are universal, inference is directly applicable to commercial species; i.e., if one species of transgenic organism can increase in fitness due to selective modification across generations, then other species can as well. We used the Purdue Orange-Red (POR) strain of medaka to produce transgenic founder lines. This strain was established from a large heterogeneous, randomly mating population of medaka homozygous for the mutant color Orange-Red. The medaka were reared using infusoria as an initial food source, followed by trout rations of AP100, trout mash and #1 trout starter as the animals got older. Mature broodstock were fed Tetra Min prior to and during breeding. Broodstock were kept in aquaria containing algae and were occasionally fed artemia.

Founders were produced using the cytoplasmic microinjection technique of Ozato et al. (1992). Eggs were collected from females within 90 minutes of fertilization. The single-celled eggs were then microinjected with approximately 200 pg of transgene DNA in 20 pl that was linearized and free of prokaryotic DNA sequences. The microinjected eggs were incubated in a custom-made hatching tank in the presence of methylene blue to prevent fungus growth (Blacklidge and Bidwell, 1992).

We inserted both the chicken ß-actin (CBA) promoter and the Atlantic salmon growth hormone promoter (sGH) driving the expression of the human growth hormone (hGH) gene. The CBA promoter and sGH promoter should differ in their effects. The CBA promoter is not tissue-specific and expression in a variety of tissues may cause unexpected side effects. The ectopic expression of GH due to a constitutive promoter like CBA may result in unusual phenotypes due to altered organ or tissue physiology. The sGH promoter may increase growth through more normal physiological regulation. The sGH promoter was PCR amplified using sequence information from Johansen et al. (1989) and a sGH/hGH construct was generously supplied to us by Dr. Geoffrey Waldbieser.

PCR was used as a first assay for identifying founders and subsequent progeny testing for germline transmission or mRNA expression. A simple protocol for rapid PCR sampling based on the method of Higuchi et al. (1989) was used. For this protocol, small pieces of fin tissue were immersed in 10 mM NaOH and boiled for 5 minutes. Ten l of that solution was added to a standard PCR reaction. A pair of 30 mer PCR primers were used for hGH transgene analysis. These primer sequences are in exons 2 and 4 and amplify a 661 bp fragment from the gene and a 360 bp fragment from the cDNA:PCR assay. These primers work well at 55C annealing temperature and do not cross react with endogenous medaka GH.

Transgene integration was verified using southern blotting. Southern blot hybridizations are used for a qualitative analysis of the transgene insertion into the genome of the animal. High molecular weight DNA was isolated from the founder and digested with restriction enzymes. DNA fragments were separated by electrophoresis in an agarose gel. The DNA was denatured and transferred to a nucleic acid binding membrane by vacuum blotting. The DNA was covalently bound to the membrane by UV cross-linking. A [³²P] labeled hGH probe was then used to hybridize to genomic DNA for identification of the target sequences. Nucleic acid hybridization protocols followed the methods of Maniatis et al. (1982).

Each confirmed founder was backcrossed to the POR strain for 2 generations before homozygous individuals were produced. Backcrossing will allow the transgene to be integrated into a heterogeneous genetic background and reduce the subsequent rate of inbreeding to less than 1% per generation.

Several aspects of medaka life history were relevant to the sexual selection experiments we conducted. Males and females become sexually mature at approximately 10 weeks of age. The sexes are sexually dimorphic in fin morphology, but not body size. A female may release between 1 to 40 eggs per day. Fertilization is external; however, following copulation, eggs remain attached to the female's vent for several hours or until the female removes the eggs by rubbing them onto a substrate. Medaka provide no parental care beyond gamete investment (Yamamoto, 1975). Females may mate daily for up to a year, and in nature all mating occurs at dawn. When placed under a constant photoperiod (10L:14D) and temperature (27C), all reproductive behavior occurs within an hour of "lights on" (when lights come on each day in the laboratory to simulate dawn).

Two size classes of males were used in the sexual selection experiments based on their standard length (distance between the snout to the end of the caudal peduncle): large males were > 1 SD from the mean of a sample of 84 measured males; small males were < 1 SD from the mean. All males were members of the same cohort measured within the same time period, approximately 2 weeks after attaining sexual maturity. Length measurements were obtained using dial calipers. Body mass was also measured with a Mettler AE-163 balance by placing individuals in a small water-filled cup for weighing.

Courtship was initiated by males and lasted for a few seconds to several minutes. A courting male would approach the female from underneath and perform a "circle dance" in which he would face the same direction as the female and swim in a tight circle (approximately 3 cm in diameter) beneath the female. Courting males may be unsuccessful for several reasons including: disruption by other males, a form of mate competition and female rejection. Females reject courting males either by swimming away or by performing a "tail drop" behavior (defined by Walter and Hamilton, 1970 as "head up") to prevent copulation. During a successful copulation, the male presses his vent directly next to female's vent and vibrates his body rapidly as the female releases eggs.

All experiments were conducted in 40 l aquaria. Two different types of mating experiments were conducted, behavioral experiments and density experiments. We maintained the same sex ratio and male size ratio in both types of experiments (2 males per female and equal numbers of large and small males, respectively). Behavioral experiments provided information on the relative importance of mate competition and mate choice; density experiments yielded data on male and female reproductive behaviors in a more realistic setting, but could not be used to distinguish the effects of the two sexual selection components. Here, we report only data on male mating success; behavioral analyses are provided in Martens et al. (in prep.). All individuals used in this experiment were haphazardly chosen from a stock population with the only criteria being that the males fit the size criteria described above, and females were extruding eggs when collected to ensure that she was gravid. Males and females were used only once throughout the experiments

Behavioral experiments involved two males (1 large and 1 small). We placed the female and each male into separate compartments in the same aquarium approximately 16 hours prior to each trial, and allowed them to acclimate to their surroundings. Each male occupied an end compartment in the aquarium; the female was in the central compartment. During this time, opaque covers were placed over the Plexiglass partitions that separated the compartments to prevent the fish from seeing each other. At "lights on" the next day, the opaque covers were removed and the female's association preference in the absence of male competition was determined on the basis of the percent time she remained near each male (Martens et al., in prep.). After this portion of the experiment was completed, the Plexiglass barriers were removed and mating behavior was observed. The three fish then were allowed to interact freely in the aquarium for another 24 hr and mating behavior was observed again at "lights on" the next day.

Four density levels (each with three replicates) were used in density experiments: 2, 3, 4, or 6 females with 4, 6, 8 or 12 males, respectively. Unlike the behavioral experiments, males and females were housed continuously in the same aquaria and thus could interact freely throughout the course of experimentation. Each day at "lights on", we observed the male courtship displays, male aggression, and copulations in one randomly chosen replicate.

RESULTS AND DISCUSSION

Production of Transgenic Fish. We injected 881 medaka eggs with the c A/hGH construct, of which 342 fry (31%) were alive by first feeding. We also injected 842 eggs with the sGH/hGH construct of which 716 fry (85%) were alive by first feeding. Seven cA/hGH transgenic and ten sGH/hGH transgenic founders were identified by PCR amplification. Of these 17 founders, two sGH/hGH and one cA/hGH were found to transmit the transgene to progeny. Eggs were collected from matings of the three founders for six days and the resulting full and half siblings from each founder were reared together in a 40l aquaria and fed four times daily for 8 weeks. The fry were weighed individually and assayed for the transgene by PCR. Transgenic progeny were only identified from one sGH/hGH founder, which transmitted the transgene at a frequency of 4.3%. The transgenic progeny had an average weight of 67 mg which was 22.9% larger than their non-transgenic full and half siblings whose average weight was 54 mg. The frequency of the transgene in the second generation of backcrossing (BC2) was approximately 1:1 and consistent with Mendelian inheritance of the transgene. Transgenic fish in BC2 were also approximately 23.6% larger than their non-transgenic sibs.

Preliminary Studies: A sample of males and females collected from the same large cohort were similar in standard length (males: 26.63 ± 3.09 mm, mean ± SD, n = 84; females: 26.05 ± 3.23 mm, n = 72; t = 1.14, P = 0.26). The sexes were also similar in mass (males: 0.32 ± 0.12 g, n = 84; females: 0.32 ± 0.11 g, n = 72; t = 0.07, P = 0.95).

The male size distribution was used to determine size cutoffs for the large and small males used in mating experiments. Exact lengths and masses were obtained for each male used in the behavioral experiments; sizes of males used in density experiments were only assessed relative to cutoff values. The 134 large males used in the behavioral experiments averaged 30.90 ± 1.65 mm in standard length and 0.46 ± 0.07 g in mass; the 134 small males used averaged 23.38 ± 1.87 mm in standard length and 0.22 ± 0.05 g in mass. The average size difference between males used in the same replicate of these experiments was 7.49 ± 2.21 mm in length and 0.24 ± 0.08 g in mass. The 134 focal females used in the experiments were chosen haphazardly from a large population and averaged 28.78 ± 2.81 mm in standard length and 0.41 ± 0.11 g in mass.

We observed 80 copulations in 135 trials of the behavioral experiments. Large males obtained 66 of the 80 copulations (83%) which revealed a significant mating advantage (Sign test; P < 0.01). The combined data from the three replicates of the four density levels in the density experiments also revealed a significant male size advantage in mating success. Large males obtained 31 of the 39 observed copulations (79.5%) (Sign test; P < 0.01). The high degree of similarity that we observed in the mating advantage of large males in these two different types of mating experiments suggests that our results should apply to a variety of social and ecological circumstances for medaka. The lack of body size dimorphism in medaka is also consistent with a large male mating advantage in nature. In many species of fish (and lower vertebrates), the sexes are usually similar in size only when the reproductive advantages males obtain in sexual selection approximate those that females obtain in increased fecundity. In our study, the size advantage in fecundity to large females was apparent in that larger females produced larger clutches, on average, each day (R² = 0.26; P = 0.016; N = 22).

Estimation of effect of sexual selection on frequency of the transgene: A model which incorporates differential viability, fecundity-fertility, and mating success on gene frequencies was given by Muir and Bell (1986), fitness parameters for which are given in Table 1. The recurrence equation for the expected frequencies of the genotypes in each generation is developed as follows:

Let pkt = frequency of kth genotype in the tth generation (i= 1, +/+; 2, +/gH; 3, gH/gH)

Wt = overall fitness of the populations in the tth generation

Wt = SiSjSk (pitfi)(pjtmj)vijk

then the expected frequencies of genotypes in the next generation are:

pk(t+1) = SiSj (pitfi)(pjtmj)vijk/Wt for k=1, 2, 3

with expected gene frequency of the transgene at generation t + 1 is:

z(t+1) = p₃(t+1) + p₂(t+1)/2

Based on results of the preliminary experiment, we set the fitness parameters for male mating advantage of 0.8 for transgenic males and 0.2 for normal. All other fitness parameters were set to 1. Resulting changes in gene frequency are therefore only due to a male mating advantage. We assumed starting frequencies of transgenic homozygous males of 1 in 1,000, 1 in 10,000, and 1 in 100,000 (to mimic an accidental release or escape of a single transgenic male), and predicted changes in gene frequency through subsequent generations (Figure 1). These results show that for every 10-fold increase in initial frequency, the time necessary for the transgene to be in the majority decreases by approximately 2.5 generations. Thus, even for an extremely rare transgenic male in a large population, the number of generations for the transgene to become the dominant form is less than 16, which can be achieved by medaka in less than 2 years. Thus, if the only difference between transgenic and wild type medaka was a male mating advantage, transgenic medaka would rapidly replace wild types in a natural population. Other factors such as reproductive or ecological advantage could further increase the rate of spread.

The above predictions concerning the rate of spread of a rare transgene in a natural population due to a male mating advantage involves many assumptions. Not the least of which is that the viability of transgenic offspring is the same as that of wild-type offspring. A more likely situation is for transgenic individuals to have reduced viability. Although we have no information on the relative viability of transgenic individuals to date, we can make some predictions using the above population genetic model by assigning various viability values to transgenic offspring relative to their wild-type counterparts. If one homozygous transgenic male is released into a natural population of 10,000 fish, the population consequences depend on the degree of viability reduction of transgenic young. These consequences can be assessed in terms of average population fitness (Figure 2), the rate of spread of the transgene (Figure 3), and the population size (Figure 4). If transgenic individuals have extremely low viability (40% relative to wild type), the transgene never becomes established in the population (Figure 3). The average population fitness is unaffected (Figure 2) as is the population size (Figure 4). If transgenic viablity is between 40% to 60% of that of wild type, the transgene remains in the population (Figure 3) but spreads slowly, achieving majority status only after many generations (>> 40). If transgenic viability is somewhat higher, between 60% to 80% of that of wild type, the transgene achieves majority status fairly quickly, after 16 to 28 generations depending on the exact viability level (Figure 3). Nevertheless, with viability of transgenic fish in the broad range of 50% to 90% of that of wild type, population size is predicted to greatly decrease (Figure 4) as a result of the genetic load imposed by the transgene. In this intermediate range, the genetic load imposed on average population fitness (Figure 2) is predicted to doom the population to extinction (Figure 4) within as few as 28 generations.

The above predictions must be interpreted cautiously. Critical assumptions of the deterministic model used are likely to be violated in nature. We are presently refining these predictions by incorporating density dependence into the model using age-structured demographic models and allowing for stochasity in population size. We predict that these refinements will rescue the population from extinction when the transgene has a low effect on viability, but that lower transgenic viabilities will still threaten population existence. As before, extremely low transgenic viability should have little or no effect on the natural population because the transgene never becomes established. In addition, predictions of both the population genetic model and the demographic models will be directly tested by the long-term maintenance of transgenic/ wild-type populations in mesocosms.

ACKNOWLEDGMENTS

We wish to express appreciation to Deb Miles for maintaining the fish and completing the PCR tests and Mark Smith for their assistance in setting up the experimental systems. This research was supported by USDA grant 93-33120-9468.

REFERENCES

Andersson M (1994) Sexual Selection. Princeton University Press.

Johansen B, Johnson OC, Valla S (1989) The complete nucleotide sequence of the growth-hormone gene from Atlantic salmon (Salmo salar). Gene 77:317-324

Maniatis T, Fritsch EF, and Sambrook J (1982) Molecular cloning a laboratory manual. Cold Spring Harbor Laboratories, New York.

Muir WM, Bell AE (1987) Multiple vital functions of the daughterless (da) gene in Drosophila melanogaster and factors influencing its expression. Genetica 72:43-54.

Muir WM, Howard RD, and Bidwell C (1994) Use of multigenerational studies to assess genetic stability, fitness, and competitive ability of transgenic Japanese medaka: I. methodology. In: Proc of Sixth International Conference on Risk Assessment Methodologies. College Park, Maryland.

Trivers RL (1972) Parental investment and sexual selection. In: Sexual Selection and the Descent of Man, 1871-1971, Campbell B (ed.), Heinemann, London. Pages 136-179.

Figure 1. Rate of increase of transgene.

Figure 2. Effect of differential viability of transgenic fish on average population fitness

Figure 3. Effect of differential viability of transgenic fish on frequency of the transgene.

Figure 4. Effect of differential viability of transgenic fish on population size.