| The yellow perch, Perca flavescens, has a native distribution throughout the Nearctic ecozone from South Carolina to Nova Scotia, westward throughout the Great Lakes region and the Mississippi Valley, and northward to the Red River Basin. The mild taste and firm flesh with low fat and phospholipid content make this species a traditional regional favorite with consumers. In addition, this species is a popular recreational angling resource. However, dramatic reductions in population sizes of yellow perch have been underway in the Great Lakes area since approximately 1950. At present, commercial fishing of yellow perch has diminished or ceased altogether in some states surrounding the Great Lakes, whereas yellow perch still has high market demand and value in its native regions. No doubt, this species holds tremendous potential for aquaculture in its native region. Despite of the recent technical advancements in yellow perch aquaculture methods, this species is still considered as an ’alternate aquaculture species’. A major constraint to the expansion of the yellow perch aquaculture industry is the slow growth rate of currently cultured populations of this species. As part of the effort to enhance yellow perch aquaculture production, our lab (Aquaculture Genetics and Breeding Laboratory, Ohio State University South Centers, USA) has initiated a selective breeding program aimed at solving the’slow growth’problem in this species.Growth-related traits are quantitative traits, which are determined by several minor genes and environmental factor. Quantitative genetics is born for animal breeding. Theories of quantitative genetics can unmask additive genetic contributions to performance traits, making the breeding program more effective. Aquatic animal breeding has a long history. In the development progress of aquatic animal breeding, a lot of breeding methods occur, like hybrid breeding, ploidy breeding, sex control breeding and molecular marker assisted breeding, etc. Compared to other breeding methods, molecular marker assisted breeding shows a large potential. Aiming to get good growth traits varieties for yellow perch early, this paper firstly built a paternity testing technology for this species. Then based on this technology, genetic parameters for growth-related traits in yellow perch were evaluated comprehensively. While two microsatellites-assisted breeding methods for selecting fast yellow perch were established and compared. The main contents included the following:(1) Construction of a paternity testing technology in yellow perch There were two procedures. The first procedure was to estimate the microsatellite markers selected for establishment of a paternity testing technology in yellow perch. According to five rules which were:1) the base number of core repeat sequence were greater than 2 (markers with 3 or 4 bp repeat were good); 2) the result of PCR amplification was good; 3) makers were with same or similar annealing temperature; 4) genotyping was in good conditions; 5) there were obviously differences in sizes of the markers selected, we finally chose eight microsatellites (i.e. YP30, YP41, YP49, YP60, YP73, YP78, YP96 and YP109) from 150 yellow perch microsatellites developed by our group (fish genetic and breeding laboratory, Ohio State University South Centers, USA) to establish a paternity testing technology in yellow perch. Three broodstock populations (2006-year population,2007-year population, and 2008-year population) were sampled. Genomic DNAs were extracted by using ammonium acetate/isoamyl alcohol method. After one-tube, single-reaction nested PCR, genotype informations of the three populations were obtained by using genetic analyzer (ABI 3130) and software GeneMapper 4.0. Evaluation results of the number of alleles (A), observed heterozygosity (Ho), expected heterozygosity (He), polymorphism information content (PIC), inbreeding coefficient (Fis), allelic richness (Rs), and Hardy-Weinberg equilibrium (HWE) showed that the 8 microsatellites we chose were highly polymorphic, indicating these microsatellites were suitable for establishment of a paternity testing technology in yellow perch. Meanwhile, the software P-Loci was also used for estimation of the selected microsatellites. Simulation results showed that:when the family number was 100, the success ratio of paternity testing could achieve 100% by using the eight microsatellites. It indicated that these microsatellites were fit for establishment of a paternity testing technology in yellow perch. The second procedure was an application assessment. 13 dams and 21 sires (one dam×two sires for twelve mating sets and one dam×one sire for one mating set) were used to establish families. By through DNA extraction, genotyping and utilization of software Cervus 3.0,94% offspring can finally assign to a single pair. A paternity testing technology in yellow perch was successfully constructed.(2) Estimation of heritabilities and genetic correlations of growth-related traits in yellow perchThirteen dams and 21 sires (one dam×two sires for twelve mating sets and one dam×one sire for one mating set) were used to produce full-sib families. The families were communally reared in ponds and tanks. Pedigree informations were obtained by using the paternity testing technology in yellow perch. By using software ASREML3.0, heritabilities and genetic correlations of growth-related traits (i.e. body weight and total length) in one-year-old and two-year-old yellow perch cultured in ponds and tanks were estimated. Heritabilities for one-year-weight and one-year-length of yellow perch cultured in ponds were 0.082±0.056 and 0.075±0.053, respectively. Heritabilities for two-year-weight and two-year-length of yellow perch cultured in ponds were 0.14±0.09 and 0.049±0.057, respectively. The low heritabilities for body weight and total length of pond-cultured yellow perch suggested that family selection was much better for growth traits breeding of pond-cultured yellow perch. Heritabilities for one-year-weight and one-year-length of yellow perch cultured in tanks could not be estimated. While heritabilities for two-year-weight and two-year-length of yellow perch cultured in tanks were high, about 0.5. Since the sample size was very small for estimating the heritabilities for body weight and total length of tank-cultured yellow perch, the results we obtained could just be reference data. The genetic and phenotypic correlations between weight and total length of one-year-old and two-year-old yellow perch from different environment (ponds and tanks) were high. Their values were from 0.907±0.014 to 0.999±0.0055. The high correlation between length and weight in this study suggested that selective breeding for increased weight in yellow perch could be achieved using an indirect selection method based on length.(3) Heritabilities and genotype by environment interaction for body weight of yellow perch174 broodfish (58(?); 116(?)) with passive integrated transponder tags (PIT) were selected to generate 58 half-sib families, which were randomly divided into 4 populations (P1, P2, P3 and P4). There were two main types of culture environments, water temperature (25℃and 16℃) and fish density (70 individuals/tank (ind./tank),50 ind./tank and 30 ind./tank). The combinations were 25℃-70 ind./tank,25℃-50 ind./tank, 25℃-30 ind./tank, and 16℃-50 ind./tank. In each combination,4 populations were reared into four tanks (0.5 diameter), respectively. At the end of year one, half of fish in each tank were randomly selected for body weight measurement and fin-cut. Pedigree informations were obtained by using the paternity testing technology in yellow perch. By using software ASREML3.0, heritabilities and genotype by environment interaction for body weight of one-year-old yellow perch were estimated. heritabilities for body weight of one-year-old yellow perch cultured with 16℃,25℃,70 ind./tank,50 ind./tank and 30 ind./tank were 1.00±0.00,0.49±0.26,0.18±0.16,0.49±0.26, and 0.94±0.35, respectively. All genetic correlations for body weight between different temperatures, and different densities were moderate (range=-0.17-0.58), suggesting the occurrence of genotype×environment effects for the trait.(4) Correlations between Growth Traits and Heterozygosity, Allelic Distance (d2) at Microsatellite Loci in the Yellow Perch1163 individuals were genotyped with eight microsatellite markers (YP30, YP41, YP49, YP60, YP73, YP78, YP96 and YP109). Using regression analyses (Pearson correlation), correlations between genetic parameters (microsatellite heterozygosity and mean square allelic distance (d2)) and total length, body weight of yellow perch reared in different culture conditions (four ponds), ages (one-year-old and two-year-old), and sexes were assessed. There were no significant associations found between heterozygosity, d2 and growth traits of yellow perch reared in different culture conditions, ages, and sexes (except one, where P value for correlation coefficients was 0.046; for all the others, P values were greater than 0.05, ranged from 0.063 to 0.975). These results suggested that i) the hypothesis of associative overdominance might be rejected in yellow perch, or ii) the microsatellites used here might be located in genes or in the proximity of genes uncoupled with growth. Moreover, another hypothesis also could be proposed that the heterozygote advantage might be found in yellow perch fingerlings since the HFCs are expected to decrease or disappear with age.(5) Study of microsatellites-assisted breeding methods for selecting fast yellow perchYellow perch broodfish used in the study were selected from the base generation of the genetic improvement program at the Ohio State University South Centers. Thirteen dams and 21 sires were used to produce the experimental fish. Each dam was put into a 55L round tank with one or two sires for spawning (one dam×two sires for twelve mating sets and one dam×one sire for one mating set). Four sires were used twice during mating. Dams spawned naturally in the tanks, resulting in eggs of each dam fertilized by either one or both sires (for one dam×two sires). Fertilized eggs obtained from different mating sets were separately incubated in 25L round tanks with flow-through well water for 11-12 days at 11-12℃. A maximum of 25 full-sib families were expected from the selective pairings. Thirteen mating sets were successfully hatched, and similar numbers of fry from each set were combined and stocked into earthen ponds for the six-week nursery phase. Subsequently, feed training was conducted in 400L round tanks for 3 weeks. A total of 6,100 feed-trained fingerlings were stocked into each of four 0.1 ha earthen ponds (labeled Pond 4,6,7 and 8) in June 2006 and communally reared for 21 months. At the end of year one, two ponds (Pond 4 and 7) were graded based on the following procedure:100 fish were randomly collected from each of the two ponds and their lengths measured and ordered to determine the size-cut-off points for the top 50% fish. Based on these cut-off points,10-20 fish were selected from each pond group for testing to properly set the grader bar gap. Then, the top 50% of fish were passively graded from the remaining fish from each of the two ponds. The graded top 50% fish were restocked to two same sized ponds for year-2 grow-out (method two, two-stage selection, TSS). The other two ungraded ponds (Pond 6 and 8) were harvested and restocked to other two ponds without grading (method one, one-stage selection, OSS).At the end of year-one rearing, all four ponds were drained and harvested. All fish from each pond were counted and group-weighed (drained weight) to the nearest 1 g to determine total biomass. A total of 150 fish were randomly sampled from each of the four ponds at the end of year one before grading selection for weight and total length measurements and fin-clipped. At the end of year-two rearing, the same sampling procedures were conducted as at the end of year one. A total of 148 and 146 fish were randomly sampled from pond 4 and 7 with OSS and 122 and 147 fish were similarly collected from pond 6 and 8 with TSS. In addition, a total of 137,127,111 and 105 fish were obtained as the top 10% largest fish from pond 4,7,6 and 8, respectively. A non-lethal biopsy (fin clip) obtained from each specimen (including broodfish and progeny) was preserved immediately in 95% ethanol for DNA analyses and subsequent parentage analysis.High success rate of paternity testing guaranteed the application of method one (OSS). Based on the pedigree information obtained by using the paternity testing technology, growth traits of yellow perch from the two breeding methods were analyzed by using SPSS 17.0 and ASREML 3.0. In the present study, within OSS genetic correlation between one-year-weight and two-year-weight were high (0.98), indicating that the growth of yellow perch recorded at year one could predict their growth for year two. In addition, mean family weights and family EBVs (estimate breeding values) for weight between year one and year two were found to be significant correlated within OSS. These results indicate that the fastest growing yellow perch families in year one would continue to be the fastest growing families in year two. Extremely significant differences (P<0.01) in weights and total lengths for random fish and the top 10% largest fish recorded at year two between the OSS and TSS were found here, suggesting year-two yellow perch selected using the TSS method were significantly heavier and longer than year-two yellow perch undergoing the OSS method. Therefore, based on the results described above we concluded that the two-stage selection method was desirable and more effective for yellow perch breeding comparing to one-stage selection in terms of improving selection efficiency and reducing costs (feed, pond/tank and labor et al.). |
PDF Full Text Request