The turbot Scophthalmus maximus(Linnaeus)is aspecies of demersal marine flatfish that naturallyinhabits the Baltic,Black, and Mediterranean Seas(Blanquer et al., 1992). It was first introduced intoChina in 1992(Wang et al., 2010; Ruan et al., 2011).The production of cultured turbot in China increasedsignificantly as a result of a technological breakthroughin the large-scale artificial breeding of this fish in1999(Ma et al., 2006). Consequently,turbot is nowan important commercial fish species in China.However,in recent years,inbreeding has led to aserious deterioration in the quality of the germplasm,resulting in high mortality and a decline in theproduction of turbot. Better breeding management,including genetic improvement,is necessary tosustain the healthy development of this industry.
The government of China has supported geneticimprovement of turbot and ,since 2006,a number ofacademic institutions have carried out continuouslarge-scale breeding programs. The Yellow SeaFisheries Research Institute has already madesignificant progress in this area,particularly inachieving faster growth rates. A fast-growing strainwas obtained by selective breeding over twogenerations. To date,however,there has been norigorous analysis of the morphological characters ofthe new strain.
To precisely characterize the morphology of thefast-growing phenotype and to identify morphologicalmarkers of this quantitative character,a slow-growingstrain was obtained. In this paper,the morphologicalcharacters of the fast-growing and slow-growingstrains of turbot at different growth stages arecompared statistically. This study provides valuable information in relation to the development of new varieties of turbot by sustained selection for fastergrowth.2 MATERIAL AND METHOD 2.1 Selection of the fast-growing and slow-growing strains of turbot
The fast-growing and slow-growing strains ofturbot were selected from a breeding program initiatedin 2007 by the Yellow Sea Fisheries ResearchInstitute,Chinese Academy of Fishery Science. InApril 2007,56 F1 full-sib families were obtainedusing a nested mating design with one male and twofemales(full-sibs have both parents in common and half-sibs have one parent in common)by ChinaTianyuan Aquaculture Ltd. An F2 breeding programwas developed based on the individual breedingvalues and inbreeding coefficients of the F1 animals and 54 full-sib families were produced in April 2010.The body weights of all F2 families were measured atintervals of 3 months between ages 3 and 27 months and the two families having the highest and lowestgrowth rates were selected through comprehensiveanalysis of these data. The criterion for selection ofthe fast-growing strain was that the mean body weightwas consistently higher than that of the slow-growingstrain throughout the period of observation.2.2 Rearing conditions
To obtain similar rearing conditions for all F1 and F2 families during the early breeding stage,measureswere taken to st and ardize both the stocking density offish and the environment. At 15,30, and 45 days postspawning,the number of larvae or juveniles in each full-sib family was st and ardized by r and om samplingto 10 000,5 000, and 2 000,respectively. At 2 monthsof age,r and om samples of 1 000 young fish from bothof the selected full-sib families were transferred toseparate 12-m3 concrete tanks. When the fish grew to5–6 cm in length(around 3 months of age),samplesof 250–300 fish were r and omly selected from eachtank for tagging using Visible Implant Elastomer(Northwest Marine Technology,Inc.,USA) and thenstocked communally. Dead fish were removed in atimely fashion. The environmental conditions werest and ardized during the larval and juvenile cultureperiod at: water temperature 13–18°C,salinity 30–40,illumination intensity 500–2 000 lx,pH 7.8–8.2, and dissolved oxygen >6 mg/L. During the period 3 to 27months,the above five indices were 15–18°C,25–30,500–1 500 lx,pH 7.6–8.2 and >6 mg/L,respectively.2.3 Statistical analysis
At 3-monthly intervals,from ages 3 to 27 months,the body length(BL),total length(TL) and bodywidth(BW)of each communally stocked fish wasmeasured with a precision of 0.01 cm. The ratios TL/BL,BW/BL and TL/BW were used as indices tocompare the morphometric dynamics of the twostrains. Differences in the three indices between thefast-growing and slow-growing strains in the samemonth were analyzed using t-tests, and differences ofeither the fast-growing or slow-growing strain amongdifferent months were analyzed by one-way analysisof variance(ANOVA) and multiple comparison.Outliers were checked using box plots and thenormality of each variable was checked using theShapiro-Wilk test. Statistical analyses were performedusing SPSS 13.0 software.3 RESULT
All data sets were normally distributed(ShapiroWilk test). In the fast-growing strain,during theperiod 3 to 27 months,the ranges of the measuredindices were: TL/BL=1.215 9–1.326 3,BW/BL=0.592 5–0.865 7 and TL/BW=1.503 9–2.065 9.For the slow-growing strain,the corresponding valueswere 1.265 1–1.337 0,0.603 8–0.825 6 and 1.577 6–2.122 1,respectively(Table 1). For TL/BL,the fast-growing and slow-growing strains exhibiteddifferent trends throughout the measurement period;e.g.,in the fast-growing strain there were maximum and minimum values at 9 and 15 months of age,respectively; by contrast,in the slow-growing strain the trend was reversed with a minimum at 9 months and a maximum at 15 months(Fig. 2). For BW/BL,the trends were similar in the fast-growing and slowgrowing strains; the mean values progressively increased,with some fluctuation(Fig. 3). In contrast,TL/BW exhibited decreasing trends in both strainsover the same period(Fig. 4), and also fluctuated; e.g.,TL/BW at age 15 months was slightly higher than thatat 12 months in both strains,before continuing todecline.
The results of a correlation analysis between themorphometric ratios and the body weights of turbot indifferent months are given in Table 2. Two of thecorrelation coefficients of TL/BL with body weightwere significantly different from zero at the 5% level(-0.232±0.011 and 0.180±0.050) and three werehighly significant at the 1% level(-0.436±0.000,-0.316±0.000 and -0.572±0.000). Six were highlysignificant negative correlations(-0.436±0.000,-0.232±0.011,-0.316±0.000,-0.572±0.000,-0.042±0.674 and -0.058±0.531) and two were lowpositive values(0.180±0.050 and 0.030±0.767).Correlations of TL/BW with body weight were all negative. Therefore,the correlations of TL/BL and TL/BW with body weights in different months wererather weak. In contrast,the correlations of BW/BLwith body weight were greater and all were highlysignificantly positive(P<0.01),except for a nonsignificant negative correlation at 3 months(-0.061±0.511). The correlation coefficient betweenBW/BL and body weight was highest at 18 months(0.652±0.000) and lowest at 24 and 27 months(0.317±0.000).
Comparison of the three morphometric ratiosbetween the fast-growing and slow-growing strainsshowed that: TL/BL was lower in the fast-growingstrain,except at 9 and 18 months of age(Table 1 and Fig. 2); BW/BL was higher in the fast-growing strain,except at 3 months(Table 1 and Fig. 3), and TL/BW ofthe fast-growing strain was consistently lower at eachmeasurement time(Table 1 and Fig. 4).4 DISCUSSION
Studies of dynamic changes in growth traits withdevelopment have generally used linear or nonlineargrowth models(Kuhi et al., 2003; Khamis et al., 2005; Roush and Branton, 2005). However,in thispaper,the fit to several linear and nonlinear timeseries models of the three indices in the two strains was imperfect(R2 values ranged from 0.800–0.89).Furthermore,inspection of the fitted curves indicatedmarked divergence from the actual changes in thethree indices. Clearly,differences in the behavior ofthe three indices could not be readily explained usingmathematical models. Hence the dynamic changes ofTL/BL,BW/BL and TL/BW were analyzed usingonly the actual data collected at different points intime. p>
The curvilinear temporal characteristics of thethree indices in the two strains clearly revealeddynamic changes and differences in the growthperformance of the two turbot groups. Although thecurves for TL/BL of both strains showed four extremevalues during the period 3 to 27 months,the shapes ofthe curves of the two strains differed. In contrast,BW/BL exhibited a similar rising trend in the twostrains and TL/BW exhibited similar falling trendsover the same period. These patterns showed that thebodies of both fast-growing and slow-growing strainsof turbot changed progressively toward a morerounded shape. These changes were particularlynoticeable during the period 3 to 6 months(Figs.2–4).TL/BL was generally lower,BW/BL was generallyhigher, and TL/BW was always lower in the fastgrowing strain than in the slow-growing strain.Because we eliminated environmental factors duringculture of the two strains,the differences in thedynamic changes in body shape, and the differencesin the three morphometric ratios,between the twostrains must be attributed to genetic factors. BW/BL and TL/BW showed very similar trends. For example,at 15 months of age,BW/BL of both strains wasslightly falling and TL/BW slightly rising,whichfurther suggests that the body shape changes of thetwo strains and the differences of body shape betweenthe two strains depended mainly on genetic factors.Changes in the body shapes of female and male turbotduring growth and development have also beencompared in relation to TL/BL,BW/BL and TL/BW and all three showed no significant differences(Wang et al., 2014),which further suggests that themorphological differences between the two strainsreflected their particular genetic characteristics and were not influenced by gender. Correlation analysisindicated negative or low positive correlationsbetween TL/BL and body weight, and negativecorrelations were observed between TL/BW and bodyweight. In contrast,highly significant large negativecorrelation coefficients were observed between BW/BL and body weight. Therefore,BW/BL could beused as a phenotypic marker during selective breedingof turbot for body weight.
Fish body shape is not constant during growth and development but displays a complex dynamic change.Phenotypic plasticity and temporal and spatialpatterns of gene expression provide a theoretical basisfor the complexity of fish body shape. Theunderst and ing of phenotypic plasticity,which refersto the ability of an organism to express different phenotypes depending on the biotic or abioticenvironment(West-Eberhard 1989; Agrawal,2001;Kerschbaumer et al., 2011),has progressedsignificantly over the past few decades,for fishes(Pigliucci,2005; Kerschbaumer et al., 2011),otheranimals(Cowley and Atchley, 1992; Atchley et al., 1994; Atchley and Zhu, 1997; Wang et al., 2006), and plants(Zhu,1995; Shi et al., 2001,2002).Environmental factors can have subtle effects on themorphometrics of fishes(Meyer,1987; Day et al., 1994; Ellis et al., 1997). Temporal and spatial patternsof gene expression indicate that genes are expressedselectively during different growth periods; differentgenes are activated and deactivated at specificdevelopmental stages(Shi et al., 2001). The presentstudy has shown that although fish body shapechanges in a complex and dynamic way during growth and development,the process exhibits a degree ofuniformity in relation to specific morphometricindices.
|Agrawal A A. 2001. Phenotypic plasticity in the interactions and evolution of species. Science, 294 (5541): 321-326.|
|Atchley W R, Xu S Z, Vogl C. 1994. Developmental quantitative genetic models of evolutionary change. Development al Genetics, 15 (1): 92-103.|
|Atchley W R, Zhu J. 1997. Developmental quantitative genetics, conditional epigenetic variability and growth in mice. Genetics, 147 (2): 765-776.|
|Blanquer A, Alayse J P, Berrada-Pkhami O et al. 1992. Allozyme variation in turbot (Psetta maxima) and brill (Scophthalmus rhombus) (Osteichthyes, Pleuronectoformes, Scophthalmidae) throughout their range in Europe. Journal of Fish Biology, 41 (5): 725-736.|
|Cowley D E, Atchley W R. 1992. Quantitative genetic models for development, epigenetic selection, and phenotypic evolution. Evolution, 46 (2): 495-518.|
|Day T, Pritchard J, Schluter D. 1994. A comparison of two sticklebacks. Evolution, 48 (5): 1 723-1 734.|
|Ellis T, Howell B R, Hayes J. 1997. Morphological differences between wild and hatchery-reared turbot. Journal of Fish Biology, 50 (5): 1 124-1 128.|
|Kerschbaumer M, Postl L, Koch M et al. 2011. Morphological distinctness despite large-scale phenotypic plasticity— analysis of wild and pond-bred juveniles of allopatric populations of Tropheus moorii. Naturwissenschaften, 98 (2): 125-134.|
|Khamis A, Ismail Z, Horan K et al. 2005. Nonlinear growth models for modeling oil palm yield growth. Journal of Mathematics and Statistics, 1 (3): 225-233.|
|Kuhi H D, Kebreab E, Lopez S et al. 2003. An evaluation of different growth functions for describing the profile of live weight with time (age) in meat and egg strains of chicken. Poultry Science, 82 (10): 1 536-1 543.|
|Ma A J, Chen C, Lei J L et al. 2006. Turbot Scophthalmus maximus : stocking density on growth, pigmentation and feed conversion. Chinese Journal of Oceanology and Limnology, 24 (3): 307-312.|
|Meyer A. 1987. Phenotypic plasticity and heterochrony in Cichlasoma managuense (Pisces, Chichlidae) and their implications for speciation in cichlid fishes. Evolution, 41 (6): 1 357-1 369.|
|Pigliucci M. 2005. Evolution of phenotypic plasticity: where are we going now? Trends in Ecology and Evolution, 20 (9): 481-486.|
|Roush W B, Branton S L. 2005. A comparison of fitting growth models with a genetic algorithm and nonlinear regression. Poult ry Science, 84 (3): 494-502.|
|Ruan X H, Wang W J, Kong J et al. 2011. Isolation and analysis of microsatellites in the genome of turbot (Scophthalmus maximus L.). African Journal of Biotechnology, 10 (4): 507-508.|
|Shi C H, Wu J G, Fan L J et al. 2001. Developmental genetic analysis of brown rice weight under different environmental conditions in indica rice. Acta Botanica Sinica, 43 (6): 603-609.|
|Shi C H, Wu J G, Lou X B et al. 2002. Genetic analysis of transparency and chalkiness area at different filling stages of rice (Oryza sativa L.). Field Crops Research, 76 (1): 1-9.|
|Wang C H, Li S F, Liu Z G et al. 2006. Developmental quantitative genetic analysis of body weight and morphological traits in red common carp, Cyprinus carpio L. Aquaculture, 251 (2-4): 219-230.|
|Wang X A, Ma A J, Huang Z H et al. 2010. Heritability and genetic correlation of survival in turbot (Scophthalmus maximus). Chinese Journal of Oceanology and Limnology, 28 (6): 1 200-1 205.|
|Wang X A, Ma A J, Huang Z H et al. 2014. Developmental differences between female and male groups in turbot (Scophthalmus maximus) breeding families. Journal of Fisheries of China, 38 (4): 464-469. (in Chinese with English abstract)|
|West-Eberhard M J. 1989. Phenotypic plasticity and the origins of diversity. Annual Review of Ecology and System atics, 20 (1): 249-278.|
|Zhu J. 1995. Analysis of conditional genetic effects and variance components in developmental genetics. Genetics, 141 (4): 1 633-1 639.|