Heritability estimates for growth‐related traits in the Pacific oyster (Crassostrea gigas) using a molecular pedigree

Pacific oyster is one of the leading species in world aquaculture, but heritability estimation applying mixed‐family approach has not been actively pursued. In this study, heritability for growth‐related traits in the Pacific oyster was first estimated by creating a single cohort of 45 families in a full‐factorial mating design consisting of nine sires and five dams. A total of 270 offspring were analysed and parentage assignment inferred by six microsatellite markers achieved 100% success. All parents contributed to the spawn and a total of 42 full‐sib families were represented. Using an animal model, heritability estimates at 12 months of age were 0.49 ± 0.25 for shell height, 0.36 ± 0.19 for shell length, 0.45 ± 0.23 for shell width and 0.35 ± 0.17 for wet weight. Genetic correlation between shell height and wet weight was quite high (0.79 ± 0.25), suggesting that direct selection of shell height, which is an easily measurable trait, also improves wet weight. The results obtained in this study indicate that growth‐related traits could be improved by exploiting additive genetic effects through selective breeding.     

Genetic parameters of growth and survival in the Pacific oyster Crassostrea gigas

The Pacific oyster Crassostrea gigas is a representative bivalve mollusc that is widely cultured in the world and is the largest molluscan group cultured in China. In order to assess the feasibility of improving survival of C. gigas through genetic selection, the heritability and genetic correlations for growth and survival traits between different life stages were examined. Genetic parameters were estimated based on intraclass correlations of 49 full‐sib families (29 half‐sib families) in larvae (4 and 20 days after fertilization) and spat (140 days after fertilization) stages. The heritability for growth traits in larvae and spat was 0.30–0.86 and 0.53–0.59, respectively, and varied with ages. The heritability of survival was low in larvae (0.13 ± 0.05 and 0.17 ± 0.04, respectively for 4 and 20 days after fertilization) but medium (0.39 ± 0.07) in spat, suggesting that selection for increasing spat survival was feasible. The genetic correlation between growth traits within age was medium to high and positive (ranging from 0.47 to 0.96, respectively, between shell length (SL) and shell height (SH) at 20 days and between SL and SH at 140 days after fertilization), suggesting that selection to improve single growth trait will cause positive response in another growth traits in C. gigas. The genetic correlations between survival and growth traits at 140 days were low but positive (ranging from 0.23 to 0.27, respectively, between survival and SH and between survival and SL at 140 days after fertilization), suggesting that selection for survival may not have a negative response in growth. Overall, this study suggests that survival traits should be taken as improving target of next selection breeding programme in C. gigas.     

Heritability of larval and juvenile growth for two stocks of Manila clam Ruditapes philippinarum

The heritability of larval and juvenile growth in natural and cultured populations of the Manila clam Ruditapes philippinarum was estimated using an unbalanced nested design and an artificial fertilization technique. A total of 33 full‐sib families and 10 half‐sib families were obtained from the natural population in Shihe (Dalian, Liaoning Province in the north‐eastern part of China), whereas 11 half‐sib families and 33 full‐sib families were generated from the cultured population in Putian (Fujian Province in the south‐eastern part of China). The offspring from each family were reared under the same hatchery and nursery conditions. Results from these studies led to estimates indicating a significantly greater influence due to the dam component compared with the sire component for both natural and cultured populations, which was mainly due to maternal effect. We found that the heritability estimates obtained from the intra‐group correlation of sire half‐sib were precise and unbiased. The narrow‐sense heritabilities of shell length for the Manila clam in the natural population (NP) at larval (at the age of 9 days) and juvenile (at the age of 30 days) stages were 0.22 ± 0.11 and 0.39 ± 0.14, respectively, whereas those in the cultured population (CP) were 0.17 ± 0.09 and 0.87 ± 0.24 respectively. The heritabilities of both NP and CP at 3 days were not significantly different from zero, and were ?0.03 for NP and ?0.14 for CP. These results suggest that selection should be highly effective in Manila clam, and that selecting either a natural or a cultured population for producing faster growing larvae and juveniles should be successful. Selective breeding could thus be used to develop high‐quality seed that would facilitate the development of a shorter culture cycle and higher yields.     

Estimation of genetic parameters for growth traits in cultured clam Meretrix meretrix (Bivalvia: Veneridae) using the Bayesian method based on Gibbs sampling

The aim of this study was to estimate the genetic parameters of the growth performance of clam Meretrix meretrix. As part of the breeding programme, 25 full‐sib families nested within eight half‐sib families were produced, planted out and tested to assess the heritability of the total body weight (TW), shell length (SL), shell height (SH) and shell width (SW) at different growth stages. Method of analysis was implemented using the Bayesian method based on the Multiple Trait using Gibbs Sampling under Animal Model program, which was used to estimate the (co)variance components of the traits and conduct genetic analysis. A total of 25 full‐sib families, each with 30–35 individuals within family, were used for the analysis. Significant positive genetic and phenotypic correlations between SL, SW, SH and TW with each other were observed. The growth traits showed high‐magnitude heritabilities, with values ranging from 0.64 to 0.85, which indicates that these traits should respond to selection and therefore should be included in genetic improvement programmes.     

Genetic parameters for resistance against Vibrio anguillarum in turbot Scophthalmus maximus

Genetic parameters for resistance to Vibrio anguillarum in Scophthalmus maximus were estimated using three different statistical models. Data were recorded from an experimental infection performed on 2,400 individuals from 30 full‐sib groups. Cross‐sectional linear model and cross‐sectional threshold probit model were used to analyse the test‐period survival; the cross‐sectional threshold logit models were used to analyse the test‐day survival. The heritability values estimated by cross‐sectional linear model (CSL), cross‐sectional threshold (probit) model (THRp) and cross‐sectional threshold (logit) model (THRl) were 0.202 ± 0.101, 0.296 ± 0.168 and 0.110 ± 0.023, respectively. The correlation coefficients between the full‐sib families’ estimated breeding values (EBVs) based on these three models were over 0.993. An almost identical ranking of families was generated using any of these models. Accuracy of selection using CSL, THRp and THRl models was 0.783, 0.789 and 0.801, respectively. Accuracy of selection based on the THRl model was higher than that based on CSL and THRp models.     

Characterization,fluctuation and tissue differences in nutrient content in the Pacific oyster (Crassostrea gigas) in Qingdao,northern China

The Pacific oyster (Crassostrea gigas) is one of the most important aquaculture species worldwide. Its meat quality is vital for consumer satisfaction, and nutrient content, especially glycogen, is closely associated with oyster fatness and meat colour. Fluctuations in nutrient content of short‐term starvation have not been previously reported, and seasonal variation in glycogen content in different tissues has rarely been reported. In the present study, we investigated these important aspects of oyster production and found that short‐term starvation (50 hr) did not significantly alter glycogen, protein or lipid content. The seasonal variation assay showed that glycogen and lipid accumulation was high in autumn and winter and that seawater temperature and protein content were inversely related to glycogen content. Glycogen content of the whole flesh was higher from January to April and was positively related to the condition index before the onset of gametogenesis. Glycogen content was higher in the gonad, labial palp and mantle compared to the gill or adductor muscle. Relative expression of genes encoding proteins involved in glycogen metabolism (glycogen synthase, glycogen phosphorylase, glycogen debranching enzyme and glycogen branching enzyme) was closely associated with the glycogen content in the corresponding tissues. Glycogen content in the gonad was regulated by glycogen metabolic and glycolysis pathway genes (6‐phosphofructo kinase, phosphoglycerate kinase, pyruvate kinase, hexokinase and glucose transporters), and stored glycogen was the main energy source for gametogenesis. These findings contribute to oyster aquaculture management and glycogen improvement and expand our understanding of glycogen metabolism in oysters.     

Heritability estimates of larval and spat shell height in the Chilean blue mussel (Mytilus chilensis Hupe 1854) produced under controlled laboratory conditions

The heritability of larval size is estimated in Mytilus chilensis, based on a nested design, involving 95 full‐sib and 19 half‐sib families (19 males each mated to 5 separated females) grown under controlled laboratory conditions. The half‐sib heritability estimates at 10, 25 and 40 days for larval and spat shell height ranged between 0.38±0.33 and 0.84±0.45, suggesting that selection for shell size would be effective.     

Heritability for growth traits in giant freshwater prawn,Macrobrachium rosenbergii (de Mann 1879) based on best linear unbiased prediction methodology

In this study, heritability was estimated for growth‐related traits of giant freshwater prawn (Macrobrachium rosenbergii) before and after morphological sexual differentiation. Estimation was made on data from 16 full‐sib and eight half‐sib families. The variance estimation was performed using a univariate mixed linear animal model and variance components were analysed following an animal model using restricted maximum likelihood procedure using average information algorithm. Heritability estimates (h²) varied considerably with ages. At 2 months old, h²for carapace length (CL; 0.35±0.15) and body weight (BW; 0.26±0.13) were higher than those estimated at 5 months old, based on mixed sex data. However, when data were sorted by sex, h² calculated from data of females were higher than those of males for CL (0.26±0.16 vs. 0.10±0.06), BW (0.28±0.17 vs. 0.12±0.08), body length (0.40±0.17 vs. 0.11±0.07), total length (0.47±0.18 vs. 0.11±0.07) and claw length (0.29±0.16 vs. 0.03±0.04). The same trend was observed for traits at 6 months old in both bulk and individual rearing.     

Growth performance and biochemical composition of the oysters Crassostrea sikamea,Crassostrea angulata and their hybrids in southern China

The Kumamoto oyster (Crassostrea sikamea) and the Portuguese oyster (C. angulata) are important aquaculture species which naturally coexist along the southern coast of China. To understand the potential feasibility of hybridization between the two species, we conducted two‐by‐two factorial cross‐experiments in Beihai (Guangxi province), and also compared the survival and growth of the hybrids to that of the two parental progenies during the grow‐out period from July 2014 to July 2015. Genetic analysis confirmed that the hybrid spats were true hybrids. Additionally, the biochemical composition of the 1‐year‐old oyster progenies was determined. In July 2015, the mean shell height of the hybrids was 42.98 ± 6.29 mm, which was higher than that of the Kumamoto oyster progeny. The cumulative survival rate of the hybrids was 26.37 ± 1.32%, which was higher than that of the progeny of the Portuguese oyster. Mean lipid content of the hybrids was 13.65 ± 1.63% of dry weight, which showed obvious heterosis compared to those of the two parental progenies. Observation of gonads revealed that all hybrids were completely fertile. Furthermore, relative expression of the lipid homeostasis genes, SREBP (sterol regulatory element‐binding proteins), PPARα (peroxisome proliferator‐activated receptor) and INSIG (insulin‐induced gene) were found to vary between parental progenies and the hybrids, thus providing a possible reason for difference in the lipid contents of these experimental groups. Overall, the hybrids were viable, rich in lipid and completely fertile and thus could serve as a promising aquaculture stock for oyster breeding in southern China.     

Heritabilities and genetic correlations of size traits at harvest size in sexually dimorphic Pacific white shrimp (Litopenaeus vannamei) grown in two environments

A captive population of Pacific white shrimp (Penaeus vannamei) replicated in two environments was evaluated for genetic variability and covariability of size traits. A total of 37 full‐sib families, each with an average of 16–20 individuals within family, were used for the analysis. There was no family by environment interaction for any of the traits. However, both fixed effects, sex and environment, were significant. The shrimp grown in environment ‘B’ (Sinaloa) grew larger and heavier than those grown in environment ‘A’ (La Paz), possibly because of the higher temperatures and lower densities in the first one during grow‐out. The females were significantly larger than the males for all traits, except for abdominal (tail) length and weight. The heritabilities, estimated through MTDFREML (multiple‐trait derivative‐free restricted maximum likelihood) after introducing environment and sex as fixed effects, were total length 0.22±0.07, abdominal length 0.23±0.07, cephalothorax length 0.17±0.06, total weight 0.17±0.06, abdominal weight 0.18±0.06, cephalothorax weight 0.15±0.06, and width of the first abdominal segment (0.14±0.05). Genetic correlations (r_g) were high between the length and weight (total and abdominal), but lower for cephalothorax weight or length with the other traits. The width of the first abdominal segment correlations with other traits was intermediate.     

Quantitative genetic properties of four measures of deformity in yellowtail kingfish Seriola lalandi Valenciennes, 1833

The main aim of this study was to estimate the heritability for four measures of deformity and their genetic associations with growth (body weight and length), carcass (fillet weight and yield) and flesh‐quality (fillet fat content) traits in yellowtail kingfish Seriola lalandi. The observed major deformities included lower jaw, nasal erosion, deformed operculum and skinny fish on 480 individuals from 22 families at Clean Seas Tuna Ltd. They were typically recorded as binary traits (presence or absence) and were analysed separately by both threshold generalized models and standard animal mixed models. Consistency of the models was evaluated by calculating simple Pearson correlation of breeding values of full‐sib families for jaw deformity. Genetic and phenotypic correlations among traits were estimated using a multitrait linear mixed model in ASReml. Both threshold and linear mixed model analysis showed that there is additive genetic variation in the four measures of deformity, with the estimates of heritability obtained from the former (threshold) models on liability scale ranging from 0.14 to 0.66 (SE 0.32–0.56) and from the latter (linear animal and sire) models on original (observed) scale, 0.01–0.23 (SE 0.03–0.16). When the estimates on the underlying liability were transformed to the observed scale (0, 1), they were generally consistent between threshold and linear mixed models. Phenotypic correlations among deformity traits were weak (close to zero). The genetic correlations among deformity traits were not significantly different from zero. Body weight and fillet carcass showed significant positive genetic correlations with jaw deformity (0.75 and 0.95, respectively). Genetic correlation between body weight and operculum was negative (?0.51, P < 0.05). The genetic correlations' estimates of body and carcass traits with other deformity were not significant due to their relatively high standard errors. Our results showed that there are prospects for genetic selection to improve deformity in yellowtail kingfish and that measures of deformity should be included in the recording scheme, breeding objectives and selection index in practical selective breeding programmes due to the antagonistic genetic correlations of deformed jaws with body and carcass performance.     

壳白长牡蛎基因型与环境互作(G×E)效应分析

为探索壳白长牡蛎品系的壳色性状和生长性状的基因型与环境互作(G×E)效应,利用巢氏设计构建全同胞家系,每个家系分成两组分别在乳山和荣成海域进行养殖。利用线性混合模型和REML法分析11月龄壳白长牡蛎生长性状和壳色性状的遗传力及G×E效应。采用最佳线性无偏预测法(BLUP法)估计壳高和L~*两个性状的育种值,并通过加权获得综合育种值来筛选优良家系。结果显示,乳山组和荣成组的壳白长牡蛎生长和壳色性状的遗传力不同,分别为(0.14±0.08)～(0.62±0.18)和(0.01±0.03)～(0.78±0.19),可能存在尺度效应。以不同环境为固定效应,综合两个环境计算出的生长和壳色性状的遗传力为(0.02±0.02)～(0.51±0.09),然而由于部分全同胞家系缺失和模型不收敛的原因,估计模型中未包括母本/共同环境效应和显性效应,上述遗传力估计值偏高。本研究中生长和壳色性状在两个环境间的遗传相关为(–0.47±0.40)～(0.75±0.18),均小于0.8,表明壳白长牡蛎品系的生长和壳色性状都具有明显的重排效应,壳白长牡蛎品系其选育需要针对不同的养殖环境培育不同适应性的选育家系。综合育种值排名前20的个体其家系来源比例表明,家系G1和G21对于乳山海域表现出特殊的适应性,而家系G4、G22和G5对荣成海域环境具有特适性,家系G2则对两个环境具有普适性。研究为壳白长牡蛎品系的良种选育提供了重要的参考依据。     

Estimating the heritability for growth-related traits in the pearl oyster,Pinctada fucata martensii (Dunker)

Twelve paternal half-sib families (or 36 full-sib families) of the pearl oyster, Pinctada fucata martensii (Dunker), were produced according to the requirements of hierarchical genetic mating design. A total of 4320 individuals, aged 15 months, were measured for seven growth-related traits. Predicated upon the additive-dominance genetic analysis model, varying genetic variance components and then heritabilities of the growth-related traits of interest were estimated using analysis of variance. Results showed that seven growth-related traits had larger additive genetic variances (P<0.05); the dominance genetic variance of shell weight (SW) was smaller (P>0.05), the dominance genetic variances of other six traits were all larger (P<0.05). Narrow- and broad-sense heritabilities for the seven traits were, respectively, 0.64 ± 0.10 and 0.78 ± 0.12 for shell length, 0.49 ± 0.06 and 0.63 ± 0.09 for shell height, 0.38 ± 0.14 and 0.54 ± 0.16 for shell breadth, 0.41 ± 0.17 and 0.56 ± 0.11 for hinge length, 0.53 ± 0.11 and 0.68 ± 0.08 for body weight, 0.35 ± 0.07 and 0.55 ± 0.08 for tissue weight and 0.67 ± 0.10 and 0.75 ± 0.16 for SW. All heritability estimates were statistically significant (P<0.05). According to these results, the mass selection procedure is suggested for the breeding of P. martensii.     

Estimates of heritability and genetic correlations for growth‐related traits in the tiger prawn Penaeus monodon

Heritabilities and genetic and phenotypic correlations were estimated for body weight (BW), body length (BL), carapace length (CL), carapace width (CW) and carapace height (CH) of the Penaeus monodon from measurements on progeny at 6 months of age resulting from a nested mating design. Estimation was made on data from 41 full‐sib and 10 half‐sib families. The estimated heritabilities for these traits were 0.13–0.24, and the significantly phenotypic and genetic correlations were 0.81–0.94 and 0.88–1.00, respectively. The present results indicate that genetic improvement through selective breeding conducted on P. monodon might be feasible and would be an efficient method to obtain a positive response to selection.     

Are strain genetic effect and heterosis expression altered with culture system and rearing environment in the Portuguese oyster (Crassostrea angulata)?

Re‐ranking of strain additive genetic effect and heterosis expression in different culture environments is not well understood in shellfish species. This study was carried out to estimate the strain additive genetic and heterotic effects on harvest body weight in a 3 × 3 complete diallel cross involving three populations of Portuguese oyster (Crassostrea angulata), two populations from Cat Ba (RIA1) and Nam Dinh, northern Vietnam, and an imported stock from China. Strip spawning was applied to produce full‐ and half‐sib families simultaneously of nine cross‐combinations for performance testing at two locations (Cat Ba Island, Hai Phong city and Van Don Island, Quang Ninh province, Vietnam) using two culture systems (‘single seed’ and ‘cultch set’). A total of 7269 individual oysters were examined over a grow‐out period of 270 days. A linear mixed model was used to estimate strain additive genetic and heterotic effects for body weight at harvest. Ranking of strains based on their additive genetic effects did not differ between the locations and culture systems. The non‐additive genetic (heterotic) effects were low and not significantly different from zero for the traits studied among culture systems. There were no significant differences in levels of heterosis for harvest body weight between the two test environments or culture systems. Based on the non‐significant heterotic and large additive genetic effects among strains used for this study, a future breeding programme in this species (C. angulata), setting aside issues of diversity, could be based simply on choice among different strains based on their performance.     

Estimation of genetic parameters for early growth trait of turbot (Scophthalmus maximus L.) using molecular relatedness

In this study, a data set of 1140 progenies from 19 families that were composed of 13 half‐sib and six full‐sib families was used in the estimation of the genetic parameters of weight for juvenile turbot at 100 days post hatch. Sixty progenies were randomly selected from each family, and 20 of these individuals were genotyped at 12 microsatellite loci. Two sorts of relatedness, including pedigree relatedness (PR) from a complete pedigree and average molecular relatedness (AMR) from molecular markers, were compared to explore the feasibility of AMR in selective breeding when only parents were known. Two different animal models, Model 1 and Model 2 which were without and with the addition of maternal and common environmental effects, respectively, were used to estimate the variance component and breeding values for weight using restricted maximum likelihood (REML) method. Thereafter, cross‐validation was applied to investigate predictive ability of model and accuracy of breeding values. Pearson correlation analysis showed that AMR was highly correlated (0.91) with PR. According to conditional Akaike Information Criterion, the best models for AMR and PR methods were Model 2 and Model 1 respectively. Heritability estimates from AMR and PR methods based on their best models were 0.19 (±0.06) and 0.66 (±0.17). Cross‐validation showed that AMR was comparable to PR in terms of predictive abilities and accuracies of breeding values based on the same model. This study therefore suggests that the AMR can be applied as alternative of PR when only parents were known.     

A novel breeding design to produce genetically protected homogenous fish populations for on‐growing

We present a novel KING breeding design to produce genetically protected homogeneous fish material for commercial producers from a breeding nucleus. KING F2‐production population is established from the nucleus, first through full‐sib mating within two unrelated high‐quality families to produce inbred F1‐progeny and then resolving the inbreeding in F2 through mating of the unrelated F1‐individuals. Owing to a small number of founders and the inbred F1‐parents, the remaining additive genetic variance is 37.5% of the original. This restricts the use of F2‐progeny to establish new breeding programmes, thereby protecting the genetic material of the nucleus. The theoretical calculations show that a concomitant decrease of phenotypic variance is possible. However, the reduction is considerable only for traits with high heritability (h² > 0.50). The method was tested with rainbow trout. The results revealed that the mean body weight of the KING‐progeny was similar, but, surprisingly, phenotypic variation (especially due to residual variance) was higher compared with either their outbred control group or the nucleus breeding population. Although further evaluation of this breeding design is needed, the results suggest that while genetic protection is achieved, the efficiency of the method to reduce phenotypic variation is limited for economically important traits with low‐to‐moderate heritability.     

Genetic parameters of growth and resistance to Polydora ciliata in the pearl oyster Pinctada fucata

The pearl oyster Pinctada fucata is an economically important oyster cultured for pearl production in China. In this study, we used an animal model to estimate the variance components and genetic parameters of growth traits and a sire‐dam threshold model to estimate the resistance of P. fucata to polychaete Polydora ciliata. The results showed that the heritability estimates were 0.3850 ± 0.1421 for shell width, 0.3478 ± 0.1292 for body weight and 0.3387 ± 0.1316 for resistance to P. ciliata, and significantly different from zero (P < 0.05). Both phenotypic and genetic correlations were high between shell width and body weight (0.935 and 0.883 respectively) (P < 0.01), but low between resistance and shell width (0.260 and 0.233 respectively)/body weight (0.152 and 0.083 respectively) (r < 0.30; P > 0.05). The higher heritabilities for shell width, body weight and resistance traits suggested that the three traits can be genetically improved through selective breeding, but the low correlation between growth and resistance traits suggested that the resistance trait can be effectively improved through index selection or through hybridization of two specialized lines, i.e. a high‐resistant line and a high growth line.     

Genetic variation in the growth traits of straight-bred and crossbred black bream (Acanthopagrus butcheri Munro) at 90 days of age

The current slow growth rate of black bream (Acanthopagrus butcheri Munro) impedes their widespread commercial aquaculture across inland southern Australia. We report initial estimates of genetic variation for growth traits at 90 days of age from a diallel cross using two black bream populations. Standard length, total length, and wet weight varied significantly among lines and among half‐sib groups within lines. Differences among half‐sib groups explained 6.8% of the total variance in standard length, 8.3% in total length, and 7.1% in wet weight, giving estimated heritabilities over all lines of 0.27±0.11 for standard length, 0.33±0.13 for total length, and 0.28±0.12 for wet weight. There was no evidence for heterosis in any traits when straight‐bred and crossbred lines were compared. There were high phenotypic (r_P=0.95–0.98) and genetic (r_G=0.63–0.69) correlations among all growth traits.     

Selective breeding can increase resistance of Atlantic salmon to furunculosis, infectious salmon anaemia and infectious pancreatic necrosis

We reasoned that by challenging large numbers of Atlantic salmon families with the causative agents of furunculosis, infectious salmon anaemia (ISA) and infectious pancreatic necrosis (IPN), we could show unequivocally that resistance to these diseases expresses moderate‐to‐high levels of additive genetic variation, and that the resistances are weakly correlated genetically. We tested this reasoning by challenging Atlantic salmon from 920 (approximately) full‐sib families with the causative agents of furunculosis and ISA, and fish from 265 of these families with the causative agent of IPN. Additive genetic variation and genetic correlations were estimated by fitting a threshold liability model to the resistances assessed as binary traits. Resistance to furunculosis, ISA and IPN was moderate –to highly heritable. The marginal posterior means for heritability on the underlying liability scale were 0.37 for resistance to ISA, and 0.55 and 0.62 for resistance to IPN and furunculosis. Genetic correlations between the resistances were weak (?0.11 to 0.07). These levels of additive genetic variation indicate that resistance to furunculosis, ISA and IPN will respond to selection. The weak genetic correlations indicate that it should be relatively easy to improve resistance to the diseases simultaneously. We believe that there is now strong evidence that selectively breeding Atlantic salmon for resistance can be highly successful.