Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Estimation of genetic parameters and genotype-by-environment interactions related to acute ammonia stress in Pacific white shrimp (Litopenaeus vannamei) juveniles at two different salinity levels

  • Xia Lu,

    Affiliations Key Laboratory of Sustainable Utilization of Marine Fisheries Resources, Ministry of Agriculture, Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Qingdao, China, Laboratory for Marine Fisheries Science and Food Production Processes, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

  • Sheng Luan,

    Affiliations Key Laboratory of Sustainable Utilization of Marine Fisheries Resources, Ministry of Agriculture, Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Qingdao, China, Laboratory for Marine Fisheries Science and Food Production Processes, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

  • Baoxiang Cao,

    Affiliations Key Laboratory of Sustainable Utilization of Marine Fisheries Resources, Ministry of Agriculture, Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Qingdao, China, Laboratory for Marine Fisheries Science and Food Production Processes, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

  • Xianhong Meng,

    Affiliations Key Laboratory of Sustainable Utilization of Marine Fisheries Resources, Ministry of Agriculture, Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Qingdao, China, Laboratory for Marine Fisheries Science and Food Production Processes, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

  • Juan Sui,

    Affiliations Key Laboratory of Sustainable Utilization of Marine Fisheries Resources, Ministry of Agriculture, Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Qingdao, China, Laboratory for Marine Fisheries Science and Food Production Processes, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

  • Ping Dai,

    Affiliations Key Laboratory of Sustainable Utilization of Marine Fisheries Resources, Ministry of Agriculture, Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Qingdao, China, Laboratory for Marine Fisheries Science and Food Production Processes, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

  • Kun Luo,

    Affiliations Key Laboratory of Sustainable Utilization of Marine Fisheries Resources, Ministry of Agriculture, Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Qingdao, China, Laboratory for Marine Fisheries Science and Food Production Processes, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

  • Xiaoli Shi,

    Affiliations Key Laboratory of Sustainable Utilization of Marine Fisheries Resources, Ministry of Agriculture, Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Qingdao, China, Laboratory for Marine Fisheries Science and Food Production Processes, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

  • Dengchun Hao,

    Affiliations Key Laboratory of Sustainable Utilization of Marine Fisheries Resources, Ministry of Agriculture, Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Qingdao, China, Laboratory for Marine Fisheries Science and Food Production Processes, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China, College of Fisheries and Life Science, Shanghai Ocean University, Shanghai, China

  • Guomin Han,

    Affiliation College of Ocean, Agricultural University of Hebei, University, Huanghua, China

  • Jie Kong

    kongjie@ysfri.ac.cn

    Affiliations Key Laboratory of Sustainable Utilization of Marine Fisheries Resources, Ministry of Agriculture, Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Qingdao, China, Laboratory for Marine Fisheries Science and Food Production Processes, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

Estimation of genetic parameters and genotype-by-environment interactions related to acute ammonia stress in Pacific white shrimp (Litopenaeus vannamei) juveniles at two different salinity levels

  • Xia Lu, 
  • Sheng Luan, 
  • Baoxiang Cao, 
  • Xianhong Meng, 
  • Juan Sui, 
  • Ping Dai, 
  • Kun Luo, 
  • Xiaoli Shi, 
  • Dengchun Hao, 
  • Guomin Han
PLOS
x

Abstract

Regarding the practical farming of Litopenaeus vannamei, the deterioration of water quality from intensive culture systems and environmental pollution is a common but troublesome problem in the cultivation of this species. The toxicities that result from deteriorating water quality, such as that from ammonia stress, have lethal effects on juvenile shrimp and can increase their susceptibility to pathogens. The toxicity of ammonia plays an important role in the frequently high mortality during the early stage on shrimp farms. However, little information is available regarding the genetic parameters of the ammonia tolerance of juveniles in the early stage, but this information is necessary to understand the potential for the genetic improvement of this trait. Considering the euryhalinity of L. vannamei and the fact that low salinity can increase the toxicity of ammonia stress, we estimated the heritability of ammonia tolerance in juveniles in 30‰ (normal) and 5‰ (low) salinity in this study using the survival time (ST) at individual level and the survival status at the half-lethal time (SS50) at the family level. In the normal and low salinity conditions and for the merged data, the heritability estimates of the ST (0.784±0.070, 0.575±0.068, and 0.517±0.058, respectively) and SS50 (0.402±0.061, 0.216±0.050, and 0.264±0.050, respectively) were all significantly greater than zero, which indicates that the ammonia-tolerance of shrimp can be greatly improved. So it might provide an alternative method to reduce mortality, help to enhance resistance to pathogens and reduce the occurrence of infectious diseases. The significant positive genetic correlation between ST and body length suggested that ammonia is more toxic to shrimp in the early stage. The medium-strength genetic correlations of the ST and SS50 between the two environments (0.394±0.097 and 0.377±0.098, respectively) indicate a strong genotype-by-environment (G×E) interaction for ammonia tolerance between the different salinity levels. Therefore, salinity-specific breeding programs for ammonia tolerance in shrimp should be purposefully implemented.

Introduction

The Pacific white shrimp Litopenaeus vannamei is an important worldwide aquaculture species [1, 2]. Due to the advantages of euryhalinity, rapid growth, and a high survival rate at high density, L. vannamei is becoming the most popular cultured species in the world and generally uses semi-intensive and intensive culture modes [3]. However, these culture practices often lead to the degradation of the water quality due to uneaten food and the waste products of the animals. Additionally, the vagaries of climates and the serious pollution of aquaculture environments jointly cause the quality of the culture water to decrease. The toxic factors associated with deteriorating water quality exhibit very strong correlations with the high mortality of L. vannamei in farms [46].

High concentrations of ammonia are the most common and greatest toxic factor in the deteriorating water quality, which does serious harm to crustaceans, mollusks and fish [4, 79]. Much research has been performed to detect the detrimental effects of ammonia to shrimp, and this research has revealed that the accumulation of ammonia in pond water reduces growth, damages the hepatopancreas and gills, increases oxygen consumption, decreases the osmoregulatory capacity, reduces the ability of the hemolymph to transport oxygen, affects the molting frequency, and even causes high mortality in shrimp [1018]. More important than all of the above, increased ammonia in the water can suppress the immune defense system of shrimp and increase their sensitivity to pathogens [1921]. We previously performed a comparative transcriptome analysis to understand the molecular mechanisms of the detrimental effects of ammonia stress in L. vannamei, and we found that the majority of the key genes involved in the response to ammonia stress are potentially involved in immune defense function [22].

Because the aquaculture environment is becoming worse, and ammonia stress causes grievous harm to shrimp aquaculture, it is necessary to seek possibilities for culturing shrimp that efficiently tolerate ammonia stress, which might represent alternative methods for reducing mortality and infectious diseases. Selective breeding programs are an effective method to gain genetic improvements and have been utilized to improve performance in terms of growth, survival, reproduction, resistances to viruses and adverse environmental stress in shrimp [2327]. Knowledge about the heritability of ammonia tolerance in shrimp is a prerequisite to understanding the potential for the genetic improvement of this trait. Previous studies have revealed that ammonia stress is more toxic to the earlier ontogenetic development of aquaculture organisms [2830], and it has been widely found to exert lethal effects on penaeid shrimp juveniles [3137]. The shrimp in the early juvenile stage are critical for successful breeding because high rates of mortality often occur during this period. We have performed an investigation to estimate the heritability of ammonia resistance in shrimp at the stage with an average body weight of 3.3 g in the normal salinity (30‰) [38], but there are no reports regarding the genetic parameters of ammonia tolerance in juvenile shrimp in the early stage. Additionally, the phenotype of a quantitative trait is determined by genetic and environmental resources and their interactions [39]; thus, genotype-by-environment (G×E) interactions will play an important role in genetic improvement [40]. L. vannamei is cultured at different salinity levels due to its euryhalinity. However, many studies have revealed that low salinity can increase the toxicity of ammonia to the shrimp [34, 35, 4143], and we have also verified that ammonia stress can influence the pathway that is involved in osmoregulation [22]. If the G×E interaction for ammonia tolerance is significant at different salinity levels, the selection response for this trait will vary across different salinity levels [44]. Based on the above reasoning, it is necessary to understand the heritability of ammonia tolerance in the early stage and the G×E interactions at different salinity levels in L. vannamei.

Considering that the majority of the production of L. vannamei occurs in seawater with normal salinity and freshwater with low salinity, in the present study, we estimated the heritability of ammonia tolerance in L. vannamei juveniles in the early stage (average body weight, 0.5 g) at two different salinity levels (30‰ and 5‰) and estimated the genetic correlations between body length and ammonia tolerance within the two environments to verify the relationship between ammonia tolerance and ontogenetic development. Additionally, in the present study, we also detected the G×E interaction between the two salinity levels. Our goal was to understand the potential for the genetic improvement of ammonia tolerance in L. vannamei in the early stage in normal and low salinity because such improvement might represent an alternative method for reducing the mortality of L. vannamei and will also be helpful for enhancing resistance to pathogen and reducing infectious diseases.

Materials and methods

Origin of the base population and selection procedure

The program was conducted at a Hebei Xinhai Aquatic Biological Technology Co., Ltd. Facility located in the Huanghua (longitude 111.3366, latitude 38.37738), Hebei province, China. The founders were collected from seven improved commercial strains that were introduced from different companies in the United States and Singapore. The virus-free individuals from the seven strains were tagged using numbered rings that were placed on one ocular peduncle and used to produce the base population (G0) via an incomplete diallel cross experiment. The G0 was constructed with 130 full-sib families using artificial insemination with 114 males and 108 females in 2011 and included 69 half-sib families. Individual estimated breeding values (EBV) for body weight and the family EBV for survival rate from the test were weighted into the selection index and used to rank the breeding candidates as described by Luan et al. [45]. The families with low selection indices (<100) were eliminated from the breeding program, and the males and females with high selection indices (>100) from the remaining families were selected to produce the next generation. During the implementation process of the present program, the selection population was closed, and the generations were discrete. The pedigree of the selection population is clear and forms a relationship matrix. The shrimp production scheme was provided in our previous study [38].

The present experimental shrimp were from 91 full-sib families (52 half-sib families) of the fifth generation (G5). The half-sib families were produced by two females mating with one male or two males mating with one female, and all of the families were produced by artificial insemination. The processes of family construction, hatching, and larvae rearing were described in detail in our previous study [46]. At the post-larvae stage of approximately 45 days, approximately 400 post-larvae were randomly selected from each family and equally transferred into tow net cages (0.5 m3) that were separately fixed in two large ponds (60 m3) for separate rearing at the same density for each family. The seawater in one of the ponds was diluted with filtered fresh water from 30‰ to 5‰, which lasted for approximately three weeks. However, the seawater in the other pond was maintained at 30‰. Aside from the salinity, the other management conditions were maintained identically between the two ponds.

Challenge with high concentration ammonia

To acquire the individual survival times for the estimations of the genetic parameters, we performed a pre-experiment to obtain the proper concentrations of ammonia in the 30‰ (normal) and 5‰ (low) salinity conditions. Based on the results of our pre-experiment, the concentrations of ammonia required to kill all of the shrimp after 72-h challenges were 32 mg/L and 18 mg/L in the normal and low salinity conditions, respectively. The ammonia solution was prepared using NH4Cl (Aldrich, Milwaukee, WI, USA).

The juveniles were too small (the average body weight of 0.50 g and average length of 3.73 cm) that they could not be tagged with visible implant elastomer (VIE) when the experiment began. Therefore, forty individuals were randomly selected from each net cage (each family) and transferred to separate 100 L tanks filled with 35 L 30‰ or 5‰ salinity water for ammonia stress tests under the same conditions of density and environmental management. After a period of three days for a temporary rearing, the ammonia acute stress experiment was initiated. During the experiment, the shrimp were given feedstuff twice per day, and the management environments were maintained identically between the tanks. The dissolved oxygen level was no less than 6 mg L-1, the pH ranged from 8.00 to 8.06, and the temperature was approximately 27±0.5℃. The tanks were cleaned daily by suction to remove feces, and the water that was removed by suction was replaced with clean water with the same concentration of ammonia. The dead shrimp were collected and recorded every hour during the experiment. When the shrimp sank, stopped movement and lost responsiveness to external stimulation, they were defined as dead and removed from the tanks with a scoop net. The family ID and death time of each dead shrimp were recorded to quantify the individual survival times (STs) and survival statuses (1 = alive, 0 = dead) at the half lethal time (SS50). Because the animals were too small that body weight measures would introduce large measurement errors due to the different degrees of excess water on the shrimp, we selected body length (BL) as an index of growth. Therefore, the individual BLs were also recorded when the shrimp died, which enabled us to investigate the relationship between ammonia tolerance and growth stage.

Statistical analysis

Variance components and heritability estimates.

An analysis of the descriptive statistics was conducted using the MEANS procedure [47]. The complete pedigree of this breeding program from G0 to G5 was used in the following analysis to account for the genetic relationships among the individuals. The common environmental effect was omitted in this analysis due to the separate rearing of the families during the experiment. The ammonia tolerance was defined by the ST (individual level) and the SS50 (family level) in the present study. The variance components and heritabilities were calculated for ST and SS50 using the data from the normal and low salinity conditions and the merged data from the two conditions. Therefore, these variables are denoted as the STH/SS50H, STL/SS50L, and STM/SS50M for the normal salinity, low salinity, and merged data, respectively.

The individual animal model was used to estimate the heritability of the ST in the shrimp that were challenged with the high concentration of ammonia. Body length exhibited a linear relationship with ST (P < 0.01) and was fit as a covariate in the model. The fitted model was as follows: (Model 1) where yijk is the observed ST of the kth individual, μ is the overall mean, Salti is the fixed effect of the ith salt level (two levels) and it was not contained when estimated for the separate environments, BLj is the covariate of the jth body length, b is the regression coefficient, ak is the random additive genetic effect of the kth individual, and eijk is the random residual effect associated with observation ijk. The phenotypic variance () was taken as the sum of all of the variance components as follows: . The heritability (h2) was calculated as the ratio between the genetic variance and the total phenotypic variance ().

A standard threshold (probit) and a sire–dam model were used in this study to estimate the heritability of the SS50. The model was written in ASReml as follows [48]: (Model 2) where yijkl is the survival status (1 = alive, 0 = dead) of the jth shrimp, λijkl is the underlying liability of yijkl, which was assumed to be a cumulative standard normal distribution, μ is the overall mean, BLj is the covariate of the jth body length, b is the regression coefficient, and Sirek and Daml are the additive genetic effects of the kth sire and the lth dam, Sire or Dam ~(0, ) (), where A is the additive genetic relationship matrix among all shrimp. The residual variance of λ was assumed to be 1. The phenotypic variance was the sum of and : . Heritability was computed as the ratio between and (). The heritability would be overestimated in this model, so the estimates for SS50 then were adjusted according to Robertson and Lerner [49].

The estimation of the heritability of BL was performed using the ASReml package [48]. The individual animal model was used to estimate the genetic parameters for the BLs of the shrimps that were challenged with high concentrations ammonia. The age at the end of the experiment exhibited a linear relationship with body length and was fit as a covariate in the model. All of the fixed effects and the covariates were statistically significant (P < 0.01). The fitted model was as follows: (Model 3) where yijk is the observed BL of the kth individual, μ is the overall mean, Salti is the fixed effect of the ith salt level (two levels), and it was not contained when estimated for the separate environments, b is the regression coefficient, Agej is the covariate of the jth individual’s age when the experiment ended, ak is the random additive genetic effect of the kth individual, and eijk is the random residual effect associated with observation ijk. The phenotypic variance () was the sum of all variance components: . The heritability was calculated as the ratio between the animals’ genetic variance and the total phenotypic variance as follows: ().

The z-scores were used to test whether the heritability estimates were significantly different from each other and whether the heritability estimates were significantly different from zero [50]. where xi and xj are the heritability estimates from different models and different salinity levels, and and are their respective standard errors. Both xj and were set to be zero when testing whether an estimate was significantly different from zero. The resulting z-score was then tested against a large-sample normal distribution.

Genotype by environment interaction estimate.

The homologous ST, SS50, and BL in the different environments were considered to be different traits; thus, the genotype by environment (G×E) interactions were estimated based on the genetic correlations of these traits between environments. The phenotypic (rp) and genetic (rg) correlations of the ST and BL values between the two salinity environments were estimated using the bivariate animal model described above. Because one animal can only be tested in one environment, there were no environmental (residual) correlations between the homologous traits, and the environmental covariance was set to be zero in the bivariate analysis [51]. Because of the convergence problem, the rg between the two salinity environments for the SS50 was calculated with the estimated breeding values. A G×E interaction was measured as the difference between the genetic correlation and 1. Thus, genetic correlations closer to 1 indicated smaller the G×E interactions. The genetic correlations between the environments were used to measure the degree of re-ranking due to G×E interactions. Additionally, the rp and rg values between the STs and BLs within environments were estimated using the bivariate animal model, which is the extension of Model 1 and Model 3, with the ASReml package [48]. The statistical significances of the genetic correlations within environments and between environments were assessed using the confidence intervals (CIs). The CIs of the genetic correlations were calculated using rg, the standard errors (SEs) and Z0.025, which is 1.96 (i.e., [rg−1.96×SE, rg+1.96×SE]). If 1 or 0 was contained within the CI, the difference between the genetic correlation and 1 or 0 was not significant.

Results

Descriptive statistics

A total of 7221 individual records, including 3624 records from the normal salinity condition and 3597 records from low salinity condition, were obtained and analyzed in the present study. The number of final testing individuals was less than the total number of starting individuals (7280), which might have been due to cannibalism during the experimental process. The numbers of samples, simple means, minima, maxima, standard deviations and the coefficients of variation for the survival rate of each family at the half-lethal time (SS50), ST, and BL are given in Table 1. The 25th percentiles, median percentiles, 75th percentiles, minima, maxima, and outliers of the SS50 and average ST values of each family are displayed in Fig 1. The results revealed the SS50 values under acute ammonia stress in the normal and low salinity conditions varied substantially between the families, but the variance was higher following exposure to acute ammonia stress in the normal salinity condition (Table 1; Fig 1A). Additionally, the STs under acute ammonia stress in the normal and low salinity conditions also varied substantially between the families and the overall individuals, but the variance was higher when analyzed at the individual level than the family level as indicated by the higher standard deviation (SD) and coefficient of variation (CV). The CVs of the STs were 27.38% and 29.27% among the 91 families in the two environments, but these values were 46.87% and 54.48% among the individuals in the two environments (Table 1). Although the concentration of ammonia (32 mg/L) was higher in the normal salinity condition than in the low salinity (18 mg/L) condition, the average ST under acute ammonia stress in the normal salinity condition (36.64 h) was substantially greater than that in the low salinity condition (24.80 h), and the variance was higher in the normal salinity condition (Table 1; Fig 1B). Regarding the BL, it also exhibited higher variance when analyzed at the individual level than at the family level, but there was no significant difference between the two environments (Table 1).

thumbnail
Fig 1.

(a) Boxplot of the survival rates of the families at the half-lethal time. (b) Boxplot of the average survival times of the families. The 25th (upper line), median (inside line) and 75th (bottom line) percentiles of the families are plotted as boxes. The minima, maxima, and the observed values are shown as -, -, and ○, respectively.

https://doi.org/10.1371/journal.pone.0173835.g001

thumbnail
Table 1. Numbers of samples/families (N) and the means, minima, maxima, standard deviations, and coefficients variation of the ST, SR50, and BL values.

https://doi.org/10.1371/journal.pone.0173835.t001

The cumulative mortality and the average survival time of each family under acute ammonia stress in the normal and low salinity conditions are displayed in Fig 2 and Fig 3, respectively. The cumulative mortalities of the families under acute ammonia stress in the low salinity condition were substantially greater than those in the normal salinity condition at each sampling point (Fig 2). All of the individuals in the low salinity condition were dead after 69 h of exposure to acute ammonia stress, but at this time, 115 individuals (3.5%) in the normal salinity condition were still alive. It showed that the average survival time exhibited large variation between the families within the same environment and between the two environments, but the average survival time of the majority of the families under acute ammonia stress in the low salinity condition was substantially lower than that of the families in the normal salinity condition (Fig 3).

thumbnail
Fig 2. Cumulative mortality of L. vannamei juveniles during the acute ammonia stress in normal and low salinity conditions.

https://doi.org/10.1371/journal.pone.0173835.g002

thumbnail
Fig 3. The average survival time for each family of L. vannamei juveniles during acute ammonia stress in normal and low salinity conditions.

https://doi.org/10.1371/journal.pone.0173835.g003

Variance components and heritability estimations

Estimates of the variance components and heritability of ST, SS50, and BL created with different models are provided in Table 2. All of the heritability estimates were significantly greater than zero (P < 0.01). According to the classification reported by Cardellino and Rovira [52] (i.e., low, 0.05–0.19; medium, 0.20–0.44; high, 0.45–0.64; and very high, >0.65), the heritability estimate of STH (0.784±0.070) was very high, the heritabilities of STL (0.575±0.068) and STM (0.517±0.058) were high, and the heritabilities of SS50H (0.402±0.061), SS50L (0.216±0.050) and SS50M (0.264±0.050) were medium. The heritability estimate of STH was significantly greater than those of STL and STM (P<0.05), but there was no significant difference between STL and STM (P>0.05). The heritability estimates of ST were all significantly greater than that of SS50 (P < 0.01). The heritability of SS50H was significantly greater than that of SS50L (P < 0.05), but there was no significant difference between SS50H and SS50M (P>0.05), and there was also no significant difference between SS50L and SS50M (P>0.05). The heritability estimates of BLH (0.346±0.052), BLL (0.386±0.054), and BLM (0.291±0.042) were all medium based on the above classification, and the values for BLH and BLL were higher than that of BLM but were not significantly different from each other (P > 0.05).

thumbnail
Table 2. Variance components and heritabilities of for ST, SS50 and BL.

https://doi.org/10.1371/journal.pone.0173835.t002

Genetic corrections within and between environments

To understand the effects of the G×E interactions on the three traits, particularly the trait of ammonia tolerance, genetic correlations were estimated between the two environments, and the results are displayed in Table 2. The phenotypic and genetic correlations among the three traits between the two environments were all positive, and their genetic correlations were higher than their respective phenotypic correlations. The estimated genetic correlation for BL between the two environments was higher (0.535±0.096) than those of ST (0.394±0.097) and SS50 (0.377±0.098) (Table 2). The genetic correlations for ST, SS50 and BL between the two environments were significantly different from 0 and 1 based on the CIs (ST, 0.20–0.58; SS50, 0.19–0.57; and BL, 0.35–0.72). These results indicate a strong re-ranking effect for ammonia tolerance and growth between the two environments.

The phenotypic and genetic correlations within the two environments for ST and BL are given in Table 3. The phenotypic and genetic correlations between the two traits within the two environments were also all positive, and their genetic correlations were greater than their respective phenotypic correlations. The phenotypic and genetic correlations between ST and BL were higher in the normal salinity condition (rp = 0.416±0.017; rg = 0.779±0.037) than in the low salinity condition (rp = 0.298±0.021; rg = 0.568±0.048).

thumbnail
Table 3. Correlation analysis based on phenotypic and breeding values between ST and BL.

https://doi.org/10.1371/journal.pone.0173835.t003

Discussion

In this study, we first reported the heritability of ammonia tolerance (adversity-stress adaptability) for juveniles in the early stage in normal salinity (30‰) and low salinity (5‰) conditions and the G×E interaction between the two environments to explore the possibility of reducing the mortality of shrimp by selection. Many studies have reported that ammonia stress is more toxic to shrimp in early stages [2830, 5355], and this finding was corroborated in the present study. In this experiment, approximately 98.6% of the individuals with an average body weight 0.5 g (average body length of 3.7 cm were dead following exposure to 32 mg/L ammonia for 72 h (pH = 8.05, temperature = 26.5±0.5℃, salinity = 30‰). However, all of the L. vannamei individuals with the an average body weight of 3.3 g were dead following exposure to 63 mg/L ammonia for 218 h (pH = 8.0, temperature = 25℃, salinity = 30‰) [38]. Sun et al. [56] reported that the half-lethal concentration of ammonia is 32.44 mg/L at 72 h (pH = 8.15, temperature = 27℃, salinity = 20‰) for L. vannamei with an average body length of 5.0 cm. Although high temperatures and low salinities can increase the toxicity of ammonia stress, the shorter survival time at lower concentrations of ammonia in our study also suggests that younger shrimp are more sensitive to ammonia toxicity. Additionally, we also observed that the lower salinity increased the sensitivity of shrimp to the toxic effects of ammonia. The lethal concentration of ammonia at 72 h in the normal salinity condition (30‰) was 32 mg/L, which is significantly higher than the 18 mg/L value observed in the low salinity condition (5‰). The average survival time of all the individuals under acute ammonia stress in the normal salinity condition was 36.61 h, which is substantially greater than the 24.80 h observed in the low salinity condition (Table 1), and the average survival time of majority of the families under acute ammonia stress in the normal salinity condition was significantly greater than that in the low salinity condition (Fig 3). The cumulative mortalities of the families in the normal salinity condition were significantly lower than those in the low salinity condition at each sampling point (Fig 2). The above results indicated that low salinity increased the toxicity of ammonia stress to shrimp, which is consistent with the results of previous studies [34, 35, 4143]. Our results suggest that the selection of ammonia-tolerant shrimp should be performed in the early stage and in low salinity water.

A high heritability of a trait suggests that rapid and highly accurate genetic improvement can be obtained by selection. Although many studies have been performed to detect ammonia toxicity for aquatic organisms [1018], very few reports are available regarding the genetic parameters related to ammonia tolerance in aquatic species; thus, there is little information against which to compare our results. Currently, the only previous report of the heritability of ammonia resistance in shrimp with an average body weight of 3.3 g that are subjected to ammonia stress in normal salinity conditions came from our group [38]. In that report, the heritabilities of ST and SS50 were 0.154±0.045 and 0.148±0.040, respectively, and these values were not significantly different from each other. In the present study, we estimated the heritability of ammonia tolerance of shrimp with an average body weight of 0.5 g under ammonia stress in normal and low salinity conditions. The heritability estimates of ST (STH = 0.784±0.070; STL = 0.575±0.068; STL = 0.517±0.058) and SS50 (SS50H = 0.402±0.061; SS50L = 0.216±0.050; SS50M = 0.264±0.050) under ammonia stress at the two salinity levels were all significantly greater than that reported in the previous study, which indicates that improving the ammonia tolerance of shrimp via selection performed at the early stage could be quite advantageous. Although we did our best to maintain the same environment between the tanks in the present study, the absent of common environmental effects because of the juveniles without communal rearing during the experiment may have contributed to the overestimation of the heritability estimates of ammonia tolerance, because the exclusion of common environmental effects can inflate heritability estimates [5759]. In our previous study, an ammonia challenge experiment was performed in four tanks, and significant tank effects were detected, so the fixed effect of the tank in the models might have reduced the additive genetic variance to a certain extent, but the absence of common environmental effects also might have resulted in overestimations of the heritability estimates [27, 60]. The higher heritability of ammonia tolerance observed in the present study might have been due to cryptic genetic variation (CGV) that emerged due to ammonia stress, which could have increased the additive genetic variance (Va, i.e., the heritable phenotypic variation). Because shrimp in the early stage are more sensitive to the ammonia stress, the CGV is more easily released. Additional information about CGV can be found in the review by Paaby and Rockman that was published in Nature Reviews Genetics [61]. Genomic matrices based on molecular information provide a more precise method for the estimation of heritability and can thus enhance the accuracy of selective breeding, and this approach has been successfully applied in animal husbandry and agriculture [62, 63]. With the development of molecular biological technology, the cost of obtaining genotypic information from individual shrimp (e.g., through sequencing-based or SNP panel-based genotyping technology) will be substantially reduced, and a few molecular genetic/genomic resources will become publically available. At this time, we will use genomic matrices for heritability estimation, which will enable families to be communal reared without tags.

In the present study, the heritability estimates of ST were all significantly higher than those for SS50, which is inconsistent with our previous results that indicated that the heritability estimates of ST and SS50 are basically the same [38]. The significantly lower heritability of SS50 compared with ST might have resulted from the smaller sample size at the family level (91) for SS50 compared to the individual level (above 3500) for ST. Additionally, the heritability estimate of ammonia tolerance at normal salinity was significantly greater than that at low salinity in the present study, which might have resulted from the incomplete release of the additive genetic variance among the individuals and families because the time to death was shorter and more intensive in the low salinity condition; after 20 h of ammonia stress, the cumulative mortality in the low salinity condition was 50%, whereas this value was only 24% in the normal salinity condition (Fig 1). The heritability estimates of ST and SS50 from the merged data from the environments were lower than those in the normal salinity condition, which indicates that salinity plays an important role in the estimation of the heritability of ammonia tolerance. After this experiment, a WSSV infection experiment was performed with the ammonia-sensitive and ammonia-tolerant populations to detect the relationship between ammonia tolerance and WSSV resistance. The results revealed that the ammonia-tolerant population also exhibited a significantly greater resistance to WSSV than the ammonia-sensitive population (this result will be published soon in an article entitled “The investigation for the susceptibility difference of ammonia-tolerant and ammonia-sensitive populations to WSSV infection in Litopenaeus vannamei”). Fortunately, the ammonia-tolerant shrimp can be improved largely due to high heritability, which might provide an alternative method to reduce mortality, aid the enhancement of resistance to pathogens, and reduce infectious diseases.

Growth is a very important trait in selective breeding programs because it is highly correlated with economic return. The estimates of the heritability of growth in L. vannamei have primarily focused on harvest body weight [27, 64], and very little information is available regarding juveniles. Although the absence of the common environmental effects could have resulted in the overestimation of the heritability of BL in the present study, the medium heritabilities (0.346±0.052, 0.386±0.054, and 0.291±0.042) in the two environments still indicated that there is potential for improving the growth of juveniles. The growth performances of the families between the two environments were not fully released due to the short communal rearing period (approximately three weeks), and thus, the heritabilities of growth observed in the two salinity levels in the present study are only preliminary results. To understand the difference in growth performance between normal and low salinity conditions, estimates of the heritability of growth in the two salinity levels is carried out in a separate experiment with a longer testing time (approximately three months) in our program, and the results of such an experiment will be published later. Previous studies have reported negative correlations between body weight and resistance traits (such as TSV resistance and WSSV resistance) in L. vanname [23, 24]. However, a strong positive correlation between body length and ammonia tolerance was detected in the present study, which is consistent with the previous studies [28, 38]. The strong positive correlation between body size and ammonia tolerance suggests that it is necessary to select ammonia-tolerant shrimp during the early stage and that selection high growth will not adversely affect ammonia tolerance performance in shrimp.

G×E interactions refer to genotypes that respond differently in different environments and thus result in different performances [6567]. Genetic correlations across environments that are significantly different from 1 indicate the existence of re-ranking effects, i.e., genotypes that should be ranked differently in different environments [68]. The results from the simulation study of Sae-Lim et al. [69] suggested that more reliable estimates of the G×E interactions can be obtained with population sizes greater than 2000 (i.e., 50–200 families and 10–40 individuals of each family) and heritabilities greater than 0.30. The above conditions are suitable to the situation in our study, and thus, the present estimates of G×E interactions are credible. Generally, selection for a trait in aquaculture is always performed in the main environment; for example, ammonia tolerance would typically be selected for in normal salinity conditions. However, the offspring of the excellent shrimp procured from selection are transferred to different commercial production environments in most cases. Therefore, strong re-ranking effects may occur and result in different responses to ammonia stress at different salinity levels if there is a strong G×E interaction for this trait, and these effects would influence selection accuracy and the total economic gains. In the present study, we detected a strong re-ranking effect (i.e., a low genetic correlation) for ammonia tolerance between the different salinity levels (Table 3), which suggests that the genetic advantage of ammonia tolerance can be significantly influenced by salinity. Additionally, the genetic correlations across environments would decrease along with increases in differences between environments [70, 71]. When a strong G×E interaction exists for a trait, it is better to perform selection in multiple environments, such as the nucleus environment and other production environments [72]. Therefore, salinity-specific breeding programs for ammonia tolerance in shrimp should be purposefully implemented.

Conclusion

Overall in this study, we first reported the heritability of ammonia tolerance in juveniles in the early stage and the G×E interaction between the normal and low salinity to explore the potential for improving ammonia tolerance in shrimp by selection. The results revealed that the heritability of ammonia tolerance was medium to high in L. vannamei juveniles, which suggests that rapid genetic gains in terms of ammonia tolerance could be obtained by increasing the selection intensity in our selective breeding program. However, the heritability of ammonia tolerance in the normal salinity condition was significantly higher than that in the low salinity condition. Additionally, a strong G×E interaction for ammonia tolerance was detected between the normal salinity and low salinity conditions, which suggests that salinity-specific breeding programs for ammonia tolerance in shrimp should be purposefully implemented and that the normal salinity environment is a better choice based on the faster rate of genetic improvement due to higher heritability. Additionally, the significantly and strong positive correction between ammonia tolerance and body size suggests that ammonia-tolerant shrimp should be selected in the early stage.

Supporting information

S1 Table. The heritability of SS50 from probit and the formula for adjusting the heritability.

https://doi.org/10.1371/journal.pone.0173835.s001

(XLSX)

S2 Table. The individual data for the analysis in the present study.

https://doi.org/10.1371/journal.pone.0173835.s002

(XLSX)

Acknowledgments

We thank Yuling Zhang and Zhangwei Kong from Ocean University of China, Guangfeng Qiang from Shanghai Ocean University, and Fanyong Zeng from Nanjing Agricultural University for assisting to complete the ammonia test experiment.

Author Contributions

  1. Conceptualization: XL SL JK.
  2. Data curation: XL SL.
  3. Formal analysis: XL JS.
  4. Funding acquisition: XL JK.
  5. Investigation: XL BXC PD DCH KL GMH XLS.
  6. Methodology: XL SL JS.
  7. Project administration: XL JK.
  8. Resources: BXC SL DCH.
  9. Software: SL JS XL.
  10. Supervision: JK SL XHM.
  11. Validation: XL.
  12. Visualization: XL.
  13. Writing – original draft: XL.
  14. Writing – review & editing: XL.

References

  1. 1. FAO Yearbook. Fishery and Aquaculture Statistics Summary tables. 2015.
  2. 2. Lu X, Luan S, Luo K, Meng XH, Li WJ, Sui J, et al. Genetic analysis of the Pacific white shrimp (Litopenaeus vannamei): heterosis and heritability for harvest body weight. Aquaculture Research 2015; 47: 3365–3375.
  3. 3. Cuzon G, Lawrence A, Gaxiola G, Rosas C, Guillaume J. Nutrition of Litopenaeus vannamei reared in tanks or in ponds. Aquaculture 2004; 235: 513–551.
  4. 4. Colt JE, Armstrong DA. Nitrogen toxicity to crustaceans, fish, and molluscs. In: Allen LJ, Kinney EC, editors. Proceedings of the Bio-Engineering Symposium for Fish Culture. Fish Culture Section, American Fisheries Society. Northeast Society of Conservation Engineers, Bethesda, MD. 1981. p. 34–47.
  5. 5. Foss A, Vollen T, Øiestad V. Growth and oxygen consumption in normal and O2 supersaturated water, and interactive effects of O2 saturation and ammonia on growth in spotted wolfish (Anarhichas minor Olafsen). Aquaculture 2003; 224: 105–116.
  6. 6. Alagappan KM, Deivasigamani D, Somasundaram ST, Kumaran S. Occurrence of Vibrio parahaemolyticus and its specific phages from shrimp ponds in east coast of India. Current Microbiology 2010; 61(4): 235–240. pmid:20140436
  7. 7. Lewis JrWM Morris DP. Toxicity of nitrite to fish: a review. Transactions of the American Fisheries Society 1986; 115: 183–195.
  8. 8. Hurvitz A, Bercovier H, Rijn JV. Effect of ammonia on the survival and the immune response of rainbow trout (Oncorhynchus mykiss, Walbaum) vaccinated against Streptococcus iniae. Fish & Shellfish Immunology 1997; 7: 45–53.
  9. 9. Cheng W, Hsiao IS, Chen JC. Effect of ammonia on the immune response of Taiwan abalone Haliotis diversicolorsupertexta and its susceptibility to Vibrio parahaemolyticus. Fish & Shellfish Immunology 2004; 17:193–202.
  10. 10. Chen JC, Liu PC, Lei SC. Toxicities of ammonia and nitrite to Penaeus monodon adolescents. Aquaculture 1990; 89: 127–137.
  11. 11. Chen JC, Ting YY, Lin JN, Lin MN. Lethal effects of ammonia and nitrite on Penaeus chinensis juveniles. Marine Biology 1990; 107: 427–431.
  12. 12. Chen JC, Lin CY. Effect of ammonia on growth of Penaeus penicillatus juveniles. Comparative Biochemistry and Physiology C Pharmacology Toxicology and Endocrinology 1992; 101: 443–447.
  13. 13. Chen JC, Lin CY. Effects of ammonia on growth and molting of Penaeus monodon juveniles. Comparative Biochemistry and Physiology C Pharmacology Toxicology and Endocrinology 1992; 101: 449–452.
  14. 14. Zin KY, Chu CJ. Acute toxicity of ammonia to Penaeus japonicus Bate juveniles. Aquaculture Research 1991; 22:259–263.
  15. 15. Chen JC, Kou YZ. Effects of ammonia on growth and molting on Penaeus japonicus juveniles. Aquaculture 1992; 104:249–260.
  16. 16. Lin H, Thuet P, Trilles JP, Mounet-Guillaume R, Charmantier G. Effects of ammonia on survival and osmoregulation of various developmental stages of the shrimp Penaeus japonicus. Marine Biology 1993; 117:591–598.
  17. 17. Chien YH. Water quality requirements and management for marine shrimp culture. In: Proceedings of the Special Session on Shrimp Farming (ed. by J. Wyban), World Aquaculture Society, Baton Rouge, LA, USA; 1992. p. 144–152.
  18. 18. Racotta IS, Hernández-Herrera R. Metabolic responses of the white shrimp, Penaeus vannamei, to ambient ammonia. Comparative Biochemistry and Physiology 2000; 125:437–443. pmid:10840218
  19. 19. Qiu DQ, Zhou XJ, Qiu MS. Study on Anti-disease ability of Litopenaeus vannamei and the biological control of vibrio parahaemolyticus bacteriophage under stresses of ammonia nitrogen. Acta Hydrobiologica Sinica 2008; 32(4):455–461.
  20. 20. Liu CH, Chen JC. Effect of ammonia on the immune response of white shrimp Litopenaeus vannamei and its susceptibility to Vibrio alginolyticus. Fish & Shellfish Immunology 2004; 16:321–334.
  21. 21. Cheng W, Hsiao IS, Chen JC. Effect of ammonia on the immune response of Taiwan abalone Haliotis diversicolor supertexta and its susceptibility to Vibrio parahaemolyticus. Fish & Shellfish Immunology 2004; 17:193–202.
  22. 22. Lu X, Kong J, Luan S, Dai P, Meng X, Cao B, et al. Transcriptome Analysis of the Hepatopancreas in the Pacific White Shrimp (Litopenaeus vannamei) under Acute Ammonia Stress. PLoS ONE 2016; 11(10): e0164396. pmid:27760162
  23. 23. Gitterle T, Salte R, Gjerde B, Cock J, Johansen , Salazar R, et al. Genetic (co)variation in resistance to White Spot Syndrome Virus and harvest weight in Penaeus (Litopenaeus) vannamei. Aquaculture 2005; 246: 139–149.
  24. 24. Argue BJ, Arce SM, Lotz JM, Moss SM. Selective breeding of Pacific white shrimp (Litopenaeus vannamei) for growth and resistance to Taura syndrome Virus. Aquaculture 2002; 204: 447–460.
  25. 25. Ibarra AM, Pérez-Rostro CI, José LRamírez, Ortega-Estrada E. Genetics of the resistance to hypoxia in postlarvae and juveniles of the Pacific white shrimp Penaeus (Litopenaeus) vannamei (Boone 1931). Aquaculture Research 2007; 38: 838–846.
  26. 26. Ibarra AM, Thomas R, Famula , Fabiola G. Heritability of vitellogenin in hemolymph, a pre-spawning selectable trait in Penaeus (Litopenaeus) vannamei, has a large genetic correlation with ovary maturity measured as oocytes mean diameter. Aquaculture 2009; 297: 64–69.
  27. 27. Castillo-Juárez H, Casares JCQ, Campos-Montes G, Villela CC, Ortega AM, Montaldo HH. Heritability for body weight at harvest size in the Pacific white shrimp, Penaeus (Litopenaeus) vannamei, from a multi-environment experiment using univariate and multivariate animal models. Aquaculture 2007; 273: 42–49.
  28. 28. Barajas FJM, Villegas RS, Clark GP, Moreno BL. Litopenaeus vannamei (Boone) post-larval survival related to age, temperature, pH and ammonium concentration. Aquaculture Research 2006; 37:492–499.
  29. 29. Zhao JH, Lam TJ, Guo JY. Acute toxicity of ammonia to the early stage-larvae and juveniles of Eriocheir sinensis H. Milne-Edwards, 1853 (Decapoda: Grapsidae) reared in the laboratory. Aquaculture Research 1997; 28:517–525.
  30. 30. Magallόn-Barajas F, Servín R, Portilllo C, García J, Lόpez B. Daily variation in short-term static toxicity of unionized ammonia in Litopenaeus vannamei (Boone) postlarvae. Aquaculture Research 2006; 37:1406–1412.
  31. 31. Cobo MD, Sonnenholzner S, Wille M, Sorgeloos P. Ammonia tolerance of Litopenaeus vannamei (Boone) larvae. Aquac Res 2014; 45: 470–475.
  32. 32. Chen JC, Lei SC. Toxicities of ammonia and nitrite to Penaeus monodon juveniles. Journal of the World Aquaculture Society 1990; 21: 300–306.
  33. 33. Wajsbrot N, Gasith A, Krom MD, Samocha TM. Effect of dissolved oxygen and the molt stage on the acute toxicity of ammonia to juvenile green tiger prawn Penaeus semisulcatus. Environ. Toxicology Chemical 1990; 9: 497–504.
  34. 34. Chen JC, Lin CY. Lethal effects of ammonia and nitrite on Penaeus penicillatus juveniles at two salinity levels. Comparative Biochemistry and Physiology 1991; 100: 466–482.
  35. 35. Chen JC, Lin CY. Lethal effects of ammonia on Penaeus chinensis Osbeck juveniles at different salinity levels. Journal of Experimental Marine Biology and Ecology 1992; 156: 139–148.
  36. 36. Ostrensky A, Wasielesky W Jr. Acute toxicity of ammonia to various life stages of the Sao Paulo shrimp, Penaeus paulensis Perez-Farfante, 1967. Aquaculture 1995; 132: 339–347.
  37. 37. Nan FH, Chen JC. Lethal effect of ammonia to juvenile Metapenaeus ensis. Journal of the Fishery society 1991; 18: 41–46.
  38. 38. Li WJ, Lu X, Luan S, Luo K, Sui J, Kong J. Heritability of body weight and resistance to ammonia in the Pacific white shrimp Litopenaeus vannamei juveniles. Chinese Journal of Oceanology and Limnology 2016; 34:1025.
  39. 39. Falconer DS. Selection in different environments: effects on environmental sensitivity (reaction norm) and on mean performance. Genetic Research 1990; 56:57–70.
  40. 40. Mulder HA, Bijma P. Effects of genotype × environment interaction on genetic gain in breeding programs. Journal of Animal Science 2005; 83:49–61. pmid:15583042
  41. 41. Chen JC, Chen CT, Cheng SY. Nitrogen excretion and changes of hemocyanin, protein and free amino acid levels in the hemolymph of Penaeus monodon exposed to different concentrations of ambient ammonia-N at different salinity levels. Marine Ecology Progress Series 1994; 110:85–94
  42. 42. Bower EC, Bidwell PJ. Ionization of ammonia in seawater: effects of temperature, pH, and salinity. Journal of the Fisheries Research Board of Canada 1978; 35:1012–1016.
  43. 43. Lin YC, Chen JC. Acute toxicity of ammonia on Litopenaeus vannamei Boone juveniles at different salinity levels. Journal of Experimental Marine Biology and Ecology 2001; 259: 109–119. pmid:11325379
  44. 44. Mulder HA, Veerkamp RF, Ducro BJ, van Arendonk JAM, Bijma P. Optimization of dairy cattle breeding programs for different environments with genotype by environment interaction. Journal of Dairy Science 2006; 89:1740–1752. pmid:16606745
  45. 45. Luan S, Wang JY, Yang GL, Luo K, Chen XF, Gao Q, et al. Genetic parameters of survival for six generations in the giant freshwater prawn Macrobrachium rosenbergii. Aquaculture Research 2015; 46:1345–55.
  46. 46. Lu X, Luan S, Luo K, Meng XH, Li WJ, Sui J, et al. Genetic analysis of the Pacific white shrimp (Litopenaeus vannamei): heterosis and heritability for harvest body weight. Aquaculture Research 2015. 47(11): 3365–3375.
  47. 47. SAS Institute Inc. SAS/STAT Software: Version 9.1.3 (TS1M3) for Microsoft Windows. SAS Institute, Cary, NC, USA. 2005.
  48. 48. Gilmour AR, Gogel BJ, Cullis BR, Thompson R. ASReml User Guide Release 3.0. VSN International Ltd, Hemel Hempstead. 2009.
  49. 49. Robertson A, Lerner IM. The Heritability of all-or-none traits: viability of poultry. Genetics 1949; 34: 395–411. pmid:17247323
  50. 50. Nguyen NH, Khaw HL, Ponzoni RW, Hamzah A, Kamaruzzaman N. 2007. Can sexual dimorphism and body shape be altered in Nile tilapia (Oreochromis niloticus) by genetic means? Aquaculture 272 (S1): S38–S46.
  51. 51. James JW. Genotype by environment interaction in farm animals. In: Adaptation and Fitness in Animal Population: Evolutionary and Breeding Perspectives on Genetic Resource Management (ed. by van der Werf J., Graser H-U., Frankham R. & Gondro C.), Springer, the Netherlands; 2009. p. 151–167.
  52. 52. Cardellino R, Rovira J. Mejoramiento genético animal. Hemisferio Sur, Buenos Aires. 1987. p 253.
  53. 53. Young-Lai WW, Charmantier-Daures M, Charmantier G. Effect of ammonia on survival and osmoregulation in different life stages of the lobster Homarus americanus. Marine Biology 1991; 110: 293–300.
  54. 54. Lin HP, Thuet P, Trilles JP, Mounet-Guillaume R, Charmantier G. Effects of ammonia on survival and osmoregulation of various development stages of the shrimp Penaeus japonicus. Marine Biology 1993; 117: 591–598.
  55. 55. Frías-Espericueta MG, Harfush-Melendez M, Osuna-Lόpez JI, Páez-Osuna F. Acute toxicity of ammonia to juvenile Penaeus vannamei Boone. Bulletin of Environmental Contamination and Toxicology 1999; 62: 646–652. pmid:10227846
  56. 56. Sun GM, Tang JH, Zhong XM. Toxicity research of ammonia nitrogen and nitrite nitrogen to Penaeus vannamei. Journal of aquaculture 2002; 1: 22–24.
  57. 57. Gallardo JA, Lhorente JP, Neira R. The consequences of including non-additive effects on the genetic evaluation of harvest body weight in Coho salmon (Oncorhynchus kisutch). Genetics Selection Evolution 2010; 42 (1):19.
  58. 58. Hoffmann AA, Merilä J. Heritable variation and evolution under favourable and unfavourable conditions. Trends in Ecology & Evolution 1999; 14: 96–101.
  59. 59. Winkelman AM, Peterson RG. Heritabilities, dominance variation, common environmental effects and genotype by environment interactions for weight and length in Chinook salmon. Aquaculture 1994; 125: 17–30.
  60. 60. Pante MJR, Gjerde B, McMillan I, Misztal I. Estimation of additive and dominance genetic variances for body weight at harvest in rainbow trout, Oncorhynchus mykiss. Aquaculture 2002; 204:383–392.
  61. 61. Paaby AB, Rockman MV. Cryptic genetic variation: evolution’s hidden substrate. Nature Reviews Genetics 2014;
  62. 62. Aguilar I, Misztal I, Johnson DL, Legarra A, Tsuruta S, Lawlor TJ. Hot topic: A unified approach to utilize phenotypic, full pedigree, and genomic information for genetic evaluation of Holstein final score. Journal of Dairy Science 2010; 93:743–752. pmid:20105546
  63. 63. Legarra A, Robert-Granié C, Manfredi E, Elsen JM. Performance of Genomic Selection in Mice. Genetics 2008; 180: 611–618. pmid:18757934
  64. 64. Andriantahina F, Liu X, Huang H. Genetic map construction and quantitative trait locus (QTL) detection of growth-related traits in Litopenaeus vannamei for selective breeding applications. PLoS ONE 2013; 8(9):e75206. pmid:24086466
  65. 65. Calus PM. Estimation of Genotype × Environment Interaction for Yield, Health and Fertility in Dairy Cattle (Ph.D. Diss.) Animal Breeding and Genetics, Wageningen University, the Netherlands 2006.
  66. 66. Mulder HA. Methods to Optimize Livestock Breeding Programs with Genotype by Environment Interaction and Genetic Heterogeneity of Environmental Variance (PhD Diss.) Animal Breeding and Genetics group, Wageningen University, the Netherlands 2007.
  67. 67. Sae-Lim P, Kause A, Mulder HA, Martin KE, Barfoot AJ, Parsons JE, et al. Genotype-by-environment interaction of growth traits in rainbow trout (Oncorhynchus mykiss): a continental scale study. Journal of Animal Science 2013; 91:5572–5581. pmid:24085417
  68. 68. Sae-Lim P, Gjerde B, Nielsen HM, Mulder H, Kause A. A reviewof genotype-byenvironment interaction and micro-environmental sensitivity in aquaculture species. Reviews in Aquaculture 2015; 7: 1–25.
  69. 69. Sae-Lim P, Komen H, Kause A. Bias and precision of estimates of genotype-by environment interaction: a simulation study. Aquaculture 2010; 310: 66–73.
  70. 70. Eknath AE, Bentsen HB, Ponzoni RW, Rye M, Nguyen NH, et al. Genetic improvement of farmed tilapias: composition and genetic parameters of a synthetic base population of Oreochromis niloticus for selective breeding. Aquaculture 2007; 273: 1–14.
  71. 71. Bentsen HB, Gjerde B, Nguyen NH, Rye M, Ponzoni RW, de Vera MSP, et al. Genetic improvement of farmed tilapias: genetic parameters for body weight at harvest in Nile tilapia (Oreochromis niloticus) during five generations of testing in multiple environments. Aquaculture 2012; 338–341:56–65.
  72. 72. Guan JT, Hu YL, Wang MS, Wang WJ, Kong J, Luan S. Estimating genetic parameters and genotype-by-environment interactions in body traits of turbot in two different rearing environments. Aquaculture 2016. 450:21–327.