- Research article
- Open Access
A SNP in the 5′ flanking region of the myostatin-1b gene is associated with harvest traits in Atlantic salmon (Salmo salar)
BMC Genetics volume 14, Article number: 112 (2013)
Myostatin (MSTN) belongs to the transforming growth factor-β superfamily and is a potent negative regulator of skeletal muscle development and growth in mammals. Most teleost fish possess two MSTN paralogues. However, as a consequence of a recent whole genome-duplication event, salmonids have four: MSTN-1 (−1a and -1b) and MSTN-2 (−2a and -2b). Evidence suggests that teleost MSTN plays a role in the regulation of muscle growth. In the current study, the MSTN-1b gene was re-sequenced and screened for SNP markers in a commercial population of Atlantic salmon. After genotyping 4,800 progeny for the discovered SNPs, we investigated their association with eight harvest traits - four body-weight traits, two ratios of weight traits, flesh colour and fat percentage - using a mixed model association analysis.
Three novel SNPs were discovered in the MSTN-1b gene of Atlantic salmon. One of the SNPs, located within the 5′ flanking region (g.1086C > T), had a significant association with harvest traits (p < 0.05), specifically for: Harvest Weight (kg), Gutted Weight (kg), Deheaded Weight (kg) and Fillet Weight (kg). The haplotype-based association analysis was consistent with this result because the two haplotypes that showed a significant association with body-weight traits, hap4 and hap5 (p < 0.05 and p < 0.01, respectively), differ by a single substitution at the g.1086C > T locus. The alleles at g.1086C > T act in an additive manner and explain a small percentage of the genetic variation of these phenotypes.
The association analysis revealed that g.1086C > T had a significant association with all body-weight traits under study. Although the SNP explains a small percentage of the variance, our results indicate that a variation in the 5′ flanking region of the myostatin gene is associated with the genetic regulation of growth in Atlantic salmon.
Myostatin (MSTN) is a potent negative regulator of skeletal muscle development and growth in mammals . Therefore, it has been an attractive candidate gene for the identification of genetic markers for growth and carcass traits in livestock species. Several mutations leading to non-functional MSTN products have been reported to cause the “double-muscling” phenotype characteristic of Piedmontese and Belgian Blue cattle [2, 3]. Additionally, a single nucleotide polymorphism (SNP) in the ovine myostatin gene has been shown to contribute to the muscular hypertrophy of Texel sheep . Selective breeding for enhanced muscle growth has indirectly increased the frequency of these naturally occurring mutations in meat-producing breeds. Moreover, as a consequence of intense selection double-muscled Belgian Blue cattle are virtually homozygous for the nt821(del11) loss-of-function mutation . The MSTN gene has also been sequenced in lower vertebrates and invertebrates, such as the zhikong scallop, the bighead carp, the Asian sea bass, the shrimp and amphioxus [6–10]. However, for the majority of non-mammalian species, the functional role of MSTN is poorly defined.
The phylogenetic analysis of MSTN genes in teleost fish suggested that a whole genome duplication event, which occurred 320-350 million years ago during early fish radiation [11, 12], resulted in two MSTN paralogues being present in modern bony fishes (MSTN-1 and MSTN-2). Additionally, in salmonids, each paralogue was duplicated once again likely due to tetraploidization (25-100 million years ago) [13, 14]. Thus, a total of four MSTN genes can be found in rainbow trout and Atlantic salmon, namely MSTN-1(−1a and -1b) and MSTN-2(−2a and -2b) paralogues. MSTN gene structure among teleosts is highly conserved, comprising three exons of comparable size (300–400 nucleotides) separated by two introns, similar to mammals . However, in contrast to mammals, where the expression of a single MSTN gene is limited primarily to skeletal muscle , teleost fish exhibit a broad pattern of differentially expressed MSTN paralogues. For example, MSTN-1 transcripts have been detected in eye, spleen, muscle and brain, among other tissues . In comparison, MSTN-2 expression pattern has been shown to be more limited and occurs mostly in the brain [18, 19]. While it is unclear exactly what functional role MSTN plays in these tissues, evidence indicates that teleost MSTN might be involved in the regulation of muscle development and growth (however, see ). In transgenic zebrafish the suppression of the MSTN gene by RNA interference has led to a double-muscled phenotype , suggesting a similar biological function to that previously described in mammals. This functionality is supported by an increase in the number of muscle fibers (hyperplasia) observed in transgenic Medaka that expresses a dominant-negative MSTN. Transgenic trout overexpressing follistatin, one of the more efficacious antagonists of MSTN, also showed enhanced muscle growth due to hyperplasia .
The Atlantic salmon (Salmo salar L.) is the most important aquaculture species in several countries, including Chile and Norway . Currently, selective breeding programs are being carried out in Norway, Canada, Iceland, Chile and the UK . An important goal of finfish aquaculture is to enhance skeletal muscle growth and improve fillet trait characteristics. Since the Atlantic salmon MSTN-1b promoter region harbours cis-regulatory elements (E-boxes) that have been shown to bind in vitro to the myogenic differentiation factor MyoD , a regulatory factor of importance for initiating the myogenic program, it is reasonable to suggest that this paralogue is a strong candidate for regulation of skeletal muscle growth. If true, variation within Atlantic salmon MSTN-1b gene may be contributing to the phenotypic variation of economically important harvest traits. Accordingly, relevant polymorphisms could then be applied as tools for marker-assisted selection (MAS) in salmon breeding programs. The objective of this study was to identify novel SNP markers on the SsMSTN-1b gene and analyse their association with growth, fatness and colour-related traits in a commercial population of Atlantic salmon.
SNP identification and genotypes
The Atlantic salmon MSTN-1b gene was re-sequenced to screen for polymorphic variation and to assess the association between this variation and harvest traits. The MSTN-contig used for primer design was built by the alignment of the SsMSTN-1b gene and two contigs of the Atlantic salmon draft genome assembly (see ; NCBI assembly GCA_000233375.1), Contig_064406 and Contig_398167. Following amplification and sequencing of PCR products from ten samples of parental salmon, amplicons generated by two primer pairs (SaMSTNb23 and SaMSTNb33) were found to contain two and one SNP(s), respectively (see Table 1). By re-aligning the PCR products containing the SNPs with SsMSTN-1b, their position in the gene was established: two were located in the upstream putative promoter region of the gene (g.1060C > A and g.1086C > T) and one in the third exon (c.3501C > G). For c.3501C > G, the substitution of alternative alleles was functionally synonymous.
A total of 4,800 animals were genotyped. Genotypes and frequencies estimated for the three MSTN loci are shown in Table 2. The genotype frequencies of g.1060C > A and g.1086C > T were consistent with HWE expectations. However, c.3501C > G was not in HW equilibrium (p < 0.001), showing a slight deficit of heterozygous animals. The correlations between alleles indicate that g.1060C > A and g.1086C > T were in strong LD (D’ 0.99, r2 0.38). There was evidence of recombination between g.1086C > T and c.3501C > G (pairwise D’ 0.52 and r2 0.13).
The population of 4,800 commercial Atlantic salmon had been measured for several traits at harvest (approximately 3 years of age). The phenotypic mean, standard deviations and ranges of the traits used in this study are given in Table 3 along with their heritability. All of the weight traits (HWT, GWT, DHWT and FLWT) showed similar and high heritabilities (~0.5), whereas yield traits (GYLD and FLYLD) exhibited low estimates (h2 < 0.05). Fat content and flesh colour heritabilities were moderate, viz. 0.17 and 0.29, respectively.
To assess the association between SNP genotype and harvest traits, a mixed model analysis was performed in 4,759 successfully genotyped individuals. A significant association was observed between g.1086C > T genotypes (p < 0.05) and all weight phenotypes (HWT, GWT, DHWT and FLWT), but not with any of the other traits. This significance was maintained when g.1086C > T was fitted either separately or simultaneously in the model with the other SNPs. Genotypes of g.1060C > A and c.3501C > G were not significantly associated with any trait when fitted either individually or simultaneously in the model.
To assess the size of effect associated with the significant MSTN-1b SNP alleles, the predicted trait values for each genotypic class of g.1086C > T were calculated. The thymine allele was associated with an increase in each of the weight traits (p < 0.05), with the additive effect of the SNP on these traits ranging from 30 to 50 g (Table 4). Whilst this effect was significant, the percentage of the additive genetic variance explained by g.1086C > T in each trait was less than 1%. The dominance effect was trivial and non-significant.
Haplotypes were constructed for the three SNPs and the association between haplotype and trait was assessed. In the Atlantic salmon population analysed, 6 haplotypes and 13 diplotypes were identified. Haplotypes with a frequency < 0.01 were excluded from further analysis. The most prevalent haplotype, hap5, accounted for 67% of all haplotypes. The most common diplotype (43% of samples) comprised hap5 homozygotes (Table 5). When haplotype combination was fitted in the model, hap4 showed a significant association with the weight traits (p < 0.05), and the association of hap5 with the same traits was highly significant (p < 0.01); the exception was the fillet weight trait (FLWT), for which only hap5 was significant (p-value = 0.031) (Table 6). These two haplotypes differ in a single nucleotide substitution at g.1086C > T, supporting the significant association of this SNP with harvest traits. The occurrence of hap5 at a dosage of two copies in the SsMSTN-1b gene was related with a decrease in 60 to 110 g in body weight traits compared to zero copies.
In this study, the SsMSTN-1b gene was re-sequenced and three novel SNPs were detected: g.1060C > A and g.1086C > T, both in the 5′ flanking region; and c.3501C > G, located in the third exon of the gene. The association analysis showed that g.1086C > T had a significant association with the weight traits under study (HWT, GWT, DHWT and FLWT), all of which show a high positive phenotypic correlation (r > 0.97). Quantitative trait loci (QTL) associated with flesh colour and growth traits have been described for Atlantic salmon [26–30]. However, SsMSTN-1b is linked to markers mapping to chr25 , where only  identified a QTL for body-weight in ~38 month-old fish. In general, QTL scans for growth traits in Atlantic salmon suggest that body-weight traits are highly polygenic. As many loci of small effect are expected to be co-regulating traits related to the growth of fish, it is possible that previous QTL mapping studies failed to detect the effect observed in the current study due to a lack of statistical power.
For aquacultural species, a significant association between MSTN polymorphisms and production traits has been detected in the bighead carp (Aristichthys nobilis), the yellow catfish (Pelteobagrus fulvidraco), the spotted halibut (Verasper variegatus), the common carp (Cyprinus carpio), the Atlantic bay scallop (Argopecten irradians) and the gilthead seabream (Sparus aurata) [7, 31–35]. Interestingly, for marine species the majority of polymorphisms that have been described as having an effect on weight traits (including the present study) are located in non-coding regions of the MSTN gene. Although this might be a reflection of low frequency of coding genetic variants in marine species [7, 8], it may also be a consequence of the incipient stage of aquacultural research and the shorter history of selective breeding for aquaculture species. For example, it is possible that loss or reduction of function mutations in coding regions of MSTN have not yet been observed and/or selected to appreciable frequencies in commercial aquaculture species. Despite the observation that mutations in the coding region of the myostatin gene are known to cause an increase in muscle mass in several mammals, non-coding mutations associated with regulatory pathways may also be underlying phenotypic variation. Moreover, “double-muscling” in Texel sheep is associated with mutations in the 3′UTR of the MSTN gene, which create illegitimate miRNA binding sites and reduce the amount of circulating MSTN protein . It remains to be investigated whether studied marine species encode a more stable MSTN protein, and whether the main effects on growth mediated by this gene are associated with gene regulation or gene structure.
The haplotype-based association analysis was consistent with the significant effect detected for g.1086C > T; the two haplotypes that showed a significant association with weight traits, hap4 and hap5 (p < 0.05 and p < 0.01, respectively), differed only by a nucleotide substitution at this locus. In accordance with the predicted trait values of the genotypes at g.1086C > T, two copies of the haplotype carrying the unfavourable SNP allele, homozygote hap5/hap5, was associated with a decrease in all body weight traits. The difference between the predicted mean of individuals carrying zero or two copies of hap5 varies between 60 g (for FLWT) to 110 g (for HWT).
Myostatin is an important target gene for aquaculture research. Some studies have explored the improvement of growth through the suppression of gene activity by over-expressing the MSTN prodomain (Mstnpro). For example,  increased growth rates of rainbow trout by immersing juveniles in bath treatments with flatfish MSTN-1pro expressed in Escherichia coli. The improvement of body mass by inhibiting the myostatin gene has also been achieved in African catfish (Clarias gariepinus), goldfish (Carassius auratus) and tilapia (Oreochromis aureus) larvae by an immersion bath treatment with a soluble form of the Active Type IIb receptor . However, a study by  showed that the positive growth responses achieved by juvenile tilapia under MSTN inhibition by immersion were not sustained until market size; after 45 weeks of exposure to flatfish Mstnpro, no significant weight or length differences between control and immersed tilapia groups were observed.
A promising alternative to short or long-term administration of MSTN inhibiting agents is marker-assisted-selection (MAS) exploiting favourable alleles at naturally segregating polymorphisms in commercial populations of fish. However, in contrast to mammals like the Belgian blue cattle, where a natural deletion of 11 base pairs on the third exon increased from 20–25% muscle mass by hyperplasia [3, 5], no genetic variants with large impacts on weight phenotypes have been identified in fish . Our results show that genetic variation at g.1086C > T has a significant association with growth traits in a commercial population of Atlantic salmon, although the proportion of variance explained by this marker is relatively small (< 1%). Nonetheless, this SNP should be evaluated further to assess its effect in other populations of salmon and to test any possible functional role on the promoter region of the SsMSTN-1b gene. In other domestic animals, it is likely that polymorphisms with large effects on growth traits are rare due to larger selection pressures moving favourable alleles towards fixation . However, Atlantic salmon and other aquaculture species are relatively recently domesticated, and polymorphisms of large effect may feasibly still be segregating in commercial populations. Therefore, future studies should aim at evaluating the effect of additional MSTN polymorphisms and other candidate genes for growth in Atlantic salmon. These results would not only be of importance in fish breeding, but also may aid insights into the physiology of muscle growth and development in fish.
The aim of the study was to discover SNPs in the Atlantic salmon MSTN-1b gene and to evaluate their association with growth and fillet traits. Three SNPs were found to be segregating in the fish that were sequenced, and these were then tested in a large commercial population. A SNP located in the upstream region of the gene (g.1086C > T) was associated with all body weight traits under study (HWT, GWT, DHWT and FLWT). The alleles in the g.1086C > T locus acted in an additive matter, with a change from a CC to a TT genotype associated with an increase of 70 to 100 g depending on the trait. These results add to the evidence that suggests SsMSTN-1b, and potentially the orthologous gene in other teleost species, plays a role in muscle development in fish. The combined effect of further QTL mapping and candidate gene studies assessing the association between genotypes and growth traits may further unravel genes of larger effect and lead to an improved understanding of the regulation of muscle growth in fish.
To discover polymorphisms in the Atlantic salmon MSTN-1b locus, specific primer pairs (Table 1) were designed to generate overlapping PCR products from a MSTN-contig that contained the entire gene (6394 bp) and ~300 bp of flanking sequence at both ends. This contig was built by aligning the complete SsMSTN-1b gene [NCBI: Acc. Num. AJ316006.2] with the Atlantic salmon draft genome assembly (; NCBI assembly GCA_000233375.1). For SNP discovery and the optimization of PCR amplifications, a panel of ten parental individuals was randomly chosen from randomly-selected families of a commercial population of Atlantic salmon (sourced from Landcatch Natural Selection, Ormsary, UK). By searching for polymorphisms in a subset of the parental samples we may have failed to detect some SNPs in our population, particularly those with a rare allele frequency (e.g. < 0.05). However, sampling 10 individuals is adequate to give a good probability of finding SNPs with a minor allele frequency above 0.1. Each PCR product was sequenced using an ABI 3730xl at ARK-Genomics (Roslin, UK) with forward and reverse primers to check for consistency of sequences. SNPs were identified by visual inspection of both chromatograms with BioEdit .
A. Animals and traits
This study was based on 4,800 fish comprising 198 commercial families from Landcatch Natural Selection Ltd. Families were created in 1999 by crossing 136 sires and 198 dams. Pedigree information was available for two previous generations for all individuals with phenotypic records. The phenotypic data was collected at the time of harvesting 3-year old fish and included Harvest Weight (kg), Gutted Weight (kg), Deheaded Weight (kg), Fillet Weight (kg), Gutted Yield (%) and Fillet Yield (%). In addition, Fat Percentage and Fillet Colour were recorded. Fat Percentage was estimated as the mean of eight readings along the animal’s body using the Torry Fatmeter (Distell Ltd). Scores for Fillet Colour ranged from 20–34 units of colour (yellow to red) and derived from the visual contrast of the fillet against the industry standard Roche colour chart (see  for details of trait collection).
B. DNA extraction and genotyping
Total genomic DNA was extracted from adipose fin tissue using a Biosprint DNA kit (QUIAGEN, Crawley, UK) following the manufacturer’s instruction. The genotyping of the discovered SNPs (see SNP Identification above) was performed on all 4,800 fish by LGC Genomics Ltd (Herts, U.K.) using a KASP assay. The KASP assay is a competitive allele-specific PCR-based genotyping system that allows high levels of assay robustness and accuracy (see technology details at http://www.kbioscience.co.uk/reagents/KASP.html). Allele-specific primers were designed and utilised by LGC Genomics based on the supplied gene sequences (see Additional file 1).
C. Statistical analysis
Test for Hardy-Weinberg equilibrium and linkage disequilibrium
Three SNPs segregating in SsMSTN-1b were further tested for departure from Hardy-Weinberg equilibrium (HWE) with a χ2 goodness of fit test. Linkage Disequilibrium (LD) was estimated for each pair of SNPs, using both the squared correlation, r2, and the normalised linkage disequilibrium coefficient, D’. Both statistical analyses were performed using Haploview 4.2 .
Mixed model association analysis
Descriptive statistics of harvest traits were performed using the SAS software (SAS Inst., Inc., Cary, NC). The heritability values of each phenotype were calculated from the result of the partitioning of variance components (h2 = Va/Vp, where Va and Vp are the additive genetic and the phenotypic variance, respectively) obtained by fitting a single-trait animal model and omitting SNP genotype as a fixed effect, using the ASReml package . To evaluate the relationship between the discovered markers and the harvest traits, SNP genotypes and haplotypes were included in the model as fixed effects.
The mixed model was as follows:
where Yik is vector of one of the traits on the individual i; μ is the overall mean of the trait; Gk is the fixed effect of the SNP genotype k (3 classes) or the effect of the haplotype (coded as the number of copies given per each haplotype described in the population: 0, 1 or 2 copies); ai is the additive effect of the ith animal; and e is the residual term. All available pedigree information was included when fitting the model. Statistical significance of the fixed effects was assessed using the Wald F-statistics with denominator degrees of freedom from the fitted model.
Predicted SNP genotype effects
For the SNP(s) that showed significant association with a harvest trait, differences between the means of each genotypic class and allelic frequencies were used to estimate additive and dominance effects . Standard errors for both effects were calculated from the variance-covariance matrix of the predicted genotype classes along with the standard errors of their differences (SED). The percentage of the additive genetic variance (%Va) explained by the SNP was determined using the standard formula: 2pq(a + d(q–p))2]/Va, where p and q are the major and minor allele frequencies of the SNP, a is the additive effect and d is the calculated dominance effect. The additive genetic variance (Va) was taken from the mixed model without fitting genotype.
Lee SJ, McPherron AC: Myostatin and the control of skeletal muscle mass. Curr Opin Genet Dev. 1999, 9: 604-607. 10.1016/S0959-437X(99)00004-0.
McPherron AC, Lee SJ: Double muscling in cattle due to mutations in the myostatin gene. Proc Natl Acad Sci U S A. 1997, 94: 12457-12461. 10.1073/pnas.94.23.12457.
Kambadur R, Sharma M, Smith TPL, Bass JJ: Mutations in myostatin (GDF8) in double-muscled Belgian blue and piedmontese cattle. Genome Res. 1997, 7: 910-915.
Clop A, Marcq F, Takeda H, Pirottin D, Tordoir X, Bibé B, Bouix J, Caiment F, Elsen J-M, Eychenne F, Eychenne F, Larzul C, Laville E, Meish F, Milenkovic D, Tobin J, Charlier C, Georges M: A mutation creating a potential illegitimate microRNA target site in the myostatin gene affects muscularity in sheep. Nat Genet. 2006, 38: 813-818. 10.1038/ng1810.
Grobet L, Martin L, Poncelet D, Pirottin D, Brouwers B, Riquet J, Schoeberlein A, Dunner S, Ménissier F, Massabanda J, Fries R, Hanset R, Georges M: A deletion in the bovine myostatin gene causes the double-muscled phenotype in cattle. Nat Genet. 1997, 17: 71-74. 10.1038/ng0997-71.
Hu X, Guo H, He Y, Wang S, Zhang L, Huang X, Roy SW, Lu W, Hu J, Bao Z: Molecular characterization of myostatin gene from zhikong scallop Chlamys farreri (Jones et preston 1904). Genes Genet Syst. 2010, 85: 207-218. 10.1266/ggs.85.207.
Liu L, Yu X, Tong J: Molecular characterization of myostatin (MSTN) gene and association analysis with growth traits in the bighead carp (Aristichthys nobilis). Mol Biol Rep. 2012, 39: 9211-9221. 10.1007/s11033-012-1794-6.
De Santis C, Evans B, Smith-Keune C, Jerry D: Molecular characterization, tissue expression and sequence variability of the barramundi (Lates calcarifer) myostatin gene. BMC Genomics. 2008, 9: 82-10.1186/1471-2164-9-82.
De Santis C, Wade NM, Jerry DR, Preston NP, Glencross BD, Sellars MJ: Growing backwards: an inverted role for the shrimp ortholog of vertebrate myostatin and GDF11. J Exp Biol. 2011, 214: 2671-2677. 10.1242/jeb.056374.
Xing F, Tan X, Zhang PJ, Ma J, Zhang Y, Xu P, Xu Y: Characterization of amphioxus GDF8/11 gene, an archetype of vertebrate MSTN and GDF11. Dev Genes Evol. 2007, 217: 549-554. 10.1007/s00427-007-0162-3.
Amores A, Force A, Yan Y-L, Joly L, Amemiya C, Fritz A, Ho RK, Langeland J, Prince V, Wang Y-L, Westerfield M, Ekker M, Postlethwait JH: Zebrafish hox Clusters and Vertebrate Genome Evolution. Science. 1998, 282: 1711-1714.
Postlethwait JH, Yan Y-L, Gates MA, Horne S, Amores A, Brownlie A, Donovan A, Egan ES, Force A, Gong Z, Goutel C, Fritz A, Kelsh R, Knapik E, Liao E, Paw B, Ransom D, Singer A, Thomson M, Abduljabbar TS, Yelick P, Beier D, Joly J-S, Larhammar D, Rosa F, Westerfield M, Zon LI, Johnson SL, Talbot WS: Vertebrate genome evolution and the zebrafish gene map. Nat Genet. 1998, 18: 345-349. 10.1038/ng0498-345.
Kerr T, Roalson EH, Rodgers BD: Phylogenetic analysis of the myostatin gene sub-family and the differential expression of a novel member in zebrafish. Evol Dev. 2005, 7: 390-400. 10.1111/j.1525-142X.2005.05044.x.
Allendorf FW, Thorgaard GH: Tetraploidy and the Evolution of Salmonid Fishes. Evolutionary Genetics of Fishes. Edited by: Turner BJ. 1984, New York: Plenum Press, 1-53.
Gabillard J-C, Biga PR, Rescan P-Y, Seiliez I: Revisiting the paradigm of myostatin in vertebrates: insights from fishes. Gen Comp Endocrinol. 2013, 194: 45-54.
McPherron AC, Lawler AM, Lee SJ: Regulation of skeletal muscle mass in mice by a new TGF-beta superfamily member. Nature. 1997, 387: 83-90. 10.1038/387083a0.
Garikipati D, Gahr S, Rodgers B: Identification, characterization, and quantitative expression analysis of rainbow trout myostatin-1a and myostatin-1b genes. J Endocrinol. 2006, 190: 879-888. 10.1677/joe.1.06866.
Ostbye T-K, Wetten O, Tooming-Klunderud A, Jakobsen K, Yafe A, Etzioni S, Moen T, Andersen O: Myostatin (MSTN) gene duplications in Atlantic salmon (Salmo salar): evidence for different selective pressure on teleost MSTN-1 and −2. Gene. 2007, 403: 159-169. 10.1016/j.gene.2007.08.008.
Maccatrozzo L, Bargelloni L, Cardazzo B, Rizzo G, Patarnello T: A novel second myostatin gene is present in teleost fish. FEBS Lett. 2001, 509: 36-40. 10.1016/S0014-5793(01)03124-6.
Lee CY, Hu SY, Gong HY, Chen MH, Lu JK, Wu JL: Suppression of myostatin with vector-based RNA interference causes a double-muscle effect in transgenic zebrafish. Biochem Biophys Res Commun. 2009, 387: 766-771. 10.1016/j.bbrc.2009.07.110.
Sawatari E, Seki R, Adachi T, Hashimoto H, Uji S, Wakamatsu Y, Nakata T, Kinoshita M: Overexpression of the dominant-negative form of myostatin results in doubling of muscle-fiber number in transgenic medaka (Oryzias latipes). Comp Biochem Physiol A Mol Integr Physiol. 2010, 155: 183-189. 10.1016/j.cbpa.2009.10.030.
Medeiros E, Phelps M, Fuentes F, Bradley T: Overexpression of follistatin in trout stimulates increased muscling. Am J Physiol Regul Integr Comp Physiol. 2009, 297: 42-10.1152/ajpregu.90964.2008.
FAO: The State of the World Fisheries and Aquaculture. 2010, Rome: FAO
Fjalestad K, Moen T, Gomez-Raya L: Prospects for genetic technology in salmon breeding programmes. Aquac Res. 2003, 34: 397-406. 10.1046/j.1365-2109.2003.00823.x.
Davidson WS, Koop BF, Jones SJ, Iturra P, Vidal R, Maass A, Jonassen I, Lien S, Omholt SW: Sequencing the genome of the Atlantic salmon (Salmo salar). Genome Biol. 2010, 11: 403-
Baranski M, Moen T, Våge D: Mapping of quantitative trait loci for flesh colour and growth traits in Atlantic salmon (Salmo salar). Genet Sel Evol. 2010, 42: 17-10.1186/1297-9686-42-17.
Reid D, Szanto A, Glebe B, Danzmann R, Ferguson M: QTL for body weight and condition factor in Atlantic salmon (Salmo salar): comparative analysis with rainbow trout (Oncorhynchus mykiss) and Arctic charr (Salvelinus alpinus). Heredity. 2005, 94: 166-172. 10.1038/sj.hdy.6800590.
Boulding E, Culling M, Glebe B, Berg P, Lien S, Moen T: Conservation genomics of Atlantic salmon: SNPs associated with QTLs for adaptive traits in parr from four trans-Atlantic backcrosses. Heredity. 2008, 101: 381-391. 10.1038/hdy.2008.67.
Houston R, Bishop S, Hamilton A, Guy D, Tinch A, Taggart J, Derayat A, McAndrew B, Haley C: Detection of QTL affecting harvest traits in a commercial Atlantic salmon population. Anim Genet. 2009, 40: 753-755. 10.1111/j.1365-2052.2009.01883.x.
Gutierrez A, Lubieniecki K, Davidson E, Lien S, Kent M, Fukui S, Withler R, Swift B, Davidson W: Genetic mapping of quantitative trait loci (QTL) for body-weight in Atlantic salmon (Salmo salar) using a 6.5 K SNP array. Aquaculture. 2012, 358–359: 61-70.
Zhu YY, Liang HW, Li Z, Luo XZ, Li L, Zhang ZW, Zou GW: Polymorphism of MSTN gene and its association with growth traits in yellow catfish (Pelteobagruse fulvidraco). Yi Chuan. 2012, 34: 72-78. 10.3724/SP.J.1005.2012.00072.
Li H, Fan J, Liu S, Yang Q, Mu G, He C: Characterization of a myostatin gene (MSTN1) from spotted halibut (Verasper variegatus) and association between its promoter polymorphism and individual growth performance. Comp Biochem Physiol B Biochem Mol Biol. 2012, 161: 315-322. 10.1016/j.cbpb.2011.12.008.
Sun Y, Yu X, Tong J: Polymorphisms in myostatin gene and associations with growth traits in the common carp (Cyprinus carpio L.). Int J Mol Sci. 2012, 13: 14956-14961. 10.3390/ijms131114956.
Guo L, Li L, Zhang S, Guo X, Zhang G: Novel polymorphisms in the myostatin gene and their association with growth traits in a variety of bay scallop, Argopecten irradians. Anim Genet. 2011, 42: 339-340.
Sanchez-Ramos I, Cross I, Macha J, Martinez-Rodriguez G, Krylov V, Rebordinos L: Assessment of tools for marker-assisted selection in a marine commercial species: significant association between MSTN-1 gene polymorphism and growth traits. Sci World J. 2012, 2012: 369802-
Lee SB, Kim YS, Oh M-Y, Jeong I-h, Seong K-B, Jin H-J: Improving rainbow trout (Oncorhynchus mykiss) growth by treatment with a fish (Paralichthys olivaceus) myostatin prodomain expressed in soluble forms in E. coli. Aquaculture. 2010, 302: 270-278. 10.1016/j.aquaculture.2010.02.027.
Carpio Y, Acosta J, Morales R, Santisteban Y, Sanchéz A, Estrada M: Regulation of body mass growth through activin type IIB receptor in teleost fish. Gen Comp Endocrinol. 2009, 160: 158-167. 10.1016/j.ygcen.2008.11.009.
Kim Y-S, Fox B, Kim KH, Lee SB, Jin HJ, Tamaru CS: Immersion bath treatment of tilapia fry with myostatin-1 prodomain does not affect tilapia growth at market size. Aquac Res. 2012, 44: 1643-1648.
Stinckens A, Georges M, Buys N: Mutations in the myostatin gene leading to hypermuscularity in mammals: indications for a similar mechanism in fish?. Anim Genet. 2011, 42: 229-234. 10.1111/j.1365-2052.2010.02144.x.
Goddard M, Hayes B: Mapping genes for complex traits in domestic animals and their use in breeding programmes. Nat Rev Genet. 2009, 10: 381-391. 10.1038/nrg2575.
Hall TA: BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp Ser. 1999, 41: 95-98.
Powell J, White I, Guy D, Brotherstone S: Genetic parameters of production traits in Atlantic salmon (Salmo salar). Aquaculture. 2008, 274: 225-231. 10.1016/j.aquaculture.2007.11.036.
Barrett JC, Fry B, Maller J, Daly MJ: Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics. 2005, 21: 263-265. 10.1093/bioinformatics/bth457.
Stephens M, Donnelly P: A comparison of bayesian methods for haplotype reconstruction from population genotype data. Am J Hum Genet. 2003, 73: 1162-1169. 10.1086/379378.
Stephens M, Smith N, Donnelly P: A new statistical method for haplotype reconstruction from population data. Am J Hum Genet. 2001, 68: 978-989. 10.1086/319501.
Gilmour AR, Gogel BJ, Cullis BR, Welham SJ, Thompson R: ASREML User Guide Release 3.0. 2009, VSN International Ltd: Hemel Hempstead, UK
Falconer DS, Mackay TFC: Introduction to Quantitative Genetics. 1996, Longmans Green: Harlow, Essex, UK, 4
The authors acknowledge staff at the ARK Genomics facility for assistance with ABI sequencing. The authors acknowledge funding from the British Biotechnology and Biological Sciences Research Council (BBSRC), grant number BB/H022007/1, the Roslin Institute’s BBSRC Institute Strategic Funding Grant, and CONICYT (Chile).
The authors declare that they have no competing interests.
CP carried out the experiments that led to SNP discovery, performed the statistical analysis and drafted the manuscript. SB and RH conceived the study and helped to draft the manuscript. AH and DG contributed to the collection and management of the harvest data, pedigrees and samples. All authors read and approved the final manuscript.