Introduction

Melanocortin-4 receptor (MC4R) is a G-protein-coupled receptor expressed in the appetite-regulating area of brain, which is associated with feed intake regulation and energy balance (Fan et al. 1997; Meidtner et al. 2010). Huszar et al. (1997) observed that knockout of MC4R induced symptoms of polyphagia, obesity, and higher levels of insulin secretion in mice. Moreover, mutations of the MC4R gene have been found to be the most common cause of hereditary obesity in humans (Carroll et al. 2005). Since MC4R links to the control of body weight via regulation of energy homeostasis, it has been considered as a potentially valuable gene for improving growth-related traits in animals.

Myostatin (MSTN) is a negative regulator of muscle development through regulating both the number and growth of muscle fibers (Lee and Mcpherron 1999). Mutations of MSTN leading to non-functional proteins have been reported to cause the “double-muscling” phenotype in cattle (Mcpherron and Lee 1997; Kambadur et al. 1997). Similarly, MSTN gene knockout was shown to cause a significant increase in muscle mass in mice, as a consequence of muscle cell hypertrophy and hyperplasia (McPherron et al. 1997). Additionally, knockdown of the MSTN gene has been shown to give rise to muscular hypertrophy in zebra fish (Lee et al. 2009). This gene has therefore been considered a potential candidate gene for identification of genetic markers and improving growth and meat quality traits in livestock and fish.

Single-nucleotide polymorphism (SNP) is a common form of variation in genes, promoters, and regulatory regions, it is widely distributed throughout genomes. Therefore, some SNPs may affect biological phenotypes through modifying the expression of genes. SNPs have become the focuses of intense research in the fields of fish genetics and breeding (Houston et al. 2012; Poćwierz-Kotus et al. 2014; Yang et al. 2016). SNPs of MSTN associated with production traits have been widely reported, both in livestock, such as sheep (Hickford et al. 2010), pig (Jiang et al. 2002), and cattle (Sellick et al. 2007), and in fish, such Atlantic salmon (Salmo salar; Peñaloza et al. 2013), common carp (Cyprinus carpio; Sun et al. 2012), spotted halibut (Verasper variegatus; Li et al. 2012), and bighead carp (Aristichthys nobilis; Liu et al. 2012). Although polymorphisms located in the MC4R gene have been widely demonstrated to be associated with growth traits (Zeng et al. 2014; Kim et al. 2000; Cai et al. 2015; Lee et al. 2013; Fontanesi et al. 2013), to date there have been few reports regarding associations between MC4R gene polymorphisms and growth quality traits in fish.

The cyprinid fish S. hollandi (Cyprinidae: Cypriniformes) is widely distributed in the south of China, including the provinces of Guangxi, Guangdong, Fujian, and Anhui. As S. hollandi has high nutritional and medicinal value, it produces high economic benefit (Cai et al. 2007). In recent years, increasing demand for S. hollandi has stimulated considerable research on this species. Most previous studies on S. hollandi have focused on genetic resources (Shu et al. 2015), rearing conditions (Lv et al. 2008), ethology (Li et al. 2011), and morphology (Wang 2013), whereas there is considerably less attention paid to analysis of marker-assisted selection.

Therefore, the aims of this study were to: (1) identify SNPs of the MSTN and MC4R genes and investigate the relationships between MSTN and MC4R polymorphisms and growth traits in S. hollandi, (2) analyze the relationships between expression levels of the MC4R and MSTN genes, their SNPs and growth traits in S. hollandi.

Materials and methods

Materials and phenotypic data collection

The experimental fish were obtained from Shaoguan Fisheries Research Institute in Guangdong, China. The parent fish of the experimental population used for association analysis were selected from the first generation of wild stock collected from the Beijiang River. All fish were hatched at the same time and cultured under the same rearing and management conditions. At the age of 1 year, 235 S. hollandi with an average weight of 134.75 ± 38.66 g were randomly selected without differentiating sexes. Five growth traits, including body weight (BWT), body length (BL), total length (TL), body depth (BD), and body width (BWH), were measured in each fish for association analysis. The part of caudal fin of each fish was collected and preserved in 95% ethanol. After completing measurements, each fish was marked and maintained with feeding under pre-measurement conditions in another pool.

PCR amplification and SNP identification

To detect MSTN and MC4R polymorphisms, two primer pairs, MSTNPF, 5′- ACAGATACGTGAATATTATC-3′, MSTNPR, 5′-TGCGCCGTTATATCTCCATG-3′, and MC4RPF, 5′-TCTTTATGAGTGAATTACTG-3′, MC4RPR, 5′-CAAAGCAGGTGCTGTGTGAG-3′, were designed using Primer Premier 5.0 software (Premier Biosoft International, Palo Alto, CA) based on the DNA sequences of S. hollandi MSTN (GenBank: KY853657) and MC4R (GenBank: KY022411) genes. PCR amplification was performed in a reaction volume of 50 µL, containing 25 µL 2 × Taq Master Mix (Dye Plus, Vazyme Biotech, Nanjing, China), 20 µL double-distilled water, 3 µL DNA solution, and 1 µL of each primer (10 µM). The PCR amplification reactions were performed using the following thermo cycle program: 94 °C for 10 min, followed by 35 cycles of 94 °C for 45 s, 55 °C for 30 s, and 72 °C for 1 min, with a final extension at 72 °C for 10 min. Amplification results were verified by 2% agarose gel electrophoresis and PCR fragments of the predicted size were purified using an agarose gel DNA Extraction kit (Generay Biotech Co., Ltd, Shanghai, China).

SNP identification, genotyping, and association analysis

Sequencing of the amplified DNA fragments was performed by Generay Biotech Co., Ltd using an ABI 3730XL sequencer (Applied Biosystems, USA). Differences of gene sequences between individuals were detected using SeqMan version 7.1.0 (DNASTAR Inc., Madison, WI, USA). SNPs were detected and genotyped through observing and comparing chromatogram files using Chromas version 2.33 (Technelysium Pty Ltd., South Brisbane, Australia). Association analyses between genotypes of the MSTN gene and the MC4R gene and the five selected growth traits were performed using post hoc multiple comparisons (the Duncan method) with SPSS 19.0 software (IBM, Armonk, NY, USA).

Association between gene expression levels and growth trait-related SNPs

On the basis of the results of association analyses, for each genotype of the growth trait-related SNPs, 12 fish were randomly collected from marked fish to analyze the association between gene expression levels and the growth trait-related SNPs. Sampled fish were anesthetized by immersion in 0.1% eugenol for 1–2 min, and these fish were then dissected using stainless steel scissors. Brain or/and muscle samples were frozen immediately in liquid nitrogen and stored at − 80 °C for RNA isolation. The brains were used to analyze the association between expression levels and the growth trait-related SNP in MC4R, whereas both muscle and brains were used for the same analysis in MSTN.

RNA isolation and the real-time PCR

Total RNA was isolated from frozen samples using RNA Isolater Total RNA Extraction Reagent (Vazyme Biotech, Nanjing, China) according to the manufacturer’s instructions. RNA concentrations were determined at 260 nm. The 260/280 ratio was used to verify the quality of the RNA in each sample. RNA samples were dissolved in diethylpyrocarbonate-treated water and stored at − 80 °C.

Real-time PCR was performed using an ABI 7000 thermal cycler in 20-µL reaction volumes containing the following components: 100 ng of cDNA, 10 µL Power SYBR Green PCR Master Mix (Vazyme Biotech Co., Ltd, Nanjing, China), 0.3 µL of each primer (10 µmolL−1), and 7.4 µL double-distilled water. The primer pairs, QMSTNF, 5′-ATGACCATGGCCACAGAGCCTG-3′, QMSTNR, 5′-CCGGTCTCAGATGAACCCAGAGC-3′, and QMC4RF, 5′-AGCCGTAGCAGACTTGTTGGTC-3′, QMC4RR, 5′-TGTTCTTGATGATGCTCTCGCG-3′, were used for RT-PCR amplification of MSTN and MC4R, respectively. All samples were analyzed in triplicate and the mean value was used to calculate mRNA expression levels. β-actin was amplified as the internal control gene with the primers F (5′-CAGCCATCCTTCTTGGGTATG-3′) and R (5′-TCTGCATACGGTCAGCAATGC-3′). The relative mRNA expression levels of each genotype were analyzed using the \({2^{ - \Delta \Delta {C_{\text{t}}}}}\) method.

Statistical analysis

Genetic analyses, including Hardy–Weinberg equilibrium (HWE), expected heterozygosity (He), and observed heterozygosity (Ho), were calculated using Haploview software (Broad Institute, America). Polymorphism information content (PIC) was calculated using PIC CALC version 0.6 (Yellow Sea Fisheries Research Institute, Qingdao, China). The effects of different SNP genotypes on the five selected growth traits were analyzed by one-way ANOVA using SPSS 19.0 software. The genotypes of SNPs significantly associated with the growth traits of S. hollandi were analyzed through post hoc multiple comparisons (the Duncan method). Analysis of the association between gene expression levels and growth trait-related SNPs was performed using Student’s t-test. The following statistical model was applied:

$$Y=u+G+e,$$

where Y is the phenotypic value of each trait, u is the population mean value of each growth trait, G is the fixed genotypic effect of each SNP, and e is the random error effect.

Results

SNP identification and genotyping

Two SNPs (MC4R-719A>G and MSTN-519C>T) were detected in the promoters of MC4R and MSTN, respectively. Chi square tests revealed that MC4R -719A>G was in HWE (p > 0.05), whereas MSTN-519C > T deviated from the HWE (p < 0.05), The He, Ho, PIC, and Hardy–Weinberg p-values of these two SNPs are shown in Table 1. The two SNPs were classified as being moderately polymorphic loci based on the following criteria: loci with PIC > 0.5 are highly polymorphic; loci with 0.5 > PIC > 0.25 are moderately polymorphic; and loci with PIC < 0.25 have low polymorphism (Vaiman et al. 1994). The allele and genotype frequencies of the SNPs are shown in Table 2.

Table 1 The genetic polymorphic information of the MC4R and MSTN SNPs of Spinibarbus hollandi
Full size table
Table 2 The allele and genotype frequencies of SNPs
Full size table

Analysis of associations between SNPs and growth traits

The results of association analyses between different SNP genotypes and the five selected growth traits are shown in Table 3. For the MC4R-719A>G SNP, the five growth traits of the GG genotype fish were significantly higher than those of the AA and AG genotype fish. Measurements of the five growth traits of the AG genotype fish were also higher than those of the AA genotype fish, although the differences were not significant. For the MSTN-519C>T SNP, the BWT of the TT genotype fish was significant higher than that of the CC and CT genotype fish, and the BD and BWH of the TT genotype fish were significantly higher than those of CC genotype fish. Measurements of the five growth traits were not significantly different between CC and CT genotype fish.

Table 3 One-way ANOVA analysis of the association between single-nucleotide polymorphisms (SNPs) of the MC4R and MSTN genes and growth traits in Spinibarbus hollandi (mean value ± standard deviation)
Full size table

Association between gene expression levels and growth trait-related SNPs

Expression levels of the MC4R gene in brain were significantly higher in the AA genotype fish than in the AG and GG genotype fish (p < 0.05) and slightly higher in the AG genotype fish than in the GG genotype fish (Fig. 1). The GG genotype fish, which had the highest BWT, had lowest expression of the MC4R gene in the brain. The AA genotype fish, which had the lowest BWT, had the highest expression of MC4R in the brain.

Fig. 1
figure 1

Body weight and MC4R mRNA expression in the brain. Different letters indicate a significant difference (p < 0.05). Black columns represent MC4R mRNA expression levels, and black dashed columns represent the body weight of each genotype. Results are expressed as means ± standard error

Full size image

Body weight and MSTN mRNA expressions in the brain were shown in Fig. 2. No significant difference was detected in the MSTN expression levels in the brain among the different genotype groups of MSTN-519C>T (p > 0.05).

Fig. 2
figure 2

Body weight and MSTN mRNA expression in the brain. Different letters indicate a significant difference (p < 0.05). Black columns represent MSTN mRNA expression levels, and black dashed columns represent the body weight of the different genotypes. Results are expressed as means ± standard error

Full size image

The expression levels of the MSTN gene in muscle were significantly lower in TT genotype fish than in the CC and CT genotype fish (p < 0.05). The BWT of TT genotype fish was significant higher than that of CC and CT genotype fish (p < 0.05), whereas there were no significant differences in BTW and expression levels of the MSTN gene between CC and CT genotype fish (Fig. 3).

Fig. 3
figure 3

Body weight and MSTN mRNA expression in muscle. Different letters indicate a significant difference (p < 0.05). Black columns represent MSTN mRNA expression levels, and black dashed columns represent the body weight of the different genotypes. Results are expressed as means ± standard error

Full size image

Discussion

In this experiment, we detected two mutations (MC4R-719A>G and MSTN-519C>T) in the promoters of the MC4R and MSTN genes of S. hollandi, respectively. Both SNPs were moderately polymorphic loci. MC4R-719A>G was in HWE, whereas MSTN-519C>T deviated from the HWE, it is possible that the MSTN-519C>T, or other variation closely linked to the SNP, was subjected to natural selection pressure. This phenomenon was also found in Argopecten irradians (Meng et al. 2017). The two SNPs are both related with growth traits in this fish. The expression levels of the MC4R gene in the brain are associated with the growth trait-related SNP, MC4R-719A>G, whereas the SNP, MSTN-519C>T, is associated with expression levels of the MSTN gene in muscle, but not in the brain.

Our results showed that the MC4R-719A>G is associated with all five growth traits of S. hollandi examined in the present study, with the GG genotype being predominant among the three genotypes. As an important candidate gene affecting growth ratio, numerous association studies have focused on the detection of MC4R gene polymorphisms associated with economic traits in domestic animals. Zhang et al. (2006) reported a missense mutation in MC4R that was significantly associated with birth weight and average daily gain in cattle. Kim et al. (2000) observed that an SNP, Asp298Asn, located in MC4R was significantly associated with growth traits in pigs, and Song et al. (2012) detected four SNPs in the 3′-UTR of MC4R that were significantly associated with birth weight in sheep. However, there have been few studies that have focused on the association between MC4R SNPs and growth quality traits in fish. Although several SNPs have previously been identified in the MC4R gene of Oreochromis niloticus (tilapia) and Takifugu rubripes, no growth trait-related SNPs were identified in the two species (Liu et al. 2009; Zhang et al. 2012). The growth trait-related SNP identified in the MC4R gene in the present study will therefore make a potentially important contribution to the analysis of molecular markers in fish.

In the present study, we observed that the MC4R-719A>G is associated with expression levels of the MC4R gene in the brain of S. hollandi. The GG allele has the lowest expression level and the GG genotype fish have the most predominant growth traits. These observations are consistent with previous observations indicating that inactivation or decreased activity of MC4R can result in an increase in body weight (Huszar et al. 1997).We hypothesized that an SNP located in the 5′-flanking region might have an impact on MC4R expression and growth performance, by affecting MC4R promoter activity. However, we cannot exclude the possibility that the MC4R SNP identified in the present study is in linkage disequilibrium with an unidentified causal mutation for MC4R expression and growth performance.

In association studies, the MSTN-519C>T was also found to be associated with growth traits of S. hollandi, with the TT genotype being predominant. Significant associations between MSTN polymorphisms and production traits have been widely reported in aquacultural species, including bighead carp (Aristichthys nobilis; Liu et al. 2012), yellow catfish (Pelteobagrus fulvidraco; Zhu et al. 2012), spotted halibut (Verasper variegatus; Li et al. 2012), common carp (Cyprinus carpio; Sun et al. 2012), Atlantic bay scallop (Argopecten irradians; Guo et al. 2011), and gilthead seabream (Sparus aurata; Sánchez-Ramos et al. 2012). Moreover, the MSTN-519C>T departed from Hardy–Weinberg equilibrium, and thus it is possible that this SNP is linked to unidentified genes that are affected by natural selection.

The MSTN-519C>T is associated with expression levels of the MSTN gene in the muscles of S. hollandi. This result indicated that MSTN is a negative regulator of growth in muscle. Knocking down MSTN using antisense morpholinos resulted in the increase of length and width of somites in juvenile zebra fish (Amali et al. 2004). Muscle hyperplasia and hypertrophy were also found in Myostatin dsRNA-microinjected zebrafish (Acosta et al. 2005). No significant association between expression level of the MSTN gene in brain and its SNP was found, suggesting that MSTN expressed in brain may plays another role at the stage of development in S. hollandi. Further investigation of this possibility, including RNA-Seq after MSTN knockdown and other functional studies, will provide important insights into MSTN function. The possibility same as above, the SNP might have a direct impact on MSTN expression, as well as is in linkage disequilibrium with unidentified loci that impact on MSTN expression.

In conclusion, the findings of this study indicate that SNPs located in the MSTN and MC4R promoters have effects on growth-related traits via modifications of gene expression. Information obtained in this study on the MC4R and MSTN SNPs may have potential applications in effective marker-assisted selection to increase body weight and lean meat percentage in S. hollandi.