Identifying Muscle Growth-Related Genes of Mandarin Fish (Siniperca chuatsi) by Transcriptome, and Exploring the Expression in Different Sizes and Compensatory Growth

How organisms display many different biochemical, physiological processes through genes expression and regulatory mechanisms affecting muscle growth is a central issue in growth and development. In Siniperca chuatsi, the growth-related genes and underlying relevant mechanisms are poorly understood, especially for difference of body sizes and compensatory growth performance. The muscle from different sizes of individuals was subjected to transcriptome analysis by RNA-sequencing. Results showed that 8,942 different expression genes (DEGs) were identified after calculating the RPKM. The DEGs involved in GH-IGF pathways, protein synthesis, ribosome synthesis and energy metabolisms, which were expressed at higher level in small individuals (S), whereas in the large fish (L) little expression was found. In repletion feeding and compensatory growth experiments, 8 more significant DEGs were used for further research (GHR2, IGFR1, 4ebp, Mhc, Mlc, Myf6, MyoD, troponin), and this research revealed different temporal and spatial expression patterns of these genes. When food was plentiful, 8 genes participated in and promoted the growth and muscle synthesis, respectively. Starvation can be shown to inhibit the expression of Mhc, Mlc and troponin, and high expression of GHR2, IGFR1 and 4ebp inhibited growth. MyoD can sense and regulate the hunger and work with Mhc and Mlc to accelerate the compensatory growth of S.chuatsi. study candidate


Conclusions
This study is helpful to understand the regulation mechanisms of muscle growth-related genes. The elected genes will contribute to the selective breeding in future as candidate genes. 4 Background Muscle growth is regulated by the genes expression nets of muscle cell proliferation and protein metabolism [1]. In vertebrates, several growth-related genes have been identified, including Myogenic Regulatory Factors (MRFs), insulin-like growth factor (IGF), IGF receptors (IGFR), growth hormone (GH), GH receptors (GHR) [2,3]. Interestingly, for mammals, muscle growth occurs mainly by hypertrophy, with little proliferation. Unlike mammals, fish muscle growth is achieved by hyperplasia and hypertrophy [4].
Furthermore, in large and fast-growing fish, both hyperplasia and hypertrophy contribute to muscle growth and continue into a large body size, whereas small and slow-growing fish largely rely on hypertrophy and the rate of muscle fiber recruitment is rather low [5][6][7].
Thus, the same genes may have different expression and regulatory mechanisms in different fish. Studies on genes expression of different sizes and compensatory growth prove fruitful for comprehending growth regulatory mechanisms of fish [8][9][10][11].
Body size is an obvious and vital characteristic of fish, which is controlled by cell number and cell size [12], and this process is tightly modulated by growth-related genes and nutrition. An adequate re-feeding following a period of starvation or unfavorable environmental conditions can occur accelerated growth, which is called compensatory growth and has been widely studied in vertebrates [10,11,13,14]. Especially, it has been reported that fish has the capable of compensatory growth [14]. In the meantime, transcriptome analysis can identify growth-related genes and expand knowledge in body size and compensatory growth, promoting the enhanced rate of food utilization and cut costs in aquaculture industry.
Transcriptome and RNA-sequencing (RNA-seq) have been applied to a substantial amount of fish biology studies, including zebrafish [15], channel catfish [16], European sea bass [17], and rainbow trout [18]. Many biological processes, including development, immune, stress response, and adaptive evolution can be mapped, annotated and understood by RNA-seq.
S.chuatsi is an important commercial fresh water fish in China. At present, the research on S.chuatsi mainly focuses on aquaculture, disease and immunity, nevertheless the growthrelated genes and mechanisms are not well understood [19][20][21]. To this end, different sizes muscle of S.chuatsi was subjected to transcriptome analysis by RNA-seq and explores the mechanisms of growth-related genes by compensatory growth in this study.
The results would be helpful to find significantly molecule markers for selective breeding.

Sequencing and annotation
The different weight of S.chuatsi was shown in  Analysis of RNA-seq data and confirmation By GO terms enrichment analysis, the biological process, cellular component and and energy metabolism gene (Glucose-6 phosphate dehydrogenase etc) (  RT   Changes in body weight during compensatory growth The SGR in different time intervals was calculated (Fig. 6A) and the weight of fish in two groups was reflected at seven time points and curves were drawn (Fig. 6B). In the first week, the weight of the experimental group decreased markedly following the induction of starvation, as indicated by a negative SGR (-0.98) which reached the lowest (-2.62) in the second week. In the third week, the body weight loss slowed down (SGR − 0.64). During the following week of re-feeding, the weight of the experimental group increased rapidly and resulted in a positive of SGR (2.99), which was significantly higher than (P < 0.01) that in the control group. The elevated SGR that characterizes compensatory growth subsequently declined back to low level during the 2 weeks of realimentation, but the SGR of experimental groups was higher than (P > 0.05) that in the control group (Fig. 6A). The bodyweight difference of experiment and controls ultimately was no significant difference, indicating that the experimental groups achieved complete compensatory growth after refeeding.
Effect of re-feeding to expression of related-growth genes During starvation, the expression of GHR2, IGFR1 and 4ebp was up-regulated, and the difference of expression increased with the increase of starvation time, and the expression was close to that of the control group following re-feeding (Fig. 7). The expression levels of Mhc, Mlc, troponin and MyoD in starvation revealed the descent tendency, however, there was no significant difference in Myf6 expression (Fig. 7).

Discussion
The growth and development of vertebrates are mainly mediated through the GH-IGF system. Especially in teleost fish, the multiple forms of GHRs, and IGFRs have been described. However, they have distinct functions and expression patterns in various fish tissues [22]. In tilapia, GHR2 participates in growth and metabolism and is highly expressed in liver and muscle,while GHR1 expression is higher in liver [23]. In seabream and rainbow trout, GHR2 expression is much higher than GHR1 in numerous tissues such as muscle, pituitary, kidney and gonad [24,25]. Interestingly, although GHR expression is affected by GH in many other fish, the opposite effects are found may due to the different expression levels. The growth responses are enhanced in loaches [26] and GH-transgenic salmon [27] and restrained in tilapia [28] and carp [29]. The weight and thick muscle fibers of GH/GHR-transgenic zebrafish is lower than the other genotypes [30]. In our experiments, however, both the enhancement and inhibition of growth occurred under different conditions, respectively. With the change of time, the expression of GHR2 among different sizes of individuals showed no significant difference (Fig. 5 GHR2), indicating that the stable expression of GHR2 was crucial to promoting growth and development.
During starvation, high expression of GHR on the surface of muscle cells can save protein consumption [23], which also has been reported in gilthead sea bream and rainbow trout [22,31]. Therefore, in S.chuatsi, GHR2 appears to inhibit growth through significant high expression during starvation (Fig. 7 GHR2) and plays a physiologically significant role in muscle metabolism, involving in the mobilization of muscle energy stores.
IGFRs can bind to IGF-1, IGF-2 and insulin, and the affinity for insulin is much lower than for IGFs in all species [32]. Furthermore, in contrast to the situation in mammals, the IGFR1 of fish muscle is greater abundance than insulin receptors (IR) [32], indicating that IGFR1 is more relevant to the regulation of muscle function than IR in fish [33], involving promoting growth, enhancing protein, increasing cell proliferation and reducing protein degradation [34]. IGFR1 was highly expressed in S at 3-month old (Fig. 5 IGFR1) (P < 0.05), suggesting IGFR1 played a more essential role for S over L in muscle, such as participating in and promoting the growth of muscle. Interestingly, prolonged fasting results in significantly reduced IGF-1 mRNA in the liver and muscle of numerous fish species [35][36][37], whereas IGFR1 increased significantly [13,35,38]. In this study, the IGFR1 expression markedly rose during fasting (Fig. 7 IGFR1), illustrating that IGFR1 expression will increase to restrict IGF-I during starvation, which is probably related to the antiapoptotic effects reported in mammals [38].
Phosphorylated of protein kinase B (PKB) is controlled by the metabolic pathway regulated by IGFR1, and PKB can mediate the AKT/mTOR/p70S6K pathway to facilitate protein synthesis and cell growth [39][40][41]. Meanwhile, mTOR can regulate translation by 4ebp [42], and the phosphorylation of 4ebp releases eIF4E to stimulate translation initiation [43]. eIF4E is able to enhance the translation of mRNAs, implicating in cell proliferation and growth [44], but it is hypothesized that overexpression of eIF4E leads to the deregulation of translational and cellular homeostasis [45]. However, the family of 4ebps can inhibite the assembly of eIF4F complex [45]. Thus, 4ebp inhibits cell growth and revert the transformed phenotype of eIF4E-overexpressing cell [45], which may save energy and prevent the malignant transformation of cells during starvation by significant expression (Fig. 7 4ebp). In conclusion, we speculate that the expression of upstream IGFR1 will promote the expression of downstream 4ebp through the above-mentioned pathways in the case of starvation, so as to cope with the food restriction. In compensatory growth experiments, the expression of IGFR1 and 4ebp was consistent ( Fig. 7 IGFR1 and 4ebp), proving the hypothesis, but further research is needed on specific regulatory mechanisms. It has been reported that fasting promotes the metabolic actions of GH rather than the growth-promoting actions [46], however, relevant aspects of researches in GHR2, IGFR1 and 4ebp are less. In the study, the functions of GHR2, IGFR1 and 4ebp also changed from growth promotion when food was abundant to growth inhibition in starvation, thus highlighting the metabolic functions of them.
Mhc expression is significantly related with growth rate [47][48][49], which also affects myofiber hyperplasia and indeterminate growth [50]. In this study, Mhc was highly expressed in S at 5-month old (Fig. 5 Mhc) (P < 0.05), and transcriptome results also showed that the growth rate of small fish was high, suggesting that Mhc can promote the growth of S through its high expression. In larval stage of S.chuatsi, the high expression of Mlc regulates muscle formation and early development [51], which also influences the tail shaft swimming character at posterior area, needing more muscle fiber and protein [52]. The significant high expression of Mlc at 5-month old in S (Fig. 5 Mlc) proved that muscle protein and muscle fiber synthesize quickly to meet physiological requirement. Myosin, the primary protein of muscles, amounts to 50% of the muscle proteins [53].
Troponin as a complex protein is largely expressed in muscle and plays a vital role in muscle contraction [54,55]. During fasting conditions, the markedly down-regulated of Mhc and Mlc indicated that muscle protein synthesis decreased (Fig. 7 Mhc and Mlc), and muscle proteins may be used as the major energy source to maintain basic metabolism which limited growth and was consistent with the observed reduced weight (Fig. 6). In the meantime, because muscle breakdown and muscle protein synthesis were inhibited, muscle contraction and motor function were restrained, which was manifested as a significant decrease in troponin expression (Fig. 7 Troponin). After re-feeding, significant up-regulation of Mhc and Mlc was observed (Fig. 7 Mhc and Mlc), showing that muscle protein was no longer used as the major energy source, and synthesis of muscle and rapid compensatory growth occurred, which resulted in increased weight [56]. As the normal synthesis and growth of muscle, muscle contraction was no longer inhibited and expression levels returned to normal (Fig. 7 Troponin).
MyoD is associated with myogenesis during developmental condition which plays a regulatory role in muscle hypertrophy and muscle mass [57,58]. Myf6 participates in myofiber differentiation by recruiting structural proteins [ 59]. From 3-month old to 5month old, MyoD and Myf6 expression had a downward trend (Fig. 5 MyoD and Myf6), which agreed with the experiment phenomenon of Zhu et al [60].It suggested that 3month old may be a rapid growth stage for S.chuatsi, accompanying with an intense myoblast differentiation, which primarily contributed to the muscle growth [60]. What puzzled us is that both MyoD and Myf6 belong to the MRFs family, but the trends of expression results were opposite in L and S (Fig. 5 MyoD and Myf6), and there was no significant difference, which may be influenced by the complex regulatory systems in vivo.
It has been reported that the reduction in muscle weight may stimulate MRFs transcription [61]. And in mammals, satellite cells can express MyoD and Myf5 in response to muscle damage [62,63]. Interestingly, during the starvation period, the expression of MyoD fell to the lowest level and then increased following the weight loss. After re-feeding, the expression gradually decreased from the highest level, and during the whole process, the expression had a certain delayed (Fig. 7 MyoD). But, the expression of Myf6 was stable throughout the process. These phenomena indicated that MyoD had a sensory and regulatory effects on the weight loss and was conducive to the muscle growth and recovery for S.chuatsi.

Conclusions
In conclusion, numerous DEGs were identified, and several significant DEGs were chosen to explore the expression in different period and compensatory growth. These genes play an important role in muscle growth in S.chuatsi. This study expands our understanding of the mechanism of compensatory growth, and will provide a reference for muscle growthrelated genes in S.chuatsi.

Experimental animals
In this study, all of S.chuatsi came from Foshan Bairong Aquatic Breeding CO., LTD. Three full-sib families A, B and C were constructed by artificial insemination. Each family contained 80 S.chuatsi in a separate tank. The experimental fish were fed under uniform conditions; they were provided adequate live fish as bait twice daily at 0900 and 1800 hours.The water temperature was maintained at 25-26 °C. After feeding 3 months in family A, 5S and 5L was performed for RNA-sEq. For B and C families, 6L and 6S fish were selected from each family at 3-month old and 5-month old for real-time PCR.
Tissue sampling All of the above experimental fish were anesthetized with tricaine methanesulfonate (MS-222, 100 mg/L) and sacrificed via decapitation for subsequent sampling, and muscle was removed immediately. The tissues were frozen in liquid nitrogen quickly, and stored at -80 °C until use. Furthermore, each experiment was performed with three independent biological replicates.

RNA extraction and library construction
Total RNA was extracted from white muscle with E.Z.N.A. total RNA kit II and detected with the concentration and quality. To acquire the entire transcriptome information, and to find out the growth-related DEGs, the muscle samples of family A were mixed with equal amount and then were divided into 2 RNA pools. PolyA mRNA was isolated by Beads with Oligo (dT) after total RNA was collected and interrupted to short fragments. Random hexamer-primer was used to synthesize the first-strand cDNA using the Qiaquick PCR Purification Kit (Qiagen). The second-strand cDNA was synthesized using buffer, dNTPs, RNaseH and DNA polymerase I, respectively (Invitrogen). Subsequently, short fragments were purified, enriched for end reparation and adding polyA, connected with sequencing adapters. After that, the suitable fragments were selected using agarose gel electrophoresis for the PCR amplification as templates. At last, the two cDNA library could be sequenced in BGI-Shenzhen using Illumina HiSeq™ 2000.

Illumina reads processing and assembly
Clean reads were screened from raw reads gained from sequencing machines by removing adaptors, unknown nucleotides larger than 5% and low quality reads (which the percentage of low Q-value ≤ 10 base was more than 20%) which would negatively affect following bioinformatics analysis. Firstly, program-Trinity combines reads with certain length of overlap to form longer fragments without N, which are called contigs. Then, these contigs were taken into further process of sequence cluster with sequence clustering software to form longer sequences without N, which are defined as unigenes.

Functional annotation and classification
The unigenes sequences were firstly aligned with a series of public databases, such as the

RT-PCR analysis
Corresponding primers were designed to vertify the results (

Availability of data and material
All data generated or analysed during this study are included in this published article.

Competing interests
The authors declare that they have no competing interests.   The number of different expression genes that the genes of large fish compared to the genes of small fish.

Figure 3
The GO analysis of different expression genes.