Figures
Abstract
In crustaceans, muscle growth and development is complicated, and to date substantial knowledge gaps exist. In this study, the claw muscle, hepatopancreas and nervous tissue of the mud crab (Scylla paramamosain) were collected at three fattening stages for sequence by the Illumina sequencing. A total of 127.87 Gb clean data with no less than 3.94 Gb generated for each sample and the cycleQ30 percentages were more than 86.13% for all samples. De Bruijn assembly of these clean data produced 94,853 unigenes, thereinto, 50,059 unigenes were found in claw muscle. A total of 121 differentially expressed genes (DEGs) were revealed in claw muscle from the three fattening stages with a Padj value < 0.01, including 63 genes with annotation. Functional annotation and enrichment analysis showed that the DEGs clusters represented the predominant gene catalog with roles in biochemical processes (glycolysis, phosphorylation and regulation of transcription), molecular function (ATP binding, 6-phosphofructokinase activity, and sequence-specific DNA binding) and cellular component (6-phosphofructokinase complex, plasma membrane, and integral component of membrane). qRT-PCR was employed to further validate certain DEGs. Single nucleotide polymorphism (SNP) analysis obtained 159,322, 125,963 and 166,279 potential SNPs from the muscle transcriptome at stage B, stage C and stage D, respectively. In addition, there were sixteen neuropeptide transcripts being predicted in the claw muscle. The present study provides a comprehensive transcriptome of claw muscle of S. paramamosain during fattening, providing a basis for screening the functional genes that may affect muscle growth of S. paramamosain.
Citation: Jiang Q, Bao C, Yang Y, Liu A, Liu F, Huang H, et al. (2017) Transcriptome profiling of claw muscle of the mud crab (Scylla paramamosain) at different fattening stages. PLoS ONE 12(11): e0188067. https://doi.org/10.1371/journal.pone.0188067
Editor: Wan-Xi Yang, Zhejiang University College of Life Sciences, CHINA
Received: June 24, 2017; Accepted: October 31, 2017; Published: November 15, 2017
Copyright: © 2017 Jiang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All transcriptome files are available from the GenBank, National Centre for Biotechnology Information (NCBI) (accession number(s): PRJNA389966).
Funding: This study was supported by the Important Science & Technology Specific Project of Zhejiang Province for Selective Breeding (2016C02055-8) and K. C. Wong Magna Fund in Ningbo University, the Spark Plan Project in Fujian Province (No. 2016S0055) and the Xiamen Southern Oceanographic Center (No. 14GZY56HJ26).
Competing interests: The authors have declared that no competing interests exist.
Introduction
The mud crab Scylla paramamosain belonging to genus Scylla, is widely distributed along inshore region in the southeast China coasts and other Asian countries [1]. Due to the short growth cycle and the large market demand, the mud crab has high commercial value in farming [2]. The aquaculture of this species has emerged more than 100 years in China and more than 30 years in other Asian countries [1, 3]. The lower yields own to many factors, such as growth, nutrition, diseases, and administration patterns [4]. Therefore, research into mechanism underlying the growth and development is of obvious importance.
In crustaceans, the growth demands periodic shedding and replacement of the exoskeleton. Thereinto, the atrophy and restoration of muscle accompanied by each molt process was crucial. Muscle growth is accompanied with the increase in the number and length of muscle fibers, which were related to the molting [5]. Previous studies have shown that the muscle fibers grow in length immediately after molting in fully differentiated lobster [6]. The large myofibril splits in molting, and then grows in the cross-sectional area during the inter-molt [6]. In crabs, it is well known that the body weight increased significantly during the fattening process, nevertheless the molecular mechanism of muscle growth is unclear so far.
The transcriptomes of various tissues have been reported in S. paramamosain, such as hemolymph, gonads, cerebral ganglia, gill and mixture of the muscle, hepatopancreas, eyestalk [4, 7–9], while the transcriptome of muscle is still unclear. This paper reported the results of next-generation sequencing (Illumina sequencing) which pay attention to claw muscle growth during mud crab fattening. Analysis were performed against the sequencing results, including De Bruijn assembly of transcriptome sequences, functional annotation, coding sequence (CDS) prediction, single nucleotide polymorphism (SNP) discovery, genes expression analysis, differential expression analysis. Data obtained in this study facilitate the in-depth understanding of the changes occurring in muscle of S. paramamosain at transcriptomic and molecular level, and could contribute to studies on specific functional genes, the construction of genetic map and identification of molecular markers in this species.
Materials and methods
Sample collection
S. paramamosain (carapace width 7.5–8.4 cm, body weight 80–130 g) were obtained from a local aquafarm, Haicang District, Xiamen, China (24˚31’24.04”N; 118˚03’18.33”E). They were reared in tanks (temperature: 27 ± 2°C; salinity: 26 ± 1 ppm), and fed with the meat of the white Pacific shrimp Litopenaeus vannamei [10]. Crabs at three different fattening stages were sampled, according to the molt cycle of the swimming crab Portunus trituberculatus [11], they attribute to stage B (the first phase of the inter-molt period, the mud crab is very thin, body weight 80–90 g, called “empty” crab.), stage C (the dominant phase of the inter-molt period, body weight 105–120 g, the mud crab is full of meat.) and stage D (the fourth phase of the pre-molt period, body weight 125–130 g, namely double shell crab.). The crabs were dissected after anesthesia on ice for 30 minutes. The claw musculatures, nervous tissue (mixed tissues of cerebral ganglia, eyestalk, and thoracic ganglia) and hepatopancreas were collected, respectively. Three biological replicates were performed in per tissue at three stages, receiving a total of 27 samples. The study does not involve endangered or protected species.
RNA extraction and cDNA library preparation
Total RNA were isolated with the Trizol Reagent (Invitrogen, USA) according to the manufacturer’s instructions. Then the extracted RNA was sequenced using the Illumina HiSeq 2500 by BMK (Beijing Co. Ltd). In short, mRNA poly (A) was separated using oligo (dT) beads. The second-strand cDNA fragments were synthesized using the purified fragments. After screening, the fragments were used for PCR amplification.
Pre-processing and de bruijn assembly
The clean reads were guaranteed by clipping adapter, trimming low-quality reads and removing ambiguous bases. After pre-processing, they were assembled by Trinity using de bruijn algorithm [12]. All clean reads for de bruijn assembly were deposited in GenBank, National Centre for Biotechnology Information (NCBI) under the Accession No. PRJNA389966.
Functional annotation
The unigenes were identified based on sequence similarity with known proteins and using a BLASTX (E≤1e-5) search against the Non-Redundant (NR), Swiss-Prot, Gene Ontology (GO), Clusters of Orthologous Groups (KOG). The KEGG Orthology of each unigene was analyzed using KOBAS 2.0 software [13]. Unigene annotation information was obtained using HMMER (E≤1e-10) search against the Pfam database with forecast of unigene amino acid sequence [14].
Structural and expression analysis
Reliable potential CDS regions from the transcript sequences were identified by TransDecoder software, based on the length of open reading frame (ORF), the Log-likelihood Score, the amino acid sequence alignments with protein domain sequence in Pfam database and other information. Only reliable, STAR mapped reads were considered for SNPs detection. SNPs were called using GATK [15]. SNPs were qualified by continuous single nucleotide mismatch within 35 bp range no more than 3 and the value of standardized serialized SNP greater than 2.0. SNP markers were divided into homozygous- (only one allele) and heterozygous-SNP (two or more alleles). SNPs density in Unigenes was computed. All clean reads were aligned with Unigene database using Bowtie [16], followed performing the estimation of expression level combined RSEM [17]. The fragments per kilobase of exon model per million mapped reads [18] (FPKM) value indicates the corresponding unigene expression abundance.
Identification and validation of differentially expressed genes
Correlation analysis was conducted with Pearson’s Correlation Coefficient (r) [19] to assess the relevance of biological replicates in the same condition. DESeq [20] was performed to analyze the differentially expressed genes (DEGs) between the samples in two conditions. Benjamini-Hochberg was employed to correct the significant value (p-value) for hypothesis test to reduce false positives. False discovery rate (FDR<0.01) and fold change (FC≥2) adopted as the key thresholds in difference expression genetic screening. Do hierarchical clusters to show differential expression patterns of gene sets under different experimental conditions. DEGs annotated in GO annotation database were enrichment analysis using the topGO software. COG statistical classification and KEGG annotation of DEGs were also performed. The significant enrichment of pathways was analyzed using the method of Fisher exact test.
Total RNA from claw muscle of specimens in three stages were extracted as described above. Approximately 2 μg RNA were used to the reverse transcription of cDNA. qRT-PCR was carry out in the 7500 Fast Real-Time PCR (Applied Biosystems) with 2×SYBR Select Master Mix (Applied Biosystems) to validate 10 DEGs (8 in the comparison of stage C and stage D, 2 in the comparison of stage B and stage C) expressed transcripts. Primers were designed using Primer5.0 Tool (PREMIER Biosoft International, Palo Alto, CA) with housekeeping gene 18s RNA as standard gene (Table 1). PCR reactions were performed under the following conditions: 95°C for 30 s, 40 cycles of 95°C for 5 s, 58°C for 30 s and 72°C for 30 s. Six biologic repetitions and three technical repetitions were performed in this study. The ultrapure water was the template in blank control. The 2-ΔΔCt method was used to calculate the gene expression, and Ct values were the mean values of six biologic replicates [21]. The relative expression ratio was represented as mean±SD. All statistical analyses were performed using SPSS 18.0, including Duncan’s multiple range tests and the significance of differences analyzed using one-way ANOVA.
Results
Transcriptome sequencing and read assembly
In this study, 27 cDNA libraries of S. paramamosain were sequenced using Illumina HiSeq 2500 platform. Clean sequencing reads and alignment statistics were showed in Table 2. The cycleQ30 percentage of all samples was more than 86.13%. The assembled transcripts (n = 183,760) had a total size of 291,529,594 bp, an average size of 1,586.47 bp and a N50 assembled transcripts with length of 3,031 bp. Nearly half of (54.56%) assembled transcripts were at the length range of 300–2000 nt (Fig 1).
Functional annotation
In this study, 23,787 transcripts mapped back to the protein database. S1 and S2 Tables show the unigenes and the annotation information. Arthropods account for the largest proportion in homology analysis of S. paramamosain transcriptome. The top three organism were Nevada termite Zootermopsis nevadensis (9.93%), Alveolate Perkinsus marinus (9.77%) and Water flea Daphnia pulex (5.54%) (Fig 2). Table 3 shows the unigenes with highest quality annotations. In addition, 16 transcripts encoding neuropeptide precursors (13 complete and 3 partial) were identified from the transcriptome data (Table 4). The deduced neuropeptides include B-type allatostatin (AST-B), short Neuropeptide F (sNPF), neuroparsin (NP), crustacean hyperglycemic hormone (CHH), orcokinin, diuretic hormone 31 (DH31), tachykinin, myosuppressin, bursicon hormone alpha subunit, putative insulin-like protein growth factor binding protein and insulin-like androgenic gland factor.
The number of aligned genes and the percentage of which in all annotated genes were recorded in brackets.
The analysis of GO terms showed that the 27,917 unigenes were in the GO domains (Fig 3). In biological domain, 19 terms contained 11,924 unigenes. The top three terms were metabolic process (285 unigenes), oxidation-reduction process (223 unigenes) and translation (214 unigenes). In cellular component domain, 4,596 unigenes were found in 16 terms. There were 429 unigenes involved in nucleus, followed by 383 unigenes in integral component of membrane and 376 unigenes in membrane. In molecular function domain, a total of 11,397 unigenes were distributed in 17 terms. The ATP binding (563 unigenes) and catalytic activity (459 unigenes) were top two terms. COG analysis showed that 7,742 transcripts were assigned to 25 COG terms, the top three COG terms of which were general function prediction (2,651 unigenes), replication, recombination and repair (1,096 unigenes) as well as transcription (862 unigenes) (Fig 4). In addition, 10,495 unigenes were classified into 220 KEGG pathways, of which 279 unigenes were in Protein processing in endoplasmic reticulum (ko: 04141), followed by 273 unigenes in oxidative Spliceosome (ko: 03040) and 270 unigenes in Purine metabolism (ko: 00230).
The x-axis indicated GO process; the y-axis on the left side indicated the percentage of the unigenes of this process in all genes; the y-axis on the right side indicated the number of unigenes in the process.
The x-axis indicated COG function; the y-axis indicated the number of unigenes.
Structural and expression analysis
A total of 56,328 CDS were predicted from 94,853 unigenes. CDS was not detected at sequence size 0–100 (bp). The top 3 levels were at sequence size 100–200 (bp), 200–300 (bp) and 300–400 (bp). As length increases, the number decreases. Therefore, the number of glimmer cds dropped significantly at sequence size 2700–2800 (bp), 2800–2900 (bp), 2900–3000 (bp). The length of CDS is shown in S1 Fig.
A total number of 159,322, 125,963 and 166,279 potential SNPs were obtained from stage B, stage C and stage D, including 143,546, 111,625 and 144,943 homoSNP, respectively (Table 5). The distribution of SNP in unigene is showed in Fig 5. SNP-free unigenes accounted for the largest proportion, followed by the unigenes with 0–1 SNP per Kb. As the density of SNPs increases in unigene, the number of unigene decreases. In addition, the correlation analysis has been performed between SNP and 121 DEGs of claw muscle transcriptomes (S5 Table). A total of 177 SNPs were found in 31 DEGs, thereinto, 105 SNPs in 12 annotated genes.
The x-axis indicated SNP number per Kb; the y-axis indicated the number of unigene.
In this study, FPKM value indicates the unigene expression abundance (Fig 6). The results showed that the biologically repeated samples of nervous tissue behaved the highest consistency with the whole of three stages. In addition, the nervous tissue has the highest overall gene expression level, followed by hepatopancreas and muscle, while the dispersion degree of gene expression was opposite. As far as muscle sample, the stage D showed the highest overall gene expression level when compared with stage B and C. However, the reproducibility of muscle samples it is best at stage C. The overall genes expression level of hepatopancreas was slightly large at stage C.
The x-axis indicated the samples; the y-axis indicated the logarithm of FPKM. The figure shows the overall level and the degree of dispersion of gene expression. The horizontal line in the box represents the expression level of 50% genes in the corresponding sample; black spots represent discrete genes, above the box for high expression, below the box for the low expression.
DEGs in the muscle transcriptome
The analysis of muscle transcriptome data revealed 121 DEGs at different stages, 63 (52.07%) of which were annotated successfully (Table 6). In the comparison of stage D and stage B, 2 genes (50% annotated, 50% unannotated) were identified and up-regulated at stage B. In the comparison of stage C and stage B, 23 genes (73.91% annotated, 26.09% unannotated) were identified, including 21 up-regulated genes and 2 down-regulated genes. In the comparison of stage C and stage D, 99 genes (46.46% annotated, 53.54% unannotated) were identified, including 29 up-regulated genes and 70 down-regulated genes. Moreover, 10 DEGs were validated by qRT-PCR, including arthrodial cuticle protein, phosphofructokinase-1 (PFK-1), enolase, arginine kinase (AK), suppressor of hairless [Su (H)], alpha crystallin family (HSPB6), glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and myosin light chain (MLC) in the comparison of stage C and stage D, and major facilitator superfamily, 3-hydroxyanthranilic acid dioxygenase in the comparison of stage B and stage C. The results show that 3 genes (GAPDH, PFK-1, and arthrodial cuticle protein) only expressed differentially with P < 0.05, while the other seven showed consistent with the sequencing results (P < 0.01) (Fig 7).
pfk-1: Phosphofructokinase-1; eno: Enolase; gapdh: Glyceraldehyde-3-phosphatede hydrogenase; Su(H): Suppressor of Hairless; hspb6: alpha crystallin family; 3-hao: 3-hydroxyanthranilic acid dioxygenase; mfs: Major Facilitator Superfamily; ak: Arginine kinase; cuti: arthrodial cuticle protein; mlc: myosin light chain; 18s RNA was used as reference gene. Values denoted by different numbers of asterisks were significantly different when compared by ANOVA (“**” p < 0.01; “*” p < 0.05).
Functional analysis of DEGs of muscle
GO terms of DEGs were analyzed. In the comparison of stage D and stage B, the gene was annotated as calcified cuticle protein like gene and classified into structural molecule activity term (GO: 0042302) in molecular function domain (Fig 8A). In comparison of stage B and C, the DEGs were assigned to 15 terms, of which 6 genes in biological process, 3 genes in cellular component and 9 genes in molecular function. “single-organism process”, “membrane” and “binding” were the top terms in the three GO domains with 4, 2 and 4 genes, respectively (Fig 8B). In comparison of stage D and stage C, 46 genes were GO-categorized into biological process (13 genes), cellular component (8 genes) and molecular function (17 genes). The “cellular process” (10 genes), “cell part” (8 genes) and “binding and catalytic activity terms” (11 genes) were the top terms in the biological process domain, cellular component and molecular function process (Fig 8C). Table 7 shows the GO terms associated with significant DEGs.
(a) The GO categories of the DEGs in stage D specimens and stage B specimens. The x-axis indicated GO process; the y-axis on the left side indicated the percentage of the DEGs; the y-axis on the right side indicated the number of the DEGs (up) and all unigenes (down) in the process; (b) The GO categories of the DEGs in stage C specimens and stage B specimens; (c) The GO categories of the DEGs in stage D specimens and stage C specimens.
COG function classification was performed with significant DEGs. In the comparison of stage D and stage C, a total number of 17 (17.2%) genes were COG-categorized into nine COG domains, 5 genes of which were in “carbohydrate transport and metabolism”, followed by 3 genes in “general function prediction only”, 2 genes in “signal transduction mechanisms” and 2 genes in “amino acid transport and metabolism”, one gene in the remaining domains (Fig 9A). In the comparison of stage B and stage C, 2 genes were found in “general function prediction only”, and 1 gene was in the rest domains, including “Transcription”, “Carbohydrate transport and metabolism”, “Amino acid transport and metabolism”, “Nucleotide transport and metabolism” and “Lipid transport and metabolism” (Fig 9B).
(a) The COG categories of the DEGs in stage D specimens and stage C specimens. The x-axis indicated COG process; the y-axis represented the number of the DEGs; (b) The COG categories of the DEGs of muscle in stage B specimens and stage C specimens.
In comparison of stage B and stage C, 3 genes were classified into cellular processes and metabolism. The genes were assigned to transport and catabolism related to lysosome (ko: 04142), amino acid and nucleotide metabolism related to purine metabolism (ko: 00230) and tryptophan metabolism (ko: 00380), respectively (Fig 10A). In comparison of stage D and stage C, 19 genes were classified into environmental information processing, genetic information processing and metabolism (Fig 10B). In environmental information processing, one gene related to signal transduction was found in Notch signaling pathway (ko: 04330); five folding, sorting and degradation related genes were found in genetic information processing, two of which were assigned to protein processing in endoplasmic reticulum (ko: 04141) and the other were assigned to RNA degradation (ko: 03018). Glycolysis/gluconeogenesis, carbon metabolism and biosynthesis of amino acids were the top three pathway in metabolism (23 genes), with four genes respectively.
(a) The KEGG pathways of the DEGs in stage C specimens and stage B specimens. The x-axis indicated the proportion of annotated DEGs in the pathway. The numbers of genes in the pathway were recorded. The y-axis represented the pathway. The same color-coded pathway was in the same KEGG category; (b) The KEGG pathways of the DEGs of muscle in stage C specimens and stage D specimens.
Discussion
A total of 127.87 Gb clean data consisting of 94,853 unigenes were successfully obtained from the mud crab S. paramamosain at three fattening stages. Previous study has shown 21,791 isotigs in the testis and ovary of the same crab species [8]. In addition, the gills of this crab species challenged with mud crab reovirus was analyzed, producing 67,279 assembled unigenes [9]. The unigenes identified in the study not only enrich transcriptomes data of S. paramamosain, but also provide the basis for studying the mechanisms of muscle growth.
In this study, 23,787 genes could be annotated. Main gene categories were found to be involved in ATP binding and catalytic activities, metabolic and oxidation-reduction processes and nucleus and membrane, which are necessary to maintain the basic life activities. The genes in these categories are relatively conservative [22, 23]. The annotation rate was lower than previous reports in yesso scallop Patinopecten yessoensis and sea cucumber Apostichopus japonicus, whose annotation rate was 27.9% and 39.1%, respectively [24, 25]. The low rate of annotated gene might be due to insufficient of the genomic data of S. paramamosain. The potentially available and unexploited genes, especially DEGs might play an important role in the process of muscle growth.
SNPs were used in trait-mapping and whole-genome association studies owing to the variability and abundance in the genome [4, 26]. SNPs, as potential markers, were even found in the specie without the full genome sequences [27, 28]. In the current study, 451,564 SNPs were called using GATK, with approximately 34-folds increase in comparison with the 13,271 SNPs in the gonads [8]. The putative SNPs could be useful in various fields of aquaculture of S. paramamosain, for example, conservation and population genetics of wild species, mapping of economically important traits, selection and breeding programmes [29]. In addition, a total of 177 SNPs were found in 31 DEGs, including 105 SNPs in 12 annotated genes. Previous study has described that SNPs in actin and CHH could affect the growth in the giant freshwater prawn Macrobrachium rosenbergii [30]. Therefore, SNPs in DEGs are speculated to affect the muscle growth during fattening process.
In crustaceans, the muscle growth and development is very complicated, and many details remain unknown to date. Muscle growth is accompanied with the increase in the number and length of muscle fibers [5]. There is few definite report on increasing the number of muscle fibers. But, once specific muscles have been established during the early stages of development, muscles continue to grow in both length and diameter [31]. In fully differentiated lobster, the muscle fibers grow in length by the addition of sarcomeres [6], and grow in cross-sectional area by the splitting of large myofibrils upon molting and enlargement during the inter-molt [6]. In crayfish Procambarus clarkii, increase in fiber length is accomplished by lengthening of existing sarcomeres [32]. Previous study speculated that the muscle growth could be stimulated by fibers stretching due to the expansion of the new exoskeleton during the post-molt period [33]. However, the mechanism involved in muscle growth is unclear in the mud crab S. paramamosain.
Extensive studies show that muscle growth responds to exercise, nutrient supply, metabolism, cytokines and endocrine factors and denervation [34]. In Drosophila, muscle growth is associated with nutrient sensing by the Insulin/Akt/TOR pathway, which was achieved by the integrated regulation of the transcription factors FOXO, Myc and Mnt [34–36]. The biosynthesis promotes formation of syncytial muscle, which can be stimulated by glycolysis [36]. In addition, glycolysis/pyruvate metabolism has a negative effect on the Notch pathway, and which were regulated by insulin pathway or target of rapamycin pathways [37]. Tixier et al. reported that seven genes (Pglym78, Pfk, Tpi, Gapdh, Pgk, Pyk, and Impl3) involved in glycolysis/pyruvate metabolism, play a vital role in the increasing of muscle fibers size and the myoblast fusion [36]. Some conserved cytokines could activate the pathways associated with muscle growth. For example, Drosophila eiger (CG12919), outstretched, unpaired 2, and unpaired 3, which regulate the muscle development by activation of the TNF signaling cascade and the Eya and JAK-STAT signaling [38, 39]. Octopamine receptor 2 causes muscle hypertrophy by regulating synthesis of cyclic-AMP and PKA [40]. The study on the interconnections between exercise and muscle mass highlighted that spargel promotes mitochondrial activity and is required for sensing the physiological effects caused by exercise [34, 41]. The muscle-nerve interactions play an important role in the formation of muscle size and structure in Drosophila [42]. In addition, a lot of factors were proven to be involved in the atrophy of muscle, such as Drosophila (CG11658, abba, CG10961, Cbl, TER94 and CG6233), p97/VCP ATPase complex, MLCs, actin and so on [43–46].
In this study, the claw muscle transcriptomes were sequenced for screening the functional genes that affect the muscle growth of S. paramamosain. The predominant catalogs of DEGs clusters were regulation of transcription, ATP binding and integral component of membrane. The results suggested that abundant DEGs involved in maintaining the basic life activities during different fattening stages. The enzymes, related to glycogen metabolism (enolase, glycogen phosphorylase) and protein metabolism (arginine kinase, aminopeptidase N), were up-regulated at stage C specimens compared to stage D specimens, which may take part in maintaining metabolic balance during fattening process. Furthermore, several DEGs have been found, which play roles in skeletal muscle differentiation, atrophy and transformation in Drosophila or mice, such as Pfk-1, Gapdh, Hspb6, Su (H), Sspn and Mlc [36, 46–49].
PFK-1 and GAPDH are the basic enzymes in glycolysis. Previous study showed that PFK-1 is regulated by signals from cell proliferation [50, 51]. In addition, PFK-1 regulates the muscle fatigue by interaction with neuronal nitric oxide synthase [52]. PFK-1 and GAPDH play important roles in the increasing of muscle fibers size and the myoblast fusion in Drosophila [36]. In this study, they were found in claw muscle transcriptomes and up-regulated at stage C specimens compared to stage D specimens. The results suggest that they could participate in the muscle growth. As it is known that the muscle fibers rely mainly on glycolysis [53], the high expression of PFK-1 and GAPDH also may be associated with higher energy metabolism at stage C specimens.In D. melanogaster, the notch signalling pathway can regulate the differentiation of muscle [54]. Su (H) functions as a transcriptional repressor, has been shown to participate in the notch signaling pathway and the myogenesis [46]. Myogenesis in crustaceans may share some similarities with Drosophila. Su (H) was found with higher expression in stage C3/4 specimens than in stage D3 specimens in muscle transcriptomic database, and it is likely to be involved in the myogenesis.
HSPB6, as a member of the small heat shock protein family, was constitutively expressed in skeletal muscle [55]. HSPB6 not only plays a role in prevention of atrophy, ischemia, hypertensive stress, and metabolic dysfunction, but also regulate muscle contraction by binding to the troponin complex [47, 56]. In the present study, twelve HSPB6 transcripts were up-regulated expression in claw muscle under stage C3/4 specimens compared to the stage D3 specimens. The claw muscular atrophy occurred mainly at the stage D3. The high expression of these genes suggest that HSPB6 could be involved in the muscle protection against atrophy by stabilizing myofibrillar proteins at stage C3/4.
In mice, Sarcospan (SSPN) plays important roles in muscular dystrophy and muscle force development by linked to sarcoglycans or as a component of the dystrophin-glycoprotein complex [48, 57, 58, 59]. SSPN was discovered in the claw muscle transcriptome that has not been identified previously in S. paramamosain. According to the function of SSPN reported in mice, SSPN may be involved in muscle atrophy of S. paramamosain. MLCs, as the myofibrillar isoform, expressed to varying degrees in different fiber types in crustacean muscle. And the contractile apparatus transformation is accompanied by change in fiber types, which can significantly affect muscle size [49]. In addition, MLCs has been found to play a key role in atrophy, which can regulate the extraction effect of ubiquitinated proteins from the myofibrils [44]. In this study, MLCs expression may be consistent with the contractile apparatus transformation.
In addition, 34 neuropeptides were found in the transcriptome data of claw muscle in S. paramamosain. Some of the neuropeptides (e.g., the AST-B, sNPF, NP, CHH and orcokinin, DH31) had been confirmed to be widely expressed in various tissues (e.g., the cerebral ganglia and ovary) of this crab species [7]. Neuropeptides may be secreted by neuroendocrine in the nervous system and autocrines/paracrines in the non-nervous system, which could regulate many physiological processes, including growth, locomotion, reproduction and metabolism [7, 60, 61]. The neuropeptides in claw muscle might function as neurotransmitters or neuromodulators to regulate muscle growth.
Conclusions
The first transcriptome analysis on the claw muscle of S. paramamosain was carried out successfully and yielded 35.46 Gb clean data. Data obtained in present study greatly contributes to the understanding of the gene expression and genome structure occurring within the claw muscle of S. paramamosain at different molting stages. Potential SNPs found in the transcriptome are useful for future selective breeding, trait-mapping, and gene localization studies. The discovery and validation of DEGs showed that these particular genes might be conducive to claw muscle atrophy or restoration in S. paramamosain.
Supporting information
S1 Table. Sequences of unigenes in S. paramamosain.
https://doi.org/10.1371/journal.pone.0188067.s001
(ZIP)
S2 Table. Unigene annotation in S. paramamosain.
https://doi.org/10.1371/journal.pone.0188067.s002
(XLS)
S3 Table. The SSRs identified in unigenes in S. paramamosain.
https://doi.org/10.1371/journal.pone.0188067.s003
(XLS)
S4 Table. The DEGs of muscle in S. paramamosain.
https://doi.org/10.1371/journal.pone.0188067.s004
(XLS)
S5 Table. The SNPs in DEGs of muscle in S. paramamosain.
https://doi.org/10.1371/journal.pone.0188067.s005
(XLSX)
S1 Fig. The length distribution of cds in S. paramamosain.
The x-axis indicated the length of cds; the y-axis indicated the number of cds.
https://doi.org/10.1371/journal.pone.0188067.s006
(TIF)
S2 Fig. Analysis of DEGs of muscle in the stage B specimens compared to the stage D specimens.
The different color dots indicate the significant DEGs (p < 0.01). Red dots indicate the DEGs with log2fold change greater than 2 (up-regulated in post-molt specimens) and green dots indicate genes with log2fold change less than -2 (down-regulated in stage B specimens).
https://doi.org/10.1371/journal.pone.0188067.s007
(TIF)
S3 Fig. The DEGs of muscle in the stage B specimens compared to the stage C specimens.
Red dots indicate the DEGs up-regulated in post-molt specimens and green dots indicate DEGs down-regulated.
https://doi.org/10.1371/journal.pone.0188067.s008
(TIF)
S4 Fig. The DEGs of muscle in the stage D specimens compared to the stage C specimens.
Red dots indicate the DEGs up-regulated in stage D specimens and green dots indicate the DEGs down-regulated.
https://doi.org/10.1371/journal.pone.0188067.s009
(TIF)
References
- 1. Ye HH, Tao Y, Wang GZ, Lin QW, Chen XL, Li SJ. Experimental nursery culture of the mud crab Scylla paramamosain (Estampador) in China. Aquaculture International. 2011; 19:313–321.
- 2. Shen Y, Lai Q. Present status of mangrove crab (Scylla serrata (Forskal) culture in China. Naga. 1994; 17:28–29.
- 3.
Keenan CP, Blackshaw A. Mud crab aquaculture and biology. ACAIR Proceedings. 1999. 78.
- 4. Ma H, Ma C, Li S, Wei J, Li X, Liu Y, et al. Transcriptome analysis of the mud Crab (Scylla paramamosain) by 454 deep sequencing: assembly, annotation, and marker discovery. Plos One. 2014; 9: e102668. pmid:25054331
- 5. Skinner DM. Breakdown and reformation of somatic muscle during the molt cycle of the land crab, Gecarcinus lateralis. J Exp Zool. 1966; 163(2):115. pmid:5972669
- 6. Haj AJE, Govind CK, Houlihan DF. Growth of lobster leg muscle fibers over intermolt and molt. J Crustacean Biol. 1984; 4(4):536–545.
- 7. Bao C, Yang Y, Huang H, Ye H. Neuropeptides in the cerebral ganglia of the mud crab, Scylla paramamosain: transcriptomic analysis and expression profiles during vitellogenesis. Sci. Rep. 2015; 5: 17055. pmid:26592767
- 8. Gao J, Wang X, Zou Z, Jia X, Wang Y, Zhang Z. Transcriptome analysis of the differences in gene expression between testis and ovary in green mud crab (Scylla paramamosain). BMC Genomics. 2014; 15: 585. pmid:25015001
- 9. Liu S, Chen G, Xu H, Zou W, Yan W, Wang Q, et al. Transcriptome analysis of mud crab (Scylla paramamosain) gills in response to mud crab reovirus (MCRV). Fish Shellfish Immunol. 2017; 60: 545–553. pmid:27492124
- 10. Huang X, Ye H, Huang H, Yang Y, Gong J. An insulin-like androgenic gland hormone gene in the mud crab, Scylla paramamosain, extensively expressed and involved in the processes of growth and female reproduction. Gen. Comp. Endocrinol. 2014; 204: 229–238. pmid:24929228
- 11. Jie S, Zhu DF, Hu ZH, Qi YZ, Wang CJ. Molt staging in the swimming crab Portunus trituberculatus. J. Fish. China. 2011; 35: 1481–1487.
- 12. Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, et al. Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat. Biotechnol. 2011; 29:644–652. pmid:21572440
- 13. Xie C, Mao X, Huang J, Ding Y, Wu J, Dong S, et al. KOBAS 2.0: a web server for annotation and identification of enriched pathways and diseases. Nucleic Acids Res. 2011; 39: 316–322.
- 14. Finn RD, Bateman A, Clements J, Coggill P, Eberhardt RY, Eddy SR, et al. Pfam: the protein families database. Nucleic Acids Res. 2014; 42: D222. pmid:24288371
- 15. Bauer DC. Variant calling comparison CASAVA1.8 and GATK. Nature Precedings. 2011.
- 16. Langmead B. Aligning short sequencing reads with Bowtie. Curr Protoc Bioinformatics. 2010; 14: 11–7.
- 17. Li B, Dewey CN. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinf. 2011; 12: 323.
- 18. Trapnell C, Williams BA, Pertea G, Mortazavi A, Kwan G, van Baren MJ, et al. Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat. Biotechnol. 2010; 28: 511–515. pmid:20436464
- 19. Garber M, Grabherr MG, Guttman M, Trapnell C. Computational methods for transcriptome annotation and quantification using RNA-seq. Nat. Methods. 2011; 8: 469. pmid:21623353
- 20. Anders S, Huber W. Differential expression analysis for sequence count data. Genome Biol. 2010; 11: R106. pmid:20979621
- 21. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001; 25: 402–408. pmid:11846609
- 22. Ghaffari N, Sanchezflores A, Doan R, Garciaorozco KD, Chen PL, Ochoaleyva A, et al. Novel transcriptome assembly and improved annotation of the whiteleg shrimp (Litopenaeus vannamei), a dominant crustacean in global seafood mariculture. Sci. Rep. 2014; 4: 7081. pmid:25420880
- 23. Tagmount A, Wang M, Lindquist E, Tanaka Y, Teranishi KS, Sunagawa S, et al. The porcelain crab transcriptome and PCAD, the porcelain crab microarray and sequence database. Plos One. 2010; 5: e9327. pmid:20174471
- 24. Hou R, Bao Z, Wang S, Su H, Li Y, Du H, et al. Transcriptome sequencing and de novo analysis for yesso scallop (patinopecten yessoensis) using 454 gs flx. Plos One. 2011; 6: e21560. pmid:21720557
- 25. Du H, Bao Z, Hou R, Wang S, Su H, Yan J, et al. Transcriptome sequencing and characterization for the sea cucumber Apostichopus japonicus (Selenka, 1867). Plos One. 2012; 7: e33311. pmid:22428017
- 26. Hui L, Xiao S, Hua Y, Zhang Z, Lv C, Zheng S, et al. Identification of immune-related genes and development of SSR/SNP markers from the spleen transcriptome of Schizothorax prenanti. Plos One. 2016; 11: e0152572. pmid:27019203
- 27. Xu J, Ji P, Zhao Z, Zhang Y, Feng J, Wang J, et al. Genome-wide SNP discovery from transcriptome of four common carp strains. Plos One. 2012; 7: e48140. pmid:23110192
- 28. Salem M, Vallejo RL, Leeds TD, Palti Y, Liu S, Sabbagh A, et al. RNA-Seq identifies SNP markers for growth traits in rainbow trout. Plos One. 2012; 7: e36264. pmid:22574143
- 29. Waiho K, Fazhan H, Shahreza MS, Moh JH, Noorbaiduri S, Wong LL, et al. Transcriptome Analysis and Differential Gene Expression on the Testis of Orange Mud Crab, Scylla olivacea, during Sexual Maturation. Plos One. 2017; 12(1):e0171095. pmid:28135340
- 30. Thanh NM, Barnes AC, Mather PB, Li Y, Lyons RE. Single nucleotide polymorphisms in the actin and crustacean hyperglycemic hormone genes and their correlation with individual growth performance in giant freshwater prawn Macrobrachium rosenbergii. Aquaculture. 2010; 301: 7–15.
- 31.
Watling L, Thiel M. The natural history of the crustacea. Oxford University Press Home Page. 2015.
- 32. Bittner GD, Traut DL. Growth of crustacean muscles and muscle fibers. J Comp Physiol. 1978; 124(3):277–285.
- 33. Mykles DL. The mechanism of fluid absorption at ecdysis in the American lobster Homarus americanus. J. Exp. Biol. 1980; 84: 89–102.
- 34. Tinkerhess MJ, Healy L, Morgan M, Sujkowski A, Matthys E, Zheng L, et al. The Drosophila PGC-1α homolog spargel modulates the physiological effects of endurance exercise. Plos One. 2012; 7(2):e31633. pmid:22348115
- 35.
Puppa MJ, Demontis F. Skeletal Muscle Homeostasis and Aging in Drosophila. Life Extension. Springer International Publishing, 2015:107–126.
- 36. Tixier V, Bataillé L, Etard C, Jagla T, Weger M, Daponte JP, et al. Glycolysis supports embryonic muscle growth by promoting myoblast fusion. Proc Natl Acad Sci U S A. 2013; 110(47):18982–18987. pmid:24191061
- 37. Saj A, Arziman Z, Stempfle D, Van BW, Sauder U, Horn T, et al. A combined ex vivo and in vivo RNAi screen for notch regulators in Drosophila reveals an extensive notch interaction network. Dev Cell. 2010; 18(5):862. pmid:20493818
- 38. Moreno E, Yan M, Basler K. Evolution of TNF signaling mechanisms: JNK-dependent apoptosis triggered by Eiger, the Drosophila homolog of the TNF superfamily. Curr Biol. 2002; 12(14):1263. pmid:12176339
- 39. Liu Y, Jakobsen JG, Amarantos I, Gilmour D, Furlong E. A systematic analysis of Tinman function reveals Eya and JAK-STAT signaling as essential regulators of muscle development. Dev Cell. 2009; 16(2):280. pmid:19217429
- 40. Navegantes LC, Resano NM, Migliorini RH, Kettelhut IC. Catecholamines inhibit Ca(2+)-dependent proteolysis in rat skeletal muscle through beta(2)-adrenoceptors and cAMP. Am J Physiol Endocrinol Metab. 2001; 281(3):E449. pmid:11500299
- 41. Tiefenböck SK, Baltzer C, Egli NA, Frei C. The Drosophila PGC-1 homologue Spargel coordinates mitochondrial activity to insulin signalling. Embo Journal. 2010; 29(1):171–183. pmid:19910925
- 42. Fernandes JJ, Keshishian H. Nerve-muscle interactions during flight muscle development in Drosophila. Dev. 1998; 125(9):1769.
- 43. Piccirillo R, Demontis F, Perrimon N, Goldberg AL. Mechanisms of muscle growth and atrophy in mammals and Drosophila. Dev Dyn. 2013. 243(2):201–215. pmid:24038488
- 44. Cohen S, Zhai B, Gygi SP, Goldberg AL. Ubiquitylation by Trim32 causes coupled loss of desmin, Z-bands, and thin filaments in muscle atrophy. J Cell Biol. 2012; 198(4):575. pmid:22908310
- 45. Paul PK, Kumar A. Targeted ablation of TRAF6 inhibits skeletal muscle wasting in mice. J Cell Biol. 2010; 191(7):1395–411. pmid:21187332
- 46. Duan H, Skeath JB, Nguyen HT. Drosophila Lame duck, a novel member of the Gli superfamily, acts as a key regulator of myogenesis by controlling fusion-competent myoblast development. Development. 2001; 128: 4489–4500. pmid:11714674
- 47. Dreiza CM, Komalavilas P, Furnish EJ, Flynn CR, Sheller MR, Smoke CC, et al. The small heat shock protein, HSPB6, in muscle function and disease. Cell Stress Chaperones. 2010; 15: 1–11. pmid:19568960
- 48. Crosbie RH, Lim LE, Moore SA, Hirano M, Hays AP, Maybaum SW, et al. Molecular and genetic characterization of sarcospan: insights into sarcoglycan-sarcospan interactions. Hum. Mol. Genet. 2000; 9: 2019–2027. pmid:10942431
- 49. Cohen S, Brault JJ, Gygi SP, Glass DJ, Valenzuela DM, Gartner C, et al. During muscle atrophy, thick, but not thin, filament components are degraded by MuRF1-dependent ubiquitylation. J. Cell. Biol. 2009; 185: 1083–1095. pmid:19506036
- 50. Yalcin A, Telang S, Clem B, Chesney J. Regulation of glucose metabolism by 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatases in cancer. Exp. Mol. Pathol. 2009; 86: 174–179. pmid:19454274
- 51.
Taylor MV. Comparison of Muscle Development in Drosophila, and Vertebrates. Muscle Development in Drosophila. Springer New York, 2006:169–203.
- 52. Fengyun Zhang, Xiaodan Qian, Cheng Yuhui, et al. Phosphofructokinase-1 Negatively Regulates Neurogenesis from Neural Stem Cells. Neurosci Bull. 2016; 32(3):205–216. pmid:27146165
- 53. Yi W, Clark PM, Mason DE, Keenan MC, Hill C, Goddard WA, et al. Phosphofructokinase 1 glycosylation regulates cell growth and metabolism. Science. 2012; 337: 975–980. pmid:22923583
- 54. Portin P. General outlines of the molecular genetics of the Notch signalling pathway in Drosophila melanogaster: a review. Hereditas. 2002; 136: 89–96. pmid:12369105
- 55. Golenhofen N, Perng MD, Quinlan RA, Drenckhahn D. Comparison of the small heat shock proteins alphaB-crystallin, MKBP, HSP25, HSP20, and cvHSP in heart and skeletal muscle. Histochem. Cell Biol. 2000; 122: 415–425.
- 56. Rembold CM, Foster DB, Strauss JD, Wingard CJ, Eyk JE. cGMP-mediated phosphorylation of heat shock protein 20 may cause smooth muscle relaxation without myosin light chain dephosphorylation in swine carotid artery. J Physiol. 2000; 524: 865–878. pmid:10790164
- 57. Hayashi K, Wakayama Y, Inoue M, Kojima H, Shibuya S, Jimi T, et al. Sarcospan: ultrastructural localization and its relation to the sarcoglycan subcomplex. Micron. 2006; 37: 591–596. pmid:16442802
- 58. Lebakken CS, Venzke DP, Hrstka RF, Consolino CM, Faulkner JA, Williamson RA, et al. Sarcospan-deficient mice maintain normal muscle function. Mol. Cell. Biol. 2000; 20: 1669–1677. pmid:10669744
- 59. Marshall JL, Chou E, Oh J, Kwok A, Burkin DJ, Crosbiewatson RH. Dystrophin and utrophin expression require sarcospan: loss of α7 integrin exacerbates a newly discovered muscle phenotype in sarcospan-null mice. Hum. Mol. Genet. 2012; 21: 4378–4393. pmid:22798625
- 60. Grimmelikhuijzen CJP, Leviev I, Carstensen K. Peptides in the nervous systems of cnidarians: structure, function, and biosynthesis. Int. Rev. Cytol. 1996; 167: 37–89. pmid:8768492
- 61. Hartenstein V. The neuroendocrine system of invertebrates: a developmental and evolutionary perspective. J. Endocrinol. 2006; 190: 555–570. pmid:17003257