Comparative Transcriptome Analysis on the Regulatory Mechanism of Thoracic Ganglia in Eriocheir sinensis at Post-Molt and Inter-Molt Stages

Eriocheir sinensis is an aquatic species found distributed worldwide. It is found in the Yangtze River of China, where the commercial fishing of this valuable catadromous aquatic species has been banned. As an important member of the phylum Arthropoda, E. sinensis grows by molting over its whole lifespan. The central nervous system of Eriocheir sinensis plays an important regulatory role in molting growth. Nevertheless, there are no reports on the regulatory mechanisms of the nervous system in E. sinensis during the molting cycle. In this study, a comparative transcriptome analysis of E. sinensis thoracic ganglia at post-molt and inter-molt stages was carried out for the first time to reveal the key regulatory pathways and functional genes operating at the post-molt stage. The results indicate that pathways and regulatory genes related to carapace development, tissue regeneration, glycolysis and lipolysis and immune and anti-stress responses were significantly differentially expressed at the post-molt stage. The results of this study lay a theoretical foundation for research on the regulatory network of the E. sinensis nervous system during the post-molt developmental period. Detailed knowledge of the regulatory network involved in E. sinensis molting can be used as a basis for breeding improved E. sinensis species, recovery of the wild E. sinensis population and prosperous development of the E. sinensis artificial breeding industry.


Introduction
Eriocheir sinensis (Phylum, Arthropod) is a valuable catadromous species found in the Yangtze River of China, the third-largest river in the world. It is distributed worldwide, including in America and Europe, as a result of its strong osmotic adjustment and movement capabilities [1]. The wild E. sinensis population has suffered from serious devastation caused by environmental pollution, overconstruction of water conservancy projects and overfishing, among other examples. The Ministry of Agriculture and Rural Affairs of the People's Republic of China has prohibited commercial fishing of wild E. sinensis. The restoration of wild E. sinensis resources is being vigorously carried out [2]. In addition, E. sinensis is a delicacy and full of nutrients. As an important economic aquatic species, the E. sinensis industry has become the pillar of the aquaculture industry [3,4]. Revealing the regulatory mechanisms underlying E. sinensis development will lay a theoretical foundation for breeding and releasing juvenile improved E. sinensis, thereby promoting the restoration of wild E. sinensis resources and the sustainable development of the E. sinensis breeding industry.
As an important member of the phylum Arthropoda, E. sinensis grows by molting over its whole lifespan. The molting cycle can be divided into four stages according to the morphological characteristics of the setae: post-molt (AB), inter-molt (C), pre-molt (D) and molt (E). During the post-molt stage, water is quickly absorbed, and the exoskeleton gradually hardens. At the inter-molt stage, the exoskeleton continues to harden and mineralize, and the muscle gradually enlarges. At the pre-molt stage, the old skeleton decomposes and is absorbed, and a new skeleton gradually forms [5].
The basic regulatory mechanism of the E. sinensis molting process has been revealed; that is, in general, the initiation and termination of molting are coordinated by ecdysone secreted by the Y organ and the molting inhibition hormone (MIH) secreted by the Xorgan/sinus gland complex in the eyestalk [6]. In recent years, research on the regulation of E. sinensis molting has been extended, mainly regarding the influence and regulatory mechanisms of feed nutrients (gamma-aminobutyric acid, L-tryptophan), culture patterns (stocking density, salinity) and key functional genes (MMP-14 and V-ATPase subunit B) in molting, growth and development [7][8][9][10][11][12]. The results indicated that gamma-aminobutyric acid supplementation can increase food intake. L-tryptophan supplementation plays a positive regulatory role in molting. The optimal stocking density for juvenile and adult E. sinensis was 42-85 crabs per m 2 . Low salinity can influence the survival of E. sinensis, and high salinity can restrain its embryo development. MMP-14 functions during E. sinensis immune response. V-ATPase subunit B plays a regulatory role in osmoregulation and is essential for E. sinensis molting [7][8][9][10][11][12]. The central nervous system is an important organ in the regulation of molting, and the neuropeptides synthesized and secreted by the central nervous system of crabs play an important regulatory role in many physiological processes, including ovary maturation and molting growth [13,14]. Nevertheless, there are no reports on the regulatory mechanisms of the nervous system in the E. sinensis molting cycle.
As mentioned, research on the mechanisms regulating E. sinensis molting is still at the initial stage. High-throughput sequencing is an efficient and essential technique for preliminarily revealing key regulatory pathways and genes involved in this process. Therefore, in this study, a comparative transcriptome analysis of E. sinensis thoracic ganglia at post-molt and inter-molt stages was carried out for the first time to reveal key regulatory pathways and functional genes at the post-molt stage. The results form a theoretical foundation for research on the regulatory network of the E. sinensis nervous system during the post-molt developmental period, and such details can be used toward breeding improved E. sinensis species, the recovery of the wild E. sinensis population and the prosperity of the E. sinensis artificial breeding industry.

Experimental Crabs and Sample Collection
One-year-old juvenile E. sinensis (average body weight was 12.5 ± 0.79 g) crabs were supplied by Jiangs' Noah's Ark Agricultural Science and Technology Co., Ltd. (Changzhou, China). Individuals of a similar size and at the same developmental stage were selected and cultured in three aquariums. Ten female E. sinensis crabs and the same number of male juveniles were cultured in the same aquarium. The aquariums were continuously aerated, and the water quality was monitored every day, including water temperature, pH, the concentration of dissolved oxygen, NH 3 -N and NO 2 . E. sinensis were fed with compound feed twice each day (at 14:00 and 17:00). The molting stage was determined according to the report from Kang et al. [5]. Cameras were installed in each aquarium to monitor molting, and the molting process was observed 24 h each day after ingestion of E. sinensis was reduced. The thoracic ganglia were collected within half an hour after molting; one male sample and one female sample at the post-molt stage were collected from each tank. Likewise, we also collected the same number of thoracic ganglia of E. sinensis at the intermolt phase. Body size parameters of sampled E. sinensis were measured before sampling.

Total RNA Extraction and ILLUMINA Sequencing
According to the manufacturer's instructions, total RNA was extracted with RNAiso reagent (TaKaRa, Kusatsu, Japan). Then, equal amounts of total RNA from the thoracic ganglia of one female and one male crab at the same developmental stage in each tank were pooled as one sample. Three samples for the post-molt stage (MP) and three samples for the inter-molt stage (MI) were finally obtained. The RNA samples were checked for quality, and the detailed operation for quantification of extracted total RNA, construction of the cDNA library and high-throughput sequencing was performed according to the methods reported in our previous study [15]. The generated raw data were submitted to NCBI (NCBI, Bethesda, MD, USA) with accession number PRJNA822878.

Data Filtering and Assembly
The raw data were filtered using NGS QC TOOLKIT V2.3.3 software (Roche, NY, USA) and analyzed, and sequences representing low-quality reads, contaminated reads, primers and adapters were removed [16]. The filtered clean data were assembled using Trinity software (v2.2.0, Singapore) [17].

Differential Expression Analysis
Differential expression analysis was carried out using the DESeq R package (1.18.0) [21]. Fold change was calculated as the ratio of the expression level of genes in the MI sample and MP sample. In addition, |log 2 fold change| > 1 and padj < 0.05 (adjusted p value) were set as the cutoff thresholds for differentially expressed genes (DEGs). The detailed method for the differential expression analysis was performed according to our previous study [15]. GO and KEGG enrichment analyses were carried out on DEGs (padj < 0.05). Finally, we obtained the top 30 GO terms and top 30 KEGG pathways. The methods are detailed in our previous study [15].

Quantitative Real-Time PCR (qPCR) Validation
The accuracy of high-throughput data was validated using qPCR. Ten DEGs were randomly selected from transcriptome data, and a qPCR experiment was performed on an ABI 7500 real-time PCR system (ABI, Waltham, MA, USA). The primers were designed with Primer Premier 6 software. Beta-actin was used as the internal reference, and the amplifications were performed according to the following program: 95 • C for 30 s and 40 cycles of 95 • C for 5 s, 60 • C for 35 s and 72 • C for 52 s. Sample detection was triplicated, and the gene expression levels were calculated using the 2 −∆∆CT method [22]. Statistical significance (p < 0.05) was calculated using one-way ANOVA and Duncan's multiple range tests (SPSS 21.0). The minimum significance level was set to 0.05.

Sequencing and Assembly of Thoracic Ganglia Transcriptome of E. sinensis
The body size parameters for E. sinensis collected at post-molt and inter-molt stages are shown in Table 1. As shown in Table 2, a total of 290,941,648 clean data were generated. Phred quality score was used as an index for the base-calling accuracy and calculated using FastQC software (Babraham, UK) v0.10.1. In this study, a Q30 value larger than 93% indicated that the base-calling accuracy for each replicate had reached 99.9% and met the requirement for further analysis. After assembly, we obtained 40,121 unigenes. Among these, 22,198 unigenes were longer than 500 bp, the max length was longer than 13,628 bp, the average length was 921.66 b, and N50 was 1209 bp.  Note: MP1-3: three replicates of thoracic ganglia of post-molt E. sinensis; MI1-3: three replicates of thoracic ganglia of inter-molt E. sinensis; valid bases: valid base ratio; Q20: ratio of bases with Phred quality score larger than 20 in raw bases.

Top 30 GO Enrichment Analysis on DEGs at Post-Molt and Inter-Molt Stages
GO defines three levels of ontologies, including molecular function (MC), biological process (BP) and cellular component (CC). As shown in Figure 1, DEGs were mostly enriched in numerous terms in the subcategory of biological process, including development regulation ('regulation of neuron death', 'fat pad development', 'seminiferous tubule development', 'positive regulation of intrinsic apoptotic signaling pathway by p53 class mediator', 'hypothalamus gonadotrophin-releasing hormone neuron development'), energy homeostasis and anti-stress response ('energy homeostasis', 'mitochondrion transport along microtubule', 'regulation of mitochondrial membrane potential') and the regulation of proteometabolism ('regulation of proteasomal protein catabolic process', 'modificationdependent protein catabolic process', 'positive regulation of protein monoubiquitination').

Top 30 KEGG Enrichment Analysis
As shown in Figure 2, the functions of the identified DEGs were mainly associated with six categories: organismal systems, metabolism, human diseases, genetic information processing, environmental information processing and cellular processes. The top 30 KEGG pathways were mainly relevant to three subcategories, namely the regulation of immune response, energy metabolism (mainly related to glycolysis and lipolysis) and neuronal signal transduction. The major pathways relevant to immune response were 'antigen processing and presentation', 'IL-17 signaling pathway', 'Th17 cell differentiation' and 'lysosome'. Energy metabolism pathways mainly involved the 'PPAR signaling pathway', 'N-glycan biosynthesis' and 'mucin-type O-glycan biosynthesis'. The main pathways related to signal transduction mainly involved the 'tight junction', 'gap junction', 'MAPK signaling pathway', 'Wnt signaling pathway' and 'endocytosis'. of proteometabolism ('regulation of proteasomal protein catabolic process', 'modificationdependent protein catabolic process', 'positive regulation of protein monoubiquitination').

Top 30 KEGG Enrichment Analysis
As shown in Figure 2, the functions of the identified DEGs were mainly associated with six categories: organismal systems, metabolism, human diseases, genetic information processing, environmental information processing and cellular processes. The top 30 KEGG pathways were mainly relevant to three subcategories, namely the regulation of immune response, energy metabolism (mainly related to glycolysis and lipolysis) and neuronal signal transduction. The major pathways relevant to immune response were 'antigen processing and presentation', 'IL-17 signaling pathway', 'Th17 cell differentiation' and 'lysosome'. Energy metabolism pathways mainly involved the 'PPAR signaling pathway', 'N-glycan biosynthesis' and 'mucin-type O-glycan biosynthesis'. The main pathways related to signal transduction mainly involved the 'tight junction', 'gap junction', 'MAPK signaling pathway', 'Wnt signaling pathway' and 'endocytosis'.

Analysis of Functional DEGs
The top 30 GO terms and top 30 KEGG pathways were comprehensively analyzed, and the key DEGs can be classified into four categories: carapace development and tissue regeneration, neuronal signal transduction, energy metabolism and homeostasis maintenance and immune and anti-stress response. The key functional DEGs are shown in Table

Analysis of Functional DEGs
The top 30 GO terms and top 30 KEGG pathways were comprehensively analyzed, and the key DEGs can be classified into four categories: carapace development and tissue regeneration, neuronal signal transduction, energy metabolism and homeostasis maintenance and immune and anti-stress response. The key functional DEGs are shown in Table 3, and all DEGs in this study are shown in [23]. The regulatory network of E. sinensis thoracic ganglia is shown in Figure 3.

Validation of Transcriptome Data by qPCR
Primers for the 10 detected DEGs are shown in [23]. As shown in Figure 4, relative expression levels of the detected DEGs as measured by qPCR were nearly consistent with those determined by high-throughput sequencing. These results indicated that the transcriptome data in this study are reliable.

Validation of Transcriptome Data by qPCR
Primers for the 10 detected DEGs are shown in [23]. As shown in Figure 4, relative expression levels of the detected DEGs as measured by qPCR were nearly consistent with those determined by high-throughput sequencing. These results indicated that the transcriptome data in this study are reliable.

Validation of Transcriptome Data by qPCR
Primers for the 10 detected DEGs are shown in [23]. As shown in Figure 4, relative expression levels of the detected DEGs as measured by qPCR were nearly consistent with those determined by high-throughput sequencing. These results indicated that the transcriptome data in this study are reliable.

Discussion
As shown in Table 3, with the comprehensive analysis of the top 30 GO and top 30 KEGG, differentially expressed KEGG and DEGs can be divided into four categories: carapace development and tissue regeneration, neuronal signal transduction, energy metabolism and homeostasis regulation and immune and anti-stress response.

Carapace Development and Tissue Regeneration at Post-Molt Stage
In this study, some genes relevant to carapace development (bursicon-alpha subunit, DDC, CHS, CS, BMP and BMPR2), skeletal muscle development (ACTA1 and ACTR2) and neuronal system development (RERE, EPHA2, ISWI and YAP1) were significantly differentially expressed. The newly formed soft carapace was gradually hardened under the regulation of various neuropeptides and factors after molting. In this study, some DEGs related to carapace formation were significantly upregulated compared with the pre-molt stage, such as bursicon-alpha subunit, DDC and CHS. Bursicon is a neuropeptide that has been shown to play a core regulatory role in carapace sclerotization during the post-molt period in the blue crab (Callinectes sapidus) and shore crab (Carcinus maenas) [24,25]. DDC catalyzes the conversion of dopa into dopamine and then initiates the subsequent process of exoskeleton sclerotization and mineralization [26]. CS catalyzes chitin biosynthesis and plays a regulatory role in the development of the cuticular layer [27]. In this study, CS was significantly upregulated to promote the formation of the new carapace. In addition, some regulatory genes relevant to bone development were also upregulated, such as BMP and BMPR2 after molting. The research on the regulatory pattern of the Y organs of the blackback land crab (Gecarcinus lateralis) during the molt cycle also indicated that some genes related to bone formation, such as BMP7 and BMP receptor 1B, were upregulated during the post-molt stage [6]. BMPs, comprising a superfamily of growth factors, are closely associated with animal growth and development, and most members of the BMP family have been shown to play an important regulatory role in biocalcification, tissue reconstruction and regeneration [28]. BMP, an important member of the TGF-beta superfamily, can activate BMPR2 and participate in BMP signaling [29].
In addition, some genes relevant to skeletal muscle development were also upregulated at the post-molt stage, such as ACTA1 and ACTR2. Alpha-actin, a major contractile constituent of skeleton muscle, plays a key regulatory role in the mediation of actin networks [30]. Research on the molting regulation of Litopenaeus vannamei indicates that some actin-related regulatory genes are upregulated after molting, such as skeletal muscle actin 6 [31]. The post-molt period is an important stage for muscle regeneration and morphological remodeling.
In this study, some novel genes functioning in the regulation of the neuronal system and connective tissue development were significantly upregulated, such as RERE, EPHA2, ISWI and YAP1. RERE acts as a transcriptional repressor for cell survival and development [32]. Its downregulation as observed in this study is expected to promote post-molt development of E. sinensis. EPHA2 functions in the regulation of brain development and angiogenesis [33]. ISWI plays a pivotal regulatory part in larval blood cell development and metamorphosis [34]. YAP1, a transcriptional regulator, plays a critical role in the regulation of tissue tension and shape [35].

Neuronal Signal Transduction after Molting
In this study, some key regulatory genes relevant to signal transduction were upregulated, such as neuroparsin and tachykinin. Neuroparsin can induce the elevation of trehalose and hemolymph lipids. It was first identified as a regulatory factor functioning in ovary maturation [36,37]. In this study, neuroparsin-2 was significantly elevated after molting. A study on the regulatory function of neuropeptides in green shore crabs during the molt cycle showed that neuroparsin-1 is involved in regulation during the molt cycle [38]. The specific biological function of neuroparsin at the post-molt developmental stage of E. sinensis remains to be elucidated through further study. Tachykinins are excitatory neuropeptides that can initiate the contraction of multiple smooth muscles, and they mediate the downstream neuronal signaling pathway together with tachykinin receptors [39,40]. In this study, tachykinin receptors, such as TKR99D and TKR86C, were upregulated after molting; similarly, tachykinin upregulation was observed in a transcriptomic analysis study of neuropeptidome in lobster Homarus americanus eyestalk ganglia [41]. In addition, other regulatory genes related to neuronal signal transduction were also upregulated, including PPP1R9B, PlEXB and RICH. PPP1R9B plays a pivotal regulatory role in dopaminergic neurotransmission [42]. PlEXB participates in axon guidance of the neuronal system [43]. RICH is an indispensable regulator for the formation of synaptic connections [44].

Regulation of Energy Metabolism and Homeostasis
In this study, regulatory genes relevant to glycan synthesis, such as UXS1, were downregulated after molting. Some genes related to lipid catabolism genes (ENGASE, MTTP and PNLIPRP2) and homeostasis regulation (HNF4A, RGN and TIM) were upregulated. A study on the energy metabolism of E. sinensis hepatopancreas and its association with molting indicated that genes relevant to glycolysis and lipolysis are upregulated during the post-molt stage [45]. UXS1, a catalyst for decarboxylation of UDP-glucuronic acid, is indispensable in tetrasaccharide biosynthesis [46]. In our study, UXS1 was found to be downregulated, which is beneficial for glycolysis and energy supply. ENGASE plays a regulatory role in the release of N-glycans from glycoprotein [47] and was upregulated for energy supply in our study. MTTP can regulate the biosynthesis of cholesteryl ester [48]. PNLIPRP2, as a lipase, participates in the hydrolyzation of triglycerides [49]. HNF4A, as a transcription factor, functions in the maintenance of the circadian rhythm of liver genes [50]. RGN can modulate Ca 2+ -dependent enzyme activities and is essential for the maintenance of calcium homeostasis [51]. TIM can determine the formation of circadian rhythm together with period circadian protein [52].
The post-molt stage is a period for carapace development and tissue regeneration, during which a sufficient energy supply is required, and glycolysis and lipolysis are vigorous. Genes related to homeostasis regulation were upregulated for the maintenance of homeostasis of signal transduction and energy metabolism during the post-molt phase.

Regulation of Immune and Anti-Stress Response
In this study, some immune-relevant regulatory genes involved in immune cell response and antigen recognition were upregulated after molting, such as Frizzle1/7, SM-PDL3B, FCN1 and ITGA4. Furthermore, some modulatory genes related to anti-stress response were also significantly expressed, including GCLC and KEAP1. The Wnt signaling pathway plays a critical regulatory role in intercellular development and differentiation of macrophages, T cells, B cells, etc. Frizzled is the receptor of Wnt protein, and it plays a positive regulatory role in immune response [53]. In this study, Frizzled 1 and Frizzled 7 were upregulated. Similarly, the transcriptome analysis of the Y organs of blackback land crabs showed that Frizzled plays an important role in the immune response during the molting cycle [54]. SMPDL3B, which is located on the surface of macrophages and dendritic cells, plays a vital role in the lipid composition of macrophages and a negative role in immunity [55]. Its downregulation, as observed in this study, is beneficial to immunity enhancement. FCN1 plays a pivotal regulatory role in the activation of the lectin pathway of the complement system [56]. ITGA4 can initiate leukocyte aggregation for an immune response [57]. Glutathione, as an important intracellular regulator, plays an important role in the maintenance of the immune system and functions in antioxidant and detoxification processes [58]. GCLC, as a key component of glutamate-cysteine ligase, participates in the rate-limiting step of glutathione biosynthesis [59]. KEAP1, as a pivotal transcription factor, plays an important regulatory role in antioxidation responses [60].
Many genes relevant to the regulation of the immune response were found to be upregulated. The reason for this is that E. sinensis is vulnerable to the invasion of pathogens during the post-molt period. At this time, the hard exoskeleton is not well-formed, and thus, immunity is enhanced to protect E. sinensis from harm. Discerning the details of the specific regulatory mechanisms awaits further study.

Application of This Study
In this study, four types of key functional genes were differentially expressed during the E. sinensis post-molt stage: These genes can be screened for breeding improved varieties of E. sinensis. In the near future, transgenic technology can be applied in the E. sinensis industry, and the characteristic improvement of E. sinensis can be performed from four aspects, namely carapace development, tissue regeneration, energy homeostasis maintenance and immune response. The obtained improved E. sinensis with a faster growth rate and stronger resistance can provide high-quality juvenile crabs for the breeding industry of E. sinensis and the proliferation and release of wild E. sinensis, thus promoting the recovery of wild E. sinensis resources.

Conclusions
In this study, a comparative transcriptome study of E. sinensis thoracic ganglia at the post-molt and inter-molt stages was carried out for the first time. The results indicate that pathways and regulatory genes related to carapace development, tissue regeneration, glycolysis and lipolysis and immune and anti-stress responses were significantly differentially expressed. At present, research on the regulatory mechanisms of the nervous system in E. sinensis during the molting cycle period is scarce. The results of this study lay a theoretical foundation for research on the regulatory networks operating in the nervous system of E. sinensis during the post-molt developmental period. Detailed knowledge of the regulatory network involved in E. sinensis molting can be used for breeding improved E. sinensis species, the recovery of the wild E. sinensis population and the lucrative development of the E. sinensis artificial breeding industry.