Integrated analysis of mRNAs and lncRNAs reveals candidate marker genes and potential hub lncRNAs associated with growth regulation of the Pacific Oyster, Crassostrea gigas

Background The Pacific oyster, Crassostrea gigas, is an economically important shellfish around the world. Great efforts have been made to improve its growth rate through genetic breeding. However, the candidate marker genes, pathways, and potential lncRNAs involved in oyster growth regulation remain largely unknown. To identify genes, lncRNAs, and pathways involved in growth regulation, C. gigas spat was cultured at a low temperature (15 ℃) to yield a growth-inhibited model, which was used to conduct comparative transcriptome analysis with spat cultured at normal temperature (25 ℃). Results In total, 8627 differentially expressed genes (DEGs) and 1072 differentially expressed lncRNAs (DELs) were identified between the normal-growth oysters (cultured at 25 ℃, hereinafter referred to as NG) and slow-growth oysters (cultured at 15 ℃, hereinafter referred to as SG). Functional enrichment analysis showed that these DEGs were mostly enriched in the AMPK signaling pathway, MAPK signaling pathway, insulin signaling pathway, autophagy, apoptosis, calcium signaling pathway, and endocytosis process. LncRNAs analysis identified 265 cis-acting pairs and 618 trans-acting pairs that might participate in oyster growth regulation. The expression levels of LNC_001270, LNC_003322, LNC_011563, LNC_006260, and LNC_012905 were inducible to the culture temperature and food abundance. These lncRNAs were located at the antisense, upstream, or downstream of the SREBP1/p62, CDC42, CaM, FAS, and PIK3CA genes, respectively. Furthermore, the expression of the trans-acting lncRNAs, including XR_9000022.2, LNC_008019, LNC_015817, LNC_000838, LNC_00839, LNC_011859, LNC_007294, LNC_006429, XR_002198885.1, and XR_902224.2 was also significantly associated with the expression of genes enriched in AMPK signaling pathway, insulin signaling pathway, autophagy, apoptosis, calcium signaling pathway, and endocytosis process. Conclusions In this study, we identified the critical growth-related genes and lncRNAs that could be utilized as candidate markers to illustrate the molecular mechanisms underlying the growth regulation of Pacific oysters. Supplementary Information The online version contains supplementary material available at 10.1186/s12864-023-09543-7.


Background
Growth is one of the most important production traits of aquaculture species [1].Growth is a complex biological trait that is affected by many endogenous and exogenous factors [2][3][4].For marine invertebrates such as oysters, growth is dramatically affected by environmental factors.The Pacific oyster (Crassostrea gigas) has been widely cultivated around the world because of its wide thermophilic range from 5 to 35 °C.The growth of oysters was increased rapidly between 20 ~ 25 °C but is suppressed under 15 ℃ [5][6][7].The nutrient accumulation of oysters varies seasonally, with high glycogen storage in early spring or winter and high lipid synthesis in summer [8].As a result, oyster growth displays seasonal variations with the fluctuation of water temperature, food availability, and other abiotic factors [9].However, the internal regulatory elements involved in the growth regulation of the Pacific oysters are largely unknown and warrant further investigation.
For a long time, improving the growth rate of aquaculture species was regarded as the top priority for aquaculture genetic improvement.Since the 1980s, transgenic technology has received increasing attention for the genetic improvement of farmed fish.The human growth hormone (GH) gene has been used to create transgenic fish, including carp, salmon, and tilapia, with rapid growth rates [10].Shellfish genetic breeding efforts have been made to primarily focus on improvement of economic traits including body weight, growth rate, and survival rate through successive artificial selection over 1~5 generations based on family and mass selection [11,12].However, these traditional breeding techniques require a significant amount of labor and material resources, so identifying the potential growth-regulating marker genes will improve the efficiency of breeding efforts.In order to reveal the phenotypic variation of growth traits in the Pacific oyster, molecular genetics approaches have been used to discover numerous microsatellite markers, SNPs, and construct several sets of genetic linkages maps [13][14][15].However, the critical genes and pathways involved in oyster growth regulation remain largely unknown.
Identifying growth-related candidate marker genes is essential for a better understanding of growth regulation mechanism and for enhancing efficiency of selective breeding in oysters.Transcriptomics techniques have been used to identify the candidate growth-related genes, pathways, and lncRNAs in many organisms [16][17][18][19][20]. Previously, we discovered that microtubule movement, as well as nucleotide and protein biosynthesis, were potentially important for oyster growth by comparative transcriptome analysis of the fast-growing "Haida No.1" oyster and wild oysters [21].Recently, lncRNAs have been identified to participate in oyster shell pigmentation [22,23], gametogenesis and reproduction [24], shell biomineralization [25], environmental temperature adaption [26], and glycogen accumulation [27].However, the candidate genes and lncRNAs that regulate the growth of C. gigas have not been investigated.Genes that regulate growth may in turn be controlled by hub lncRNAs, therefore, identifying these candidate lncRNAs may enable the identification of the candidate marker genes that are responsible for oyster growth control.
This study aims to identify genes, lncRNA, and important pathways that are potentially involved in oyster growth regulation through the comparative transcriptome analysis of oysters cultured at different temperatures.On this basis, we hope to reveal the molecular mechanism of these genes and lncRNAs participating in oyster growth regulation by detecting their expression patterns at different cultured temperatures and nutrient levels.Our study provides a set of genes and lncRNAs which can be used as marker genes of growth in selective breeding programs of the Pacific oyster.

Identification and characterization of lncRNAs
Initial analyses identified 18,969 novel lncRNAs through the integrative computational pipeline as shown in Fig. 1A.The information on the novel lncRNAs and annotated lncRNAs were provided in two separate files (Additional file 2 and Additional file 3).The lncRNAs were classified based on their location in the genome, including intronic lncRNAs (class code "i"), long intergenic noncoding RNAs (lincRNAs, class code "u"), and antisense lncRNAs (class code "x").Of all the novel lncRNAs, the intronic lncRNAs, lincRNAs, and anti-sense lncRNAs account for 68.5%, 25.2%, and 6.4%, respectively.Expression analysis indicated that the expression levels of lncRNAs were lower than that of the protein-coding mRNAs (Fig. 1B).The gene length, the predicted ORF length, and the exon number were all significantly different between lncRNAs and protein-coding mRNAs.Overall, the exon number of the lncRNAs is smaller than that of the protein-coding mRNAs (Fig. 1C), the lengths of both transcripts and predicted ORFs are shorter in lncRNAs than that of the proteincoding mRNAs (Fig. 1D and E).The features of the novel lncRNAs are similar to the annotated lncRNAs but different from the protein-coding mRNAs.

Screening potential genes and lncRNAs involved in the oyster growth regulation
The correlation coefficient of gene expression among samples showed that Pearson's R 2 values of the six samples were bwtween 0.87 and 0.931.Apparently, higher correlation coefficient was observed between samples within group than that between the two groups, normalgrowth oysters and slow-growth oysters (Additional file 4: Fig. S1).Differential expression analysis between the normal-and slow-growth oysters allowed the identification of a total of 8627 DEGs and 1072 DELs (Additional file 1: Table S2 and Table S3).Of these, 3026 DEGs and 202 DELs were expressed significantly higher (P-adj < 0.05 and log 2 FC > 1) in normal-growth oysters, whereas 4701 DEGs and 870 DELs were expressed significantly higher (P-adj < 0.05 and log 2 FC <-1) in slowgrowth oysters (Additional file 4: Fig. S2).

GO and KEGG analysis of the DEGs
GO terms such as "nucleoside biosynthetic process", "pyrimidine-containing compound metabolic process", "sodium ion homeostasis", and the "tryptophan metabolic processes" were significantly enriched (P < 0.05) from the DEGs expressed at higher levels in normal-growth oysters (Fig. 2A, Additional file 1: Table S4).In contrast, the "multivesicular body organization", "vesicle targeting, to, from or within Golgi", "negative regulation of TOR signaling", and several cell proliferation-related processes were significantly enriched (P < 0.05) from the DEGs expressed at higher levels in slow-growth oysters (Fig. 2B, Additional file 1: Table S5).KEGG enrichment analysis revealed that the DEGs were significantly enriched (P < 0.05) in the pathways related to cell metabolisms and signal transduction, such as the "GTP-binding proteins", "apoptosis", "autophagy", "endocytosis", "pyrimidine metabolism", "AMPK signaling pathway", "insulin signaling pathway", and "calcium signaling pathways" (Fig. 2C, Additional file 1: Table S6).The lengths of the transcripts and predicted ORFs were shorter than that of mRNAs Fig. 2 Function analysis of the DEGs in the normal-and slow-growth oysters.(A) GO enrichment of the DEGs that were highly expressed in the normalgrowth oysters.(B) GO enrichment of the DEGs that were highly expressed in the slow-growth oysters.(C) KEGG analysis of the DEGs in normal-and slow-growth oysters [75][76][77].The blue bar represented the genes highly expressed in slow-growth oysters, and the red bar represented the genes highly expressed in normal-growth oysters

Validation of the regulatory relationship between lncRNAs and their target genes
A set of DELs and DEGs with the highest log 2 FC, including SREBP1/p62-LNC_001270, FAS-LNC_006260, CaM-LNC_011563, CDC42-LNC_003322, PIK3CA-LNC_012905, and TBUA1A-LNC_015438, were selected for validation by using qRT-PCR (Table 1).As shown in Fig. 5, the expression profiles of the DELs-DEGs pairs were consistent with the RNA-seq results.The target genes were under significantly positive or negative regulation of the lncRNAs which were consistent with the target gene prediction results (Fig. 5A -F).Furthermore, in the fasting and re-feeding treatment of the oysters, expression of LNC_001270 and LNC_006260 Fig. 4 Transcriptional relationship between the DEGs and DELs.(A) Cis-regulatory networks of DEGs-DELs that participated in the growth regulation.The yellow line represented antisense lncRNAs, the green lines represented lncRNAs located downstream of target genes, and the purple lines represented lncRNAs located upstream of target genes.(B) Trans-regulatory networks of DEGs-DELs that participated in the growth regulation.The pink represented genes highly expressed in normal-growth oysters, and the blue represented genes highly expressed in slow-growth oysters was decreased with fasting treatment (fasting for 5 and 14 days, referred to as F5 and F14) while increased with re-feeding treatment (refeeding after 1 and 6 days, referred to as R1 and R6).Their target genes, including SREBP1, p62, and FAS, showed opposite expression patterns (Fig. 5G, H).Expression of LNC_011563 and LNC_003322 was also decreased with fasting treatment and increased after re-feeding, which was consistent with the expression of their target genes including CaM and CDC42 (Fig. 5I, J).The expression of LNC_012905 was increased with fasting and decreased with re-feeding, which showed an opposite expression pattern in comparison with the expression of the PI3KCA gene (Fig. 5K).In contrast, expression of LNC_003736 was also increased with fasting and decreased with re-feeding, but showed a consistent expression pattern with the TUBA1A target gene (Fig. 5L).

Cell metabolism provides basic energy supply for oyster growth
Under optimal growth conditions, growth factors and their receptor-mediated pathways promote cell proliferation and cell metabolism [28].In this study, the insulin signaling pathway, ILPR-mediated PI3K/AKT signaling pathway, and TOR signaling pathway were activated to promote oyster growth by regulating glycogen metabolism and protein synthesis, with high expression levels of ILPR, PI3KCA, S6Kβ1, GSK3β, and RAPTOR being observed in normal-growth (NG) oysters (Fig. 6A).The activity of the insulin receptor depends on several factors.For instance, PTPepsilonM has been shown to play a role in insulin-induced glucose metabolism via direct dephosphorylation and inactivation of IR [29].Since it is highly expressed in normal-growth oysters, the whole insulin signaling pathway is likely activated in the normal-growth oysters.Furthermore, Ca 2+ and calmodulin (CaM) contribute to the phosphorylation of the beta subunit of the insulin receptor and regulation of the binding activity of insulin to its receptor.Fluctuations in cytosolic Ca 2+ concentrations or inhibition of calcium-calmodulin actions are all thought to affect the secretion of peptides and neurotransmitters [30][31][32][33].Highly expressed CaM genes in normal-growth oysters might facilitate the activity of insulin signaling and promote the secretion of insulin-like peptides.Current knowledge suggests that biogenic amines, including octopamine (OA), dopamine (DA), and serotonin (5-HT) are all involved in growth regulation by stimulating the release of hormones, such as insulin-like peptides, and activating the process of energy mobilization [34,35].Tyrosine metabolism and tryptophan metabolic process/serotonin biosynthetic processes were also increased in normal-growth oysters compared to slow-growth oysters in our study.Tyrosine is an amino acid precursor of norepinephrine and dopamine, and the roles of dopamine and serotonin in Fig. 6 Genes and pathways involved in growth regulation of the oyster.Genes were enriched in signal transduction of AMPK and insulin signaling pathways (A), autophagy (B), pyrimidine metabolism (C), and apoptosis (D).The pink represented genes that were highly expressed in normal-growth oysters, the blue represented genes that were highly expressed in slow-growth oysters, the gray represented unchanged genes, the yellow represented the genes that were both highly expressed in the normal-and slow-growth oysters (these genes possessed same gene name but different gene ID), the green represented the key factors in these pathways, the arrows represented a promoting effect and the "-|" represented an inhibiting effect regulating the secretion of insulin or insulin-like peptides have been widely studied [36,37].All these clues suggest that the insulin-like peptides and the associated factors such as neurotransmitter, kinase, and CaM could work together to control oyster growth by regulating the metabolism activity and the neuroendocrine activity.
Under unfavorable growth conditions, such as nutrient deprivation and temperature stress, AMP-activated protein kinase (AMPK) mediates multiple metabolic processes to replenish cellular ATP supplies.This includes positive regulation of the fatty acid oxidation and autophagy process and negative regulation of ATP-consuming biosynthetic processes including glycogen, lipid, and protein synthesis [38].In addition, with deficiency of exogenous carbohydrates, gluconeogenesis is activated to provide glucose through utilization of amino acids, lactate, pyruvate, and glycerol.The gluconeogenic enzyme, PEPCK, allows hepatic parenchymal cells to produce glucose from pyruvate and other precursors derived from the citric acid cycle [39], while PHKG2 functions to mediate the glycogenolysis process through phosphorylating and activating glycogen phosphorylase [40].Both were highly expressed in the slow-growth oysters.Furthermore, the lipid metabolism-related genes such as the HSL, SREBP1, FAS, ACC, SCD, and HMG-CoA genes were all highly expressed in slow-growth oysters, suggesting a need for stimulation of the gluconeogenesis and the fatty acids metabolism activity at low temperature or other growth inhibited condition to support normal survival (Fig. 6A).With high expression levels of the p62, SNAP29, and ATG genes under growth-inhibited conditions (Fig. 6B), the autophagy process was also activated to promote the degradation of macromolecules and organelles in order to maintain the energy balance under the limited nutrient conditions [41].Furthermore, activation of AMPK under environmental stress is always closely related to the inhibition of PI3K, AKT, and ERK, or inhibition at the level of translation elongation and an increase in autophagy markers.Exposure of cells to ambient stress affects the activity of signaling networks previously implicated in metabolic and growth factor signaling [42].Our previous study also found that the growth is negatively affected under the experiment temperature (15 ℃) where the food intake activity and metabolism of nutrients were influenced.The genes that are in response to lower temperatures could also be involved in the growth regulation process [43].Our findings suggest that two important energy-sensing pathways, insulin and AMPK signaling, play critical roles in the modulation of energy supply and consumption to sustain normal oyster growth.

Cell proliferation, apoptosis, and survival processes are related to oyster growth
During adequate nutrition, growth factors use glucose as energy to promote cell proliferation, but in nutrientstress conditions, the lack of glycogen not only affects cell metabolism but also affects cell proliferation [44].Cell division requires adequate amounts of purine and pyrimidine nucleotides for nucleic acid synthesis.Pyrimidine ribonucleotides are essential for DNA and RNA synthesis, cell growth, and proliferation.They provide the necessary building blocks for nucleic acids and precursors for cell membrane synthesis [45].In this study, the pyrimidine metabolism-related genes were highly expressed under normal growth conditions, indicating that cell proliferation was active (Fig. 6C).Once the nutrients were limited, cells reduced metabolism, went into cell cycle arrest, and eventually induced apoptosis.Apoptosis is an energy-dependent process that eliminates damaged cells and is always accompanied by cytoskeleton rearrangement [46].The TNF signaling and downstream caspase 3, caspase 7, caspase 8, actin, and tubulin genes were highly expressed in slow-growth oysters, indicating that the apoptosis process disrupted the stability of the cytoskeleton and inhibited the oyster growth (Fig. 6D).The NF-κB signaling pathway is a key mediator that regulates cell fate in response to various environmental stimuli or growth factors deprivation by driving the upregulation of proliferative and anti-apoptotic transcripts [47].In this study, several anti-apoptotic genes, such as BIRC, IAPs, XIAP, and Bcl2, were highly expressed in slow-growth oysters and functioned to resist the apoptosis signals induced by adverse environmental conditions (Fig. 6D).Although TNF is a classical pro-inflammatory cytokine, it is also found to be produced from adipose tissue, and associated to obesity-associated metabolic disease [48].In addition, the crosstalk between TNF and insulin signaling through the transcription factor GATA6 was revealed in a previous study [49].The expression of the apoptosis-related protein, ABC transport, GSK-3β, NF-κB, and TOR, are together controlled by the PI3K-AKT signaling and play important roles in the balance of cell metabolism and proliferation [50].We speculated that the TNF and NF-κB signaling pathways induced cell apoptosis and were activated to balance the cell metabolism and cell proliferation of oysters under growth-inhibited conditions.

Endocytosis-associated signaling transduction plays an indispensable role in oyster growth
Endocytosis plays an essential role in controlling the activity and quantity of the membrane receptors, including the G protein-coupled receptors (GPCRs) and the receptor tyrosine kinases (RTKs), thereby increasing or decreasing receptor-mediated signaling transduction [51].Receptors can be recycled to the plasma membrane for signal transduction or be retained in multivesicular bodies (MVBs) and translocate to the lysosomes for degradation.In the EGFR-mediated pathways, ligand binding induces endocytosis of EGFR into endosomes and eventually into the lysosomes for degradation, which reduces the number of active EGFR molecules, attenuates the signaling, and finally influences cell proliferation and individual growth [52].In this study, the MVB genes, VPS genes, CHMP genes, and the EGFR gene were all highly expressed in slow-growth oysters, which suggested that the endocytosis-mediated lysosome degradation might inhibit the oyster growth through attenuating the growth-related signaling transduction.
Ubiquitination of downstream signaling molecules on the plasma membrane, such as the Ras GTPases, also leads to endocytosis and signal downregulation, which impacts organ development [53,54].We found that several E3 ubiquitin ligase genes and the genes of GTPbinding protein were significantly highly expressed under growth-inhibited conditions.Furthermore, the endocytic pathway was reported to be sensitive to nutrient availability.Endocytosis of non-essential proteins indeed may assist survival under conditions of limited energy supply.In yeast, low glucose levels stimulate the endocytosis of plasma membrane proteins to the lysosome and block their recycling to the cell surface, thereby promoting endocytic flux to the vacuole and providing energy for cell metabolism [55].Therefore, we propose that endocytosis-induced signaling transduction, ubiquitination, and plasma membrane protein degradation processes might govern the growth of oysters by balancing the environmental input with endogenous signaling pathways.

LncRNAs participate in transcriptional regulation of the growth-related genes
The lncRNAs are critical regulators of gene expression in growth, development, and environmental stress [56][57][58][59].In this study, we presented a catalog of lncRNAs expressed in C. gigas, including 18,969 novel lncRNAs and 2396 known lncRNAs.Analysis of their target genes allowed identification of 265 DEGs-DELs cis-regulatory pairs and 618 DEGs-DELs trans-regulatory pairs that play indispensable roles in cell metabolism and proliferation.We found that the expressions of SREBP1/p62-LNC_001270, CDC42-LNC_003322, CaM-LNC_011563, FAS-LNC_006260, PIK3CA-LNC_012905, and TUBA1A-LNC_003736 were all influenced by environmental temperature and nutrient level.Once the expression of LNC_001270 decreased under conditions of low temperature and fasting treatment, its target gene p62 was upregulated.As the LNC_001270 is located at the antisense strand of its target gene, LNC_001270 may hybridize with sense RNA as RNA duplexes or establish complex configurations as RNA-DNA duplexes and triplexes [60] to inhibit the expression of p62, and eventually influence the autophagy process.As a master regulator of lipid homeostasis [61], the transcription factor SREBP1 is located downstream of the LNC_001270, thus we speculate that the LNC_001270 may also participate in lipid metabolism.The CDC42 gene, which belongs to the Rho GTPases family, had been reported to be involved in the regulation of the cell cycle, cell survival, actin cytoskeleton organization, and even membrane trafficking [56].Recent studies found that the CDC42 also participated in insulin synthesis, insulin granule mobilization, exocytosis-mediated insulin secretion, and the β cell proliferation [62,63].Regulation of CDC42 is associated with miR-NAs, lncRNAs, and even post-translational modifications [64,65].In this study, we identified LNC_003322 from the upstream of CDC42 whose expression was under the positive regulation of LNC_003322.Furthermore, LNC_011563 is located downstream of its target gene CaM and positively regulates the expression of CaM with the fluctuation of culture temperature and nutrient levels.The insulin signaling downstream key kinase PI3KCA, fatty acid, and cholesterol metabolism-related genes SREBP1 and FAS were all altered by food abundance under negative regulation of LNC_012905, LNC_001270, and LNC_006260.Of all the trans-acting lncRNAs, the expression of XR_9000022.2, LNC_008019, LNC_015817, LNC_000838, LNC_00839, LNC_011859, LNC_007294, LNC_006429, XR_002198885.1,and XR_902224.2was closely related to the expression of genes enriched in AMPK signaling, insulin signaling, pyrimidine metabolism, calcium signaling pathway, endocytosis, apoptosis, and tyrosine metabolism pathways.The lncRNAs may serve as functional hub to play a muti-function in regulating growth-related genes in oysters.

Conclusions
In conclusion, cell metabolism, cell proliferation, and the signaling transduction processes are all important for oyster growth and are easily affected by environmental factors (Fig. 7).Under normal growth conditions, the insulin-like peptides and the ILPR-mediated pathways are active to regulate the processes related to cell metabolism and cell proliferation.In this process, several factors, including the kinase, the CaM, and the neuropeptides, all affect the activity of the insulin signaling pathway.While under growth-inhibited conditions, AMPK signaling is activated to block energy consumption, and autophagy is also activated to maintain the balance of energy metabolism.These processes are always energy-consuming and hindered the growth of oysters.In addition, the internalization of the membrane receptors such as the RTKs and GPCRs is crucial for growth-related signaling transduction.The lncRNAs could serve as functional hub to play crucial roles in the transcription regulation of growthrelated genes.

Growth-inhibited oyster model
The C. gigas spat with similar size (shell height around 2 mm) were initially reared at 25 °C with a 12 h: 12 h light: dark photoperiod.To build the growth-inhibited model, we randomly selected C. gigas spat for culture at a lower temperature (15 ℃).This slow-growth spat (SG) was used for comparative transcriptome analysis with spat cultured at a normal temperature (25 ℃, normal growth, NG).For each cultured temperature, three random replicates were allocated, and the survival, feeding situation, and shell height of the oyster spat were monitored daily.After 10 days, the shell height of the spat cultured at 25 ℃ was significantly higher (4 mm ± 0.3 mm) than that of the spat cultured at 15 ℃ (2 mm ± 0.3 mm) [43].

RNA extraction, library construction, and sequencing
Total RNA was extracted from the soft body of normal-growth oysters and slow-growth oysters by using the Trizol reagent (Invitrogen, USA) according to the manufacturer's instructions.The RNA purity and concentration were evaluated with the Nano Photometer® spectrophotometer (IMPLEN, CA, USA) and Qubit® RNA Assay Kit in Qubit® 2.0 Fluorometer (Life Technologies, CA, USA).The RNA integrity was assessed with the RNA Nano 6000 Assay Kit of the Bioanalyzer 2100 system (Agilent Technologies, CA, USA).For library construction, a total amount of 20 ng RNA per sample was used as input.Firstly, ribosomal RNA was removed with the Epicentre Ribo-zero™ rRNA Removal Kit (Epicentre, USA), and the rRNA-free residue was cleaned up by ethanol precipitation.Subsequently, sequencing libraries were generated using the rRNA-depleted RNA by using NEBNext® Ultra™ Directional RNA Library Prep Kit for Illumina® (NEB, USA) following the manufacturer's recommendations.Briefly, fragmentation was carried out using divalent cations under elevated temperature in Fig. 7 The proposed regulatory network involved in growth regulation of the oyster with the influence of environmental factors.① In slow-growth oysters, internalization and degradation of the membrane receptor (RTKs and GPCRs) through the MVB and lysosome cause signaling termination.In normal-growth oysters, some of the GPCRs recycle to the membrane and activate the PKC and PLC, ultimately increasing calcium metabolism and insulin-like peptides (ILPs) release.Meanwhile, the secretion of ILPs is highly dependent on glucose levels, neurotransmitters, and neuropeptide activity.② Under normal growth conditions, the ILPs and the ILPR-mediated PI3K/AKT, MPAK/ERK, and TOR signaling pathways play indispensable roles in cell and organism growth, cell cycle regulation, and cell metabolism.Under growth-inhibited conditions, activation of AMPK signaling influences fatty acid metabolism, gluconeogenesis, autophagy, and cell metabolism processes.③ Activation of the ILPR-mediated PI3K/AKT signaling pathway also regulates the expression of CASP and the activity of NF-κB signaling, ultimately causing cell apoptosis or survival NEBNext First Strand Synthesis Reaction Buffer (5X).First-strand cDNA was synthesized using a random hexamer primer and M-MuLV Reverse Transcriptase (RNase H-).Second-strand cDNA synthesis was performed using DNA polymerase I and RNase H.In the reaction buffer, dTTPs were replaced by dUTP.The remaining overhangs were converted into blunt ends via exonuclease/ polymerase activities.After adenylation of the 3' ends of the DNA fragments, a NEBNext adaptor with a hairpin loop structure was ligated to prepare for hybridization.To select cDNA fragments of 150 ~ 200 bp in length, the library fragments were purified with the AMPure XP system (Beckman Coulter, Beverly, USA).Thereafter, 3 µl of USER Enzyme (NEB, USA) was incubated at 37° C for 15 min, followed by 5 min at 95 °C before proceeding to PCR.The PCR reaction was performed with Phusion High-Fidelity DNA polymerase over 15 cycles.The library quality was assessed on the Agilent Bioanalyzer 2100 system.Finally, the product was performed on a cBot Cluster Generation System using the TruSeq PE Cluster Kit v3-cBot-HS (Illumina) according to the manufacturer's instructions.After cluster generation, the libraries were sequenced on an Illumina Hiseq 2500 platform for 150 bp paired-end reads.

Quality control and transcriptome assembly
Clean reads were obtained after trimming the reads with adapters, low quality, and uncertain 'N' with the ratio of 'N' > 10% using fastp [66].Reference genome and gene model annotation files were downloaded from the genome website directly (RefSeq: GCF_902806645.1, https://ftp.ncbi.nlm.nih.gov/genomes/refseq/invertebrate/Crassostrea_gigas/latest_assembly_versions/GCF_902806645.1_cgigas_uk_roslin_v1/).An index of the reference genome was built, and paired-end clean reads were aligned to the reference genome using HISAT2 (v2.0.4).The HISAT2 was run with "--rnastrandedness RF", and other parameters were set as default [67].The mapped reads of each sample were assembled with StringTie (v2.1.1)[68] in a referencebased approach.Transcripts from all samples were then merged together with StringTie merge mode to build a consensus set of transcripts across samples.Transcript abundances were estimated and read coverages were generated using StringTie.Then, the assembled transcripts were compared to the known genes recorded in the database using Cufflinks compare [69].The unknown transcripts were used for further lncRNA analysis.

LncRNAs identification
The pipeline of novel lncRNAs prediction was as follows: firstly, the unknown transcripts were selected for identification of lncRNAs, including those exonic overlap on the opposite strand ("x"), those fully contained within a reference intron ("i"), and those located in intergenic regions ("u").Then, transcripts with the exon number ≥ 2, length > 200 bp, and FPKM ≥ 0.5 were chosen, and the transcripts that overlapped with known protein-coding genes on the same strand were discarded.Finally, the CNCI (Coding-Non-Coding-Index) (v2), CPC (v0.1) (--e value le-10), and PFAM (Pfam-scan) (v1.3) (--E 0.001; --domE 0.001; --pfamB) [70][71][72] were used to filter transcripts with coding potential.The remaining transcripts were considered reliable lncRNAs.Furthermore, lncRNAs and mRNAs were compared and analyzed in terms of structure and sequence characteristics, including the transcript length, exon number, and predicted ORF length.The ORF sequence of mRNA was extracted by the annotation of known gene structure, and the ORF sequence of lncRNA was predicted by EMBOSS: getorf.The ORF sequences were converted into protein sequences, and then the length distribution was obtained.

Identification and functional annotation of differentially expressed genes
The differential expression analysis was carried out using the R package DESeq (1.18.0), and the transcripts with P-adjust < 0.05 and | log 2 (fold change) | >1 were assigned as differentially expressed genes [73].Furthermore, GO and KEGG enrichment analysis of DEGs were performed with ClusterProfiler (v4.1.4)[74][75][76][77].The GO terms and KEGG pathways with a corrected P-value < 0.05 were considered significantly enriched.After enrichment, all the significantly enriched GO terms were clustered based on semantic similarity by using the R package rrvgo (threshold = 0.7) [78].

Prediction of target genes
To identify cis-acting lncRNAs that can regulate neighboring coding genes, coding genes within the regions 10 kb upstream or downstream of the DELs were selected.The prediction of trans-target genes was based on the co-expression relationship between DEGs and DELs.We calculated the Pearson correlation coefficients (r) using the R function "cor.test" and selected the DEGs-DELs pairs with | r |>0.99 for further analysis.Finally, the DEGs-DELs interaction networks were constructed and visualized with Cytoscape software [79].

Fasting and re-feeding experiment
For the fasting and re-feeding experiment, ninety 8-month-old C. gigas were randomly divided into three groups and starved for 14 days, then refed with frozen Chlorella ad libitum.Samples were collected before fasting, on day 5 and day 14 during fasting, and 1 h, and 6 h after re-feeding, respectively.At each time point, eight tissues including the labial palp, gill, mantle, digestive gland, hematocyte, heart, visceral ganglia, and adductor muscle from six oysters were rapidly excised and frozen in liquid nitrogen for further RNA extraction and cDNA synthesis.Furthermore, the cDNA used for real-time PCR analysis was derived from a mixture of the eight tissues at each sampling point consisting of an equal amount of each tissue.

qRT-PCR analysis
To validate the transcriptome regulatory relationship between the DEGs and DELs, cDNA from the temperature treatment and fasting/re-feeding treatment samples were all used for qRT-PCR.The primer sets were designed using Primer Express software (Applied Biosystems, USA), and the sequences were shown in Additional file 1: Table S10.All real-time PCRs were carried out in a LightCycler 480 real-time PCR machine (Roche, Switzerland) with a mixture of 5 µL 2×SYBR Premix ExTaq (Qiagen, Germany), 1.0 µL of diluted cDNA, 3 µL of PCR-grade water, and 0.5 µL of each 10 µM primer.The PCR was initiated by denaturation at 95 °C for 30 s; followed by 40 amplification cycles at 95 °C for 15 s and 60 °C for 30 s. Dissociation protocols were used to measure the melting curves.The relative expression level was calculated with the 2 −ΔΔCt method [80], and the data were expressed as the mean ± SD.Statistical significance was determined by the one-way ANOVA and Student's t-test for multiple groups and two groups comparison, and P < 0.05 was considered statistically significant.

Fig. 1
Fig. 1 Identification and characterization of lncRNAs.(A) An integrative computational pipeline for the systematic identification of the lncRNAs.(B) The expression levels of lncRNAs were lower than that of the protein-coding mRNAs.(C) The exon number of lncRNAs was smaller than that of mRNAs.(D, E) The lengths of the transcripts and predicted ORFs were shorter than that of mRNAs

Fig. 5
Fig. 5 Validation of the regulatory relationship between DEGs and DELs.(A-F) The expression patterns of the DEGs-DELs pairs in normal-and slowgrowth oysters.Data are presented as means ± SD of the three independent replicates.(G-L) The expression patterns of the DEGs-DELs pairs during the fasting and re-feeding process.C represented the control group, F5 and F14 represented 5 and 14 days after fasting treatment; R1 and R6 represented 1 and 6 h after re-feeding.Data are expressed as the mean ± SD (n = 6)

Table 1
List of DEGs-DELs pairs in normal-and slow-growth oysters