Contents lists available at ScienceDirect

Aquaculture

journal homepage: www.elsevier.com/locate/aquaculture

Transcriptome analysis of differentially expressed genes in the fore- andhind-intestine of ovate pompano Trachinotus ovatus

Ke-Tao Lina,b, Wen-Xiong Wangb,c, Hui-Ting Ruana,b, Jia-Ge Daia,b, Ji-Jia Suna,b, Li Liua,b,⁎,Xian-De Huanga,b,⁎

a College of Marine Sciences, South China Agricultural University, 483 Wushan Road, Guangzhou 510642, Chinab Joint Laboratory of Guangdong Province and Hong Kong Region on Marine Bioresource Conservation and Exploitation, College of Marine Sciences, South ChinaAgricultural University, Guangzhou 510642, Chinac Department of Ocean Science, Hong Kong University of Science and Technology, Clearwater Bay, Kowloon, Hong Kong

A R T I C L E I N F O

Keywords:Ovate pompanoIntestineTranscriptomeDifferentially expressed genes

A B S T R A C T

The fish intestine is a highly complex organ and plays an important role in fish immunity and nutrition. Tocharacterize the molecular functions of the intestine of ovate pompano, we compared the gene expression dif-ferences and explored the molecular functions of the fore- and hind-intestine (FI and HI) using transcriptomicanalysis. Our analysis yielded a total of approximately 21 Gb (gigabytes) of clean bases and 144M of clean reads(mean of 24M for each sample). Most of the 61,572 genes in the FI and HI were not differentially expressed, andonly 105 genes were identified as the differentially expressed genes (DEGs). Among the 61,572 genes, the genesthat exhibited the highest expression levels were mostly related to various digestive enzymes. Among the 105DEGs, 86 were annotated as immune- and absorption-related genes in ovate pompano. Detailed analysis of theseDEGs in FI and HI indicated that the FI required more abundant genes to handle the attacks and digestion,whereas the HI recruited more genes to deal with the absorption, suggesting that FI focused mainly on immunityand digestion, and the HI focused on absorption. Our results provided an insight on the molecular mechanisms ofthe function of intestine in ovate pompano.

1. Introduction

The intestine is a multifunctional organ critical for nutrient ab-sorption, pathogen recognition, and intestinal microbiome regulation.Many fish diseases in the farming sector are caused by intestinal da-mage, which leads to serious economic loss (Groschwitz and Hogan,2009). Intestinal health and normal function are thus critical for fishhealth for several reasons. First, animals farmed in aquaculture aresusceptible to intestinal infections due to water environment and highstocking densities common in many farming practices. Second, in-testinal health is usually manageable to some degrees by adding variousadditives and drugs to commercial pellet feeds. Finally, the intestinalmicro-ecosystem allows for microbial colonization by symbionts andresponds to dietary manipulations (Martin et al., 2016).

Recent advances in high-throughput technologies such as RNA se-quencing (RNA-seq) have revolutionized the study of fish intestine atthe level of whole transcriptome, which benefits both model and non-model species of fish at ever-decreasing costs. It also provides a pow-erful tool for exploring the complex molecular mechanisms of intestinal

health (Li et al., 2012; Xia et al., 2013; Qian et al., 2014). To date, thetranscriptomic studies of fish intestine mainly focus on the followingfour areas: 1) intestinal immune function and responses to diseasechallenge, such as parasitic infection, viral response, bacterial response,immunostimulants, and vaccination; 2) effects of alternative plantprotein on fish intestinal health; 3) role of intestine in osmoregulatoryfunction and responses to environmental stress; and 4) complex inter-actions between fish intestine and its microbial inhabitants (Martinet al., 2016). Furthermore, intestinal immunity and nutrition are closelyrelated to feeding habits and intestinal microbiota (Buddington et al.,1997; Wang et al., 2018), and different intestinal segments have dif-ferent functions in fish (Rombout et al., 2011). For example, most of themacromolecules and foreign antigens were absorbed in fish hind in-testine (Lokka et al., 2014; Brugman, 2016), where about 95–99% ofshort-chain fatty acids (SCFA) were rapidly absorbed (Titus andAhearn, 1988). Oral vaccines could be degradated in the fish foregut(Rombout et al., 2011; Lovmo et al., 2017). In European sea bass, dif-ferent intestinal segments were compared using high-quality sequen-cing to generate a precise functional map of the intestine (Calduch-

https://doi.org/10.1016/j.aquaculture.2019.04.078Received 22 December 2018; Received in revised form 26 April 2019; Accepted 27 April 2019

⁎ Corresponding authors at: College of Marine Sciences, South China Agricultural University, 483 Wushan Road, Guangzhou 510642, China.E-mail addresses: liuli@scau.edu.cn (L. Liu), huangxd@scau.edu.cn (X.-D. Huang).

Aquaculture 508 (2019) 76–82

Available online 29 April 20190044-8486/ © 2019 Published by Elsevier B.V.

T

Giner et al., 2016).The ovate pompano, Trachinotus ovatus, a carnivorous fish, belongs

to the genus Trachinotus (Percfonnes: Carangidae) and is one of themost economically important warm-water cultured marine fish in theworld due to its fast growth, good flesh quality and suitability for cageculture. In recent years, ovate pompano farming has been developedrapidly and widely along the southern coast of China, includingGuangdong, Guangxi, Fujian and Hainan Provinces. Previous studies onovate pompano mainly focused on nutrition and disease (Zhang et al.,2014; Yu et al., 2016; Tan et al., 2017). In a recent study, a tran-scriptome analysis was performed on various tissues in T. ovatus, whichprovided a valuable genomic resource for novel gene discovery, geneexpression and regulation studies, as well as the identification of ge-netic markers (Xie et al., 2014). In spite of its economic importance,very few studies examined the intestine of ovate pompano. The eluci-dation of intestinal molecular function such as difference in differentparts of intestine could help to better understand fish intestinal im-munity and nutrition.

In the present study, we therefore compared the gene expressiondifferences and explored the molecular functions in the fore- and hind-intestine (FI and HI) (three replicates each) using transcriptome ana-lysis to characterize the intestinal molecular functions in the ovatepompano, which may provide insight into the intestinal molecularbiology of the ovate pompano and other fish.

2. Materials and methods

2.1. Sample collection

Nine ovate pompano (body weight: 500 ± 50 g) were purchasedfrom the Wushan farmer's market (Guangzhou, Guangdong, China).After fish were randomly divided into three groups and sacrificed, theintestinal contents were scraped off with tweezers and discarded, andthe fore- and hind-intestinal segments were removed by surgical scis-sors and rinsed in sterile 1× phosphate-buffered saline solution. Foreach replicate, the FI segments were collected from three fish as men-tioned above and pooled as one sample, and the HI segments were alsoprepared using similar method. Finally, a total of six samples (three FIsamples and three HI samples) were frozen immediately in liquid ni-trogen and then stored at −80 °C. All procedures about the handling ofexperimental fish met the China Law for Animal Health Protection andInstructions (Ethics approval No. SCXK (YU2005-0001)).

2.2. Library construction and sequencing

Total RNA was extracted with a Trizol RNA extraction kit(Invitrogen, USA) following the manufacturer's instructions, and eu-karyotic mRNA was enriched by Oligo (dT) beads. Then, the enrichedmRNA was fragmented and reverse transcribed into cDNA with randomprimers. After the second-strand cDNA was synthesized by DNA poly-merase I, RNase H, dNTP and buffer, purified with QiaQuick PCR ex-traction kit (Qiagen, GER), and end repaired, then the poly(A) wasadded and ligated to Illumina sequencing adapters. Following size se-lection by agarose gel electrophoresis, the ligation products were am-plified by PCR and sequenced using Illumina HiSeqTM 4000 by GeneDenovo Biotechnology Co. (Guangzhou, China).

2.3. De novo assembly

To acquire high-quality clean reads, reads were further filtered toremove low quality reads with> 50% of low quality (Q-value≤ 10)bases and reads containing adapters or> 10% of unknown nucleotides.Transcriptome de novo assembly was carried out using Trinity, theshort-reads assembling program (Haas et al., 2013) with default para-meters. First, Inchworm assembled reads by a greedy k-mer based ap-proach to collect linear contigs. Next, Chrysalis clusters above related

contigs and builds a de Bruijn graphs. Then, Butterfly analyzes thepaths taken by reads and read pairings based on de Bruijn graph, andoutputs alternatively spliced isoform and transcripts derived fromparalogous genes. Finally, the longest transcripts were combined to-gether and assembled into unigenes. Meanwhile, BUSCO (Bench-marking Universal Single-Copy Orthologs) (version 3.0.1) was used toevaluate the quality of transcriptomic assembly with a single copy of anorthologous gene from vertebrata_odb9.

2.4. Unigene expression analysis and annotation

The unigene expression was calculated and normalized to the readsper kb per million reads (RPKM) method (Mortazavi et al., 2008),which eliminated the bias of different gene lengths and abundance onthe calculation of gene expression. RPKM was calculated as:RPKM= (1,000,000 ∗ C)/(N ∗ L/1000), where C is the number of readsthat are uniquely mapped to one unigene, L is the length (base) of theunigene, and N is the total number of reads that are uniquely mapped toall unigenes.

Unigenes were annotated by the BLASTx program (http://www.ncbi.nlm.nih.gov/BLAST/) with an E-value threshold of 10−5 to theNCBI non-redundant protein (Nr) database (http://www.ncbi.nlm.nih.gov), the Swiss-Prot protein database (http://www.expasy.ch/sprot),Gene Ontology (GO), and the Kyoto Encyclopaedia of Genes andGenomes (KEGG) database (http://www.genome.jp/kegg). Proteinfunctional annotations were based on the best alignment results (lowestE-value), which were chosen to determine the sequence direction ofgenes. ESTScan program (Iseli et al., 1999) was used to determine thesequence direction and predict the coding regions. Gene Ontology (GO)annotation of unigenes was analyzed by Blast2GO software (Conesaet al., 2005), then functional classification of unigenes was performedusing WEGO software (Ye et al., 2006; Ye et al., 2018). The KyotoEncyclopaedia of Genes and Genomes (KEGG) annotation of unigeneswas used by KEGG Automatic Annotation Server (http://www.genome.jp/tools/kaas/).

2.5. Analysis of differentially expressed genes (DEGs)

To remove the rRNA mapped reads, short reads alignment toolBowtie2 (2.2.8) (Langmead and Salzberg, 2012) was used for mappingreads to ribosome RNA (rRNA) database, and the remaining reads weremapped to the reference transcriptome by default parameters. Then themapping ratio (Unique Mapped Reads number+Multiple MappedReads number)/All Reads number) and the number of expressive genewere calculated.

After the gene expression levels of the three replicates were nor-malized and equalized by RPKM, analysis of differentially expressedgenes (DEGs) across the FI and HI was performed using the DEGseq Rpackage (Wang et al., 2010a) and adjusted using q-value (< 0.01). Toidentify DEGs across samples or groups, the edgeR package (http://www.r-project.org/) was used.

Genes were regarded as significantly differentially expressed when afold change>2 (|log2FC| > 1, namely |log2(HI-gene_RPKM/FI-gene_RPKM)| > 1) and a false discovery rate (FDR)< 0.05. To de-termine the classifications of DEGs, enrichment analysis of GO func-tions and KEGG pathways were performed. After mapping the DEGs toGO terms or pathways, gene numbers for every term or pathway werecalculated.

3. Results

3.1. De novo transcriptome assembly and quality evaluation

Six samples (three fore-intestine and three for hind-intestine sam-ples) were sequenced using the Illumina HiSeq 4000 platform, whichyielded a total of approximately 21 Gb (gigabytes) of clean bases and

K.-T. Lin, et al. Aquaculture 508 (2019) 76–82

77

approximately 144M clean reads (mean of 24M for each sample). Thedetailed statistics of each sample are listed in Table S1, which includesthe total bases, Q20, Q30, read numbers, GC percentage, adapternumber, and low-quality number before and after filtration. The uniquemapped ratio (unique mapped reads/total reads) was approximately87.7%. Raw read data were uploaded to the NCBI Sequence ReadArchive (SRA) under the accession number SRP150471.

After using the Trinity software, de novo sequence assembly re-sulted in 61,575 unigenes (a total of 72,577,876 assembled bases) with2491 bp N50 and an average length of 1178 ranging from 201 bps to26,003 bps (Table 1). The unigene number and their length distributionindicated that 20,388 unigenes were>1000 bp in length (Fig. 1).These results showed good quality of the assembled unigenes.

To further evaluate completeness of the unigenes, comparison ofT. ovatus unigenes to the orthologs of closely related species(Larimichthys_crocea, Oreochromis_niloticus and Stegastes_partitus) wasperformed. As shown in Table S2 and Fig. S1, there were> 10,000unigenes with the ratio ≥0.9 and coverage> 80%, and most ofT. ovatus unigenes could effectively cover the orthologs of closely re-lated species. The ratio close to 1 or the percentage close to 100 in-dicated the good completeness of the transcriptome assembly.Meanwhile, to check the possible contamination, the list of speciesfound in the NR blast results was analyzed and counted as in Table S3.Furthermore, BUSCO was performed and the result as in Fig. S2 showed

that among total 2586 BUSCOs, complete BUSCOs was 2237 (86.5%)(complete and single-copy BUSCOs: 2152 (83.2%), complete and du-plicated BUSCOs: 85 (3.3%)), fragmented BUSCOs was 205 (7.9%) andmissing BUSCOs was 144 (5.6%).

3.2. Statistics of expressed genes

A total of 61,575 unigenes were obtained in the intestine. Amongthe 61,572 genes, 49,855, 45,912 and 54,971 genes were obtained fromthe three FI samples and represented 59,837 genes, and 53,203, 52,714and 54,773 genes were obtained from the three HI samples and re-presented 60,124 genes. Of the 61,572 genes, 31,493 genes were notannotated, and 30,079 genes were annotated using the Nr databases.The detailed statistics are listed in Tables 1 and 2.

Further, no difference in expression was observed for most of the61,572 genes in the FI and HI. However, a significant difference wasobserved for 105 genes as indicated by the RPKM value (FI_rpkm andHI_rpkm values) (Table S4 and S5). Among the 30,079 sample genesannotated by the Nr Databases, 12,961 FI and 12,347 HI genes hadRPKM values that were< 1, and 746 FI and 767 HI had RPKM valuesthat were 0, which were represented by 0.001 to calculate the geneexpression. The median RPKM values were 1.58 and 1.81 for the FI andHI, respectively, and the mean RPKM values were 18.4 and 16.96 forthe FI and HI, respectively. Among the 31,493 genes not annotated bythe Nr Databases, RPKM values were< 1 for 25,610 FI and 23,928 HIgenes and 0 (0.001) for 989 FI and 681 HI genes. The median valueswere 0.43 and 0.56, and the mean values were 1.51 and 1.84 for FI andHI genes, respectively. The median and mean values for the annotatedgenes were about 4 times and 10 times higher than the non-annotatedgenes, respectively.

Furthermore, for 105 genes that were differentially expressed in theFI and HI, 86 genes were annotated, and 19 genes were not annotatedby the Nr Databases (Fig. 2). In addition, 58 genes were significantlyup-regulated, and 47 genes were down-regulated. For the annotated 86genes, 46 were significantly up-regulated, and 40 were down-regulated.For the non-annotated 19 genes, 12 were significantly up-regulated,

Table 1Statistics for the unigene assembly.

Terms Numerical values

Total assembled bases (bp) 72,577,876GC percentage 46.73N50 2491Max length (bp) 26,003Min length (bp) 201Average length (bp) 1178Unigenes number 61,575

Fig. 1. Length distribution of all unigenes of T. ovatus.

K.-T. Lin, et al. Aquaculture 508 (2019) 76–82

78

and 7 were down-regulated (Fig. 2). These results indicate that therewas little significant differential expression between the FI and HIsamples (105/61572 genes).

3.3. GO functions and KEGG pathways of DEGs

The possible DEG functions between the FI and HI were furtherdetermined using the GO functions and KEGG pathways. The GO clas-sification of DEGs in the FI and HI was determined and shown in Fig. 3.107 and 68 DEGs were grouped into biological process category, 80 and29 DEGs into the cellular component category, and the 40 and 36 intomolecular function category. Most of the up-regulated GO-terms weregreater than the down-regulated GO-terms. In sequential order, the topfive up-regulated GO-terms were ‘binding’, ‘cell part’, ‘cell’, ‘biologicalregulation’, and ‘single-organism process’. To determine the most af-fected pathways in the FI compared to the HI, 21 DEGs were mapped to24 KEGG pathways and listed in Table S6. Then, KEGG enrichmentanalysis (corrected Rich Factor > 0.05) was performed, and ‘neu-roactive ligand-receptor interaction’ (Rich Factor: 5/395) and ‘PPARsignaling pathway’ (Rich Factor: 2/99) were identified as the top twopathway enrichments. Other pathway terms were only one as the nu-merator and some of the DEGs.

3.4. Detailed DEGs in the fore- and hind-intestine

Despite the significant difference for some DEGs especially based onthe formula of |log2(HI-gene_RPKM/FI-gene_RPKM)| > 1, such differ-ence may be caused by error due to the lower original FI and HI RPKMvalues. To make the results more reliable, these DEGs with the lowlyoriginal FI and HI RPKM values were further removed using the sum(3.39) of the FI and HI median RPKM values (1.58 and 1.81) as athreshold. To gain a global view of the DEGs in the FI and HI, 61 of 86annotated DEGs were further clustered and represented as a heatmap(Fig. 4). Considering the lack of genomic and related ncRNA data forthe ovate pompano, the non-annotated 19 genes are listed in Table S5but not shown in the heatmap.

Based on Fig. 4 and Table S5, the down-regulated DEGs were clas-sified into three categories, 1) the immune effecter factors, such asacidic mammalian chitinase (AMCase)-like (RPKM value: 648.32 in the FIVs 0.85 in the HI), natterin-3-like, natterin-3-like isoform X1, Lectin, andeosinophil peroxidase-like, 2) the digestive and absorption related genesin the intestine, such as trypsinogen-like protein 3, alkaline phosphatase-like, meprin A subunit beta-like, peptide YYa, ammonium transporter Rhtype B-like, aquaporin-10-like, carboxypeptidase O-like, and insulin-likepeptide INSL5; and 3) other genes, such as sodium-dependent phosphatetransport protein 2B, and fetuin-B-like. Similarly, the up-regulated DEGsalso fell into three categories: 1) the immune gene collectrin 2) the di-gestive and absorption related genes, including ileal fatty acid-bindingprotein (iFABP) (RPKM value: 6.17 in the FI Vs 4430.41 in the HI), in-osine-uridine preferring nucleoside hydrolase-like, NADPH oxidase orga-nizer 1-like, and cathepsin B; and 3) other genes, such asileal sodium/bileacid cotransporter and telomerase reverse transcriptase. The details arelisted in Table S5. These results indicated that the FI required moreabundant genes to handle the attacks (the immune related effecterfactors) and digestion (the digestive related genes), and the HI recruitedmore genes to deal with the absorption (the absorption related genes),suggesting that FI focused on immunity and digestion, and the HIconcentrated on absorption in the ovate.

In addition, many genes were not differentially expressed at sig-nificant levels between the FI and HI, but they displayed the high ex-pression level (RPKM value) ranging from thousands to tens of thou-sands both in the FI and HI, and mainly annotated as various enzymessuch as trypsin-2 and trypsin-2-like, chymotrypsin A- and B-like, chymo-trypsin-like elastase family member 2A, elastase-1-like, pancreatic elastase,carboxypeptidase B, ubiquitin-like, elongation factor 1 alpha, trypsin-3, 40Sribosomal protein S2, cytochrome c oxidase subunit III (mitochondrion),apolipoprotein A-I, collagen alpha-1(I) chain-like, actin (aortic smoothmuscle), and selenoprotein P isoform X1 (Table S4). To some degrees, thehigh expression levels of these genes (most of which were proteaseenzymes) indicated that they had multiple functions, such as functionsrelated to digestion and absorption in both the FI and HI.

4. Discussion

The intestine is a highly complex organ that plays a key role in fishimmunity and nutrition. In recent years, intestinal microbiome alsoattracted increasingly attention. Exploring the intestinal molecularmechanisms is necessary to better understand the intestine's multiplefunctions, including interaction of the intestine and its microbiome.Several high-throughput profiling (transcriptome) studies have beenperformed to study the intestinal responses to various situations, suchas disease challenge, feeds, environmental stress and developmentalfactors (Martin et al., 2016). In this study, we focused on the intestine ofthe ovate pompano, T. ovatus, and sequenced six transcriptomes of theFI and HI (each tissue was divided into three replicates), which yieldedapproximately 21 Gb of clean bases. Finally, a total number of 61,575unigenes and 61,572 genes were obtained. Fewer than half (30079) ofthe genes were functionally annotated by various databases, such as Nrdatabases, and approximately half of genes (31493) did not completelymatch the known genes, which was reflected in the gene number in the

Table 2Statistics of the annotated gene number.

Sample Genes num (Ratioa) Group Genes num (ratioa) All ref/samples genes num (ratioa) Annotation genes/non-annotation genes

FI1 49,855 (80.97%) FI 59,837 (97.18%) 61,572/61249 (99.48%) 30,079/31493FI2 45,912 (74.57%)FI3 54,971 (89.28%)HI1 53,203 (86.41%) HI 60,124 (97.65%)HI2 52,714 (85.61%)HI3 54,773 (88.96%)

a Ratio: All Samples Genes Num/All Ref Genes Num.

Fig. 2. Histogram of the DEG statistics for the fore- and hind-intestine. The up-regulated and down-regulated genes are shown by the histogram, and the an-notated and non-annotated genes are shown with different shades of gray.

K.-T. Lin, et al. Aquaculture 508 (2019) 76–82

79

intestine of the ovate pompano. Some of the non-annotated genes mayhave been assembly artifacts, unknown genes, or long non-coding RNAs(lncRNAs) (Mattick, 2011) that required for further verification.

Non-annotated genes included unknown genes, error cluster se-quences, and ncRNA such as lncRNA, the expression of which was as-sociated with lower than normal mRNA (Derrien et al., 2012). Fur-thermore, there was almost no significant difference in expressionbetween the FI and HI with only 105 DEGs among the 61,572 genes.The genes that exhibited the highest expression coexisted in the FI andHI with no significant difference and were mainly associated withvarious enzymes such as trypsin-2, trypsin-2-like, chymotrypsin A- andB-like, and chymotrypsin-like elastase family member 2A. The medianand mean RPKM values of the FI and HI samples also confirmed thisresult.

Although intestinal gene expression under different conditions wasreported previously (Martin et al., 2016), there are few DEGs studies onthe different intestinal segments (Wang et al., 2010b; Calduch-Gineret al., 2016). The results of this study suggested that the functions of theDEGs in the FI were primarily related to immunity and digestion, andthe functions of DEGs in the HI were primarily related to absorption inthe ovate pompano. Compared to the HI, 45 annotated genes in the FIsuch as AMCase-like, natterin-3-like and trypsinogen-like protein 3, weredown-regulated. Similar result was observed in the European sea bass

(Dicentrarchus labrax) where an AMCase-like was identified (Calduch-Giner et al., 2016). Meanwhile, AMCase has been implicated in theinduction of an interleukin-13 (IL-13)-mediated pathway via a T helper-2 (Th2)-specific protein and may be an important mediator of IL-13–induced responses. In addition, it may function as a major protease-resistant glycosidase to constitutively degrade the chitin substrates aspart of the host's defense against chitin-containing pathogens in themammalian digestive system (Zhu et al., 2004; Ohno et al., 2016).Some genes were also up-regulated. For example, a sharp increase inthe expression of> 700× was observed for iFABP. iFABP belongs to themembers of FABP family, which binds long-chain fatty acids with highaffinity and function as central regulators of whole-body metaboliccontrol. They are involved in reversibly binding intracellular hydro-phobic ligands and trafficking them throughout cellular compartmentsand specific proteins, and they directly regulate the cognate nucleartranscription factor activity (Storch and Corsico, 2008; Thumser et al.,2014). Additionally, there were also some DEGs, such as Hox and vi-tamin K, whose expressions were not detected in the FI and HI. To somedegrees, these DEGs indicated that the gene levels were different in theFI and HI of the ovate pompano, which sheds light on the differentfunctions and mechanisms of action in the intestine, and contributes toidentify and study the key indicators of intestinal functions.

Furthermore, the intestinal microbiota had been revealed to

Fig. 3. The GO classification of the DEGs in the fore- and hind-intestine.

K.-T. Lin, et al. Aquaculture 508 (2019) 76–82

80

regulate multiple host metabolic pathways, and interact with intestinalimmune responses during health and disease (Round and Mazmanian,2009; Nicholson et al., 2012; Kamada et al., 2013). As the reciprocalmechanisms between the host and intestinal microbiota, studies of theintestinal microbiota would promote to further understand the functionof DEGs in the intestinal different segments. Therefore, in our otherworks, 16S rRNA gene sequencing was also applied to investigate themicrobial composition and distribution in the FI and HI of the ovatepompano (data not shown). Combined with the transcriptional and 16SrRNA sequencing studies, it may be better to identify key indicators ofintestinal gene functions and vital bacterial population of the intestinalmicrobiota.

In conclusion, we obtained the fore- and hind-intestinal (three re-plicates each) transcriptomes of the ovate pompano and determinedboth the similarities and differences in fore-and hind-intestinal geneexpression. Of the 61,572 genes in the FI and HI, no difference in ex-pression was observed with the exception of 105 DEGs. The genes thatdemonstrated the highest expression levels were mostly related tovarious digestive enzymes. Among the 105 DEGs, 86 were annotated as

genes relating to immunity and absorption in the ovate pompano. Theresults of Detailed DEGs indicated that the FI required more abundantgenes to handle the attacks (the immune related effecter factors) anddigestion (the digestive related genes), and the HI recruited more genesto deal with the absorption (the absorption related genes). Our resultsmay provide insight on the intestinal molecular biology of the ovatepompano.

Supplementary data to this article can be found online at https://doi.org/10.1016/j.aquaculture.2019.04.078.

Acknowledgments

We thank the anonymous reviewers for their comments on thiswork. This study was supported by the National Infrastructure ofFishery Germplasm Resources (2018DKA30470).

References

Brugman, S., 2016. The zebrafish as a model to study intestinal inflammation. Dev. Comp.Immunol. 64, 82–92.

Fig. 4. Heatmap of the DEGs in the fore- and hind-intestine. Log2-transformed RPKM value for each gene has been taken the average of three values, and is shownusing a colour scale.

K.-T. Lin, et al. Aquaculture 508 (2019) 76–82

81

Buddington, R.K., Krogdahl, A., Bakke-Mckellep, A.M., 1997. The intestines of carni-vorous fish: structure and functions and the relations with diet. Acta Physiol. Scand.Suppl. 638, 67–80.

Calduch-Giner, J.A., Sitja-Bobadilla, A., Perez-Sanchez, J., 2016. Gene expression pro-filing reveals functional specialization along the intestinal tract of a carnivorousTeleostean fish (Dicentrarchus labrax). Front. Physiol. 7, 359.

Conesa, A., Gotz, S., Garcia-Gomez, J.M., Terol, J., Talon, M., Robles, M., 2005. Blast2GO:a universal tool for annotation, visualization and analysis in functional genomicsresearch. Bioinformatics 21, 3674–3676.

Derrien, T., Johnson, R., Bussotti, G., Tanzer, A., Djebali, S., Tilgner, H., Guernec, G.,Martin, D., Merkel, A., Knowles, D.G., Lagarde, J., Veeravalli, L., Ruan, X., Ruan, Y.,Lassmann, T., Carninci, P., Brown, J.B., Lipovich, L., Gonzalez, J.M., Thomas, M.,Davis, C.A., Shiekhattar, R., Gingeras, T.R., Hubbard, T.J., Notredame, C., Harrow, J.,Guigo, R., 2012. The GENCODE v7 catalog of human long noncoding RNAs: analysisof their gene structure, evolution, and expression. Genome Res. 22, 1775–1789.

Groschwitz, K.R., Hogan, S.P., 2009. Intestinal barrier function: molecular regulation anddisease pathogenesis. J. Allergy Clin. Immunol. 124, 3–20.

Haas, B.J., Papanicolaou, A., Yassour, M., Grabherr, M., Blood, P.D., Bowden, J., Couger,M.B., Eccles, D., Li, B., Lieber, M., MacManes, M.D., Ott, M., Orvis, J., Pochet, N.,Strozzi, F., Weeks, N., Westerman, R., William, T., Dewey, C.N., Henschel, R., LeDuc,R.D., Friedman, N., Regev, A., 2013. De novo transcript sequence reconstruction fromRNA-seq using the Trinity platform for reference generation and analysis. Nat. Protoc.8, 1494–1512.

Iseli, C., Jongeneel, C.V., Bucher, P., 1999. ESTScan: a program for detecting, evaluating,and reconstructing potential coding regions in EST sequences. Proc. Int. Conf. IntellSyst. Mol. Biol. 138–148.

Kamada, N., Seo, S.U., Chen, G.Y., Nunez, G., 2013. Role of the gut microbiota in im-munity and inflammatory disease. Nat. Rev. Immunol. 13, 321–335.

Langmead, B., Salzberg, S.L., 2012. Fast gapped-read alignment with Bowtie 2. Nat.Methods 9, 357–359.

Li, C., Zhang, Y., Wang, R., Lu, J., Nandi, S., Mohanty, S., Terhune, J., Liu, Z., Peatman, E.,2012. RNA-seq analysis of mucosal immune responses reveals signatures of intestinalbarrier disruption and pathogen entry following Edwardsiella ictaluri infection inchannel catfish, Ictalurus punctatus. Fish Shellfish Immunol. 32, 816–827.

Lokka, G., Austbo, L., Falk, K., Bromage, E., Fjelldal, P.G., Hansen, T., Hordvik, I.,Koppang, E.O., 2014. Immune parameters in the intestine of wild and reared un-vaccinated and vaccinated Atlantic salmon (Salmo salar L.). Dev. Comp. Immunol. 47,6–16.

Lovmo, S.D., Speth, M.T., Repnik, U., Koppang, E.O., Griffiths, G.W., Hildahl, J.P., 2017.Translocation of nanoparticles and Mycobacterium marinum across the intestinalepithelium in zebrafish and the role of the mucosal immune system. Dev. Comp.Immunol. 67, 508–518.

Martin, S.A., Dehler, C.E., Krol, E., 2016. Transcriptomic responses in the fish intestine.Dev. Comp. Immunol. 64, 103–117.

Mattick, J.S., 2011. The central role of RNA in human development and cognition. FEBSLett. 585, 1600–1616.

Mortazavi, A., Williams, B.A., McCue, K., Schaeffer, L., Wold, B., 2008. Mapping andquantifying mammalian transcriptomes by RNA-Seq. Nat. Methods 5, 621–628.

Nicholson, J.K., Holmes, E., Kinross, J., Burcelin, R., Gibson, G., Jia, W., Pettersson, S.,2012. Host-gut microbiota metabolic interactions. Science 336, 1262–1267.

Ohno, M., Kimura, M., Miyazaki, H., Okawa, K., Onuki, R., Nemoto, C., Tabata, E.,

Wakita, S., Kashimura, A., Sakaguchi, M., Sugahara, Y., Nukina, N., Bauer, P.O.,Oyama, F., 2016. Acidic mammalian chitinase is a proteases-resistant glycosidase inmouse digestive system. Sci. Rep. 6, 37756.

Qian, X., Ba, Y., Zhuang, Q., Zhong, G., 2014. RNA-Seq technology and its application infish transcriptomics. OMICS 18, 98–110.

Rombout, J.H., Abelli, L., Picchietti, S., Scapigliati, G., Kiron, V., 2011. Teleost intestinalimmunology. Fish Shellfish Immunol 31, 616–626.

Round, J.L., Mazmanian, S.K., 2009. The gut microbiota shapes intestinal immune re-sponses during health and disease. Nat. Rev. Immunol. 9, 313–323.

Storch, J., Corsico, B., 2008. The emerging functions and mechanisms of mammalian fattyacid-binding proteins. Annu. Rev. Nutr. 28, 73–95.

Tan, X., Sun, Z., Huang, Z., Zhou, C., Lin, H., Tan, L., Xun, P., Huang, Q., 2017. Effects ofdietary hawthorn extract on growth performance, immune responses, growth- andimmune-related genes expression of juvenile golden pompano (Trachinotus ovatus)and its susceptibility to Vibrio harveyi infection. Fish Shellfish Immunol. 70, 656–664.

Thumser, A.E., Moore, J.B., Plant, N.J., 2014. Fatty acid binding proteins: tissue-specificfunctions in health and disease. Curr. Opin. Clin. Nutr. Metab. Care 17, 124–129.

Titus, E., Ahearn, G.A., 1988. Short-chain fatty acid transport in the intestine of a her-bivorous teleost. J. Exp. Biol. 135, 77–94.

Wang, L., Feng, Z., Wang, X., Wang, X., Zhang, X., 2010a. DEGseq: an R package foridentifying differentially expressed genes from RNA-seq data. Bioinformatics 26,136–138.

Wang, Z., Du, J., Lam, S.H., Mathavan, S., Matsudaira, P., Gong, Z., 2010b. Morphologicaland molecular evidence for functional organization along the rostrocaudal axis of theadult zebrafish intestine. BMC Genomics 11, 392.

Wang, A.R., Ran, C., Ringø, E., Zhou, Z.G., 2018. Progress in fish gastrointestinal mi-crobiota research. Rev. Aquac. 10, 626–640.

Xia, J.H., Liu, P., Liu, F., Lin, G., Sun, F., Tu, R., Yue, G.H., 2013. Analysis of stress-responsive transcriptome in the intestine of Asian seabass (Lates calcarifer) usingRNA-seq. DNA Res. 20, 449–460.

Xie, Z., Xiao, L., Wang, D., Fang, C., Liu, Q., Li, Z., Liu, X., Zhang, Y., Li, S., Lin, H., 2014.Transcriptome analysis of the Trachinotus ovatus: identification of reproduction,growth and immune-related genes and microsatellite markers. PLoS One 9, e109419.

Ye, J., Fang, L., Zheng, H., Zhang, Y., Chen, J., Zhang, Z., Wang, J., Li, S., Li, R., Bolund,L., Wang, J., 2006. WEGO: a web tool for plotting GO annotations. Nucleic Acids Res.34, W293–W297.

Ye, J., Zhang, Y., Cui, H., Liu, J., Wu, Y., Cheng, Y., Xu, H., Huang, X., Li, S., Zhou, A.,Zhang, X., Bolund, L., Chen, Q., Wang, J., Yang, H., Fang, L., Shi, C., 2018. WEGO2.0: a web tool for analyzing and plotting GO annotations, 2018 update. NucleicAcids Res. 46, W71–W75.

Yu, Y., Wei, S., Wang, Z., Huang, X., Huang, Y., Cai, J., Li, C., Qin, Q., 2016.Establishment of a new cell line from the snout tissue of golden pompano Trachinotusovatus, and its application in virus susceptibility. J. Fish Biol. 88, 2251–2262.

Zhang, Q., Yu, H., Tong, T., Tong, W., Dong, L., Xu, M., Wang, Z., 2014. Dietary sup-plementation of Bacillus subtilis and fructooligosaccharide enhance the growth, non-specific immunity of juvenile ovate pompano, Trachinotus ovatus and its disease re-sistance against Vibrio vulnificus. Fish Shellfish Immunol. 38, 7–14.

Zhu, Z., Zheng, T., Homer, R.J., Kim, Y.K., Chen, N.Y., Cohn, L., Hamid, Q., Elias, J.A.,2004. Acidic mammalian chitinase in asthmatic Th2 inflammation and IL-13 pathwayactivation. Science 304, 1678–1682.

K.-T. Lin, et al. Aquaculture 508 (2019) 76–82

82