Immune gene discovery by expressed sequence tag (EST) analysis of hemocytes in the ridgetail white prawn Exopalaemon carinicauda.

The ridgetail white prawn Exopalaemon carinicauda is one of the most important commercial species in eastern China. However, little information of immune genes in E. carinicauda has been reported. To identify distinctive genes associated with immunity, an expressed sequence tag (EST) library was constructed from hemocytes of E. carinicauda. A total of 3411 clones were sequenced, yielding 2853 ESTs and the average sequence length is 436 bp. The cluster and assembly analysis yielded 1053 unique sequences including 329 contigs and 724 singletons. Blast analysis identified 593 (56.3%) of the unique sequences as orthologs of genes from other organisms (E-value < 1e-5). Based on the COG and Gene Ontology (GO), 593 unique sequences were classified. Through comparison with previous studies, 153 genes assembled from 367 ESTs have been identified as possibly involved in defense or immune functions. These genes are categorized into seven categories according to their putative functions in shrimp immune system: antimicrobial peptides, prophenoloxidase activating system, antioxidant defense systems, chaperone proteins, clottable proteins, pattern recognition receptors and other immune-related genes. According to EST abundance, the major immune-related genes were thioredoxin (141, 4.94% of all ESTs) and calmodulin (14, 0.49% of all ESTs). The EST sequences of E. carinicauda hemocytes provide important information of the immune system and lay the groundwork for development of molecular markers related to disease resistance in prawn species. Crown

Introduction

The ridgetail white prawn Exopalaemon carinicauda is one of the major economic shrimp species, which naturally distributes in the coasts of Yellow Sea and Bohai Sea, China. The ridgetail white prawn contributes to one third of the gross outcome of the polyculture ponds in eastern China [1]. Given its high commercial interest, studies focus on the diseases causing large mortality and“milky shrimp” disease, with Hematodinium confirmed as its main causative agents, has received increasing attention [1]. However, there are no effective measures to take for control of this disease. The basic problem is a lack of enough knowledge of the immune system and defense mechanisms of crustacean [2-4]. Therefore, better understanding of immune system and identification of immune-related genes will be important for the health management in crustacean farming.

Expressed sequence tags (ESTs) are partial sequences of cDNAs typically produced by rapid single-pass sequencing of randomly chosen clones from a cDNA library [5]. ESTs analysis is an efficient approach for gene discovery and exploitation for molecular markers [6-11]. In addition, EST analysis could also be used in investigation of gene expression patterns in different tissues [12], and generation of genetic linkage maps and physical maps [13,14]. The research protocol has been successfully used to discover immune related genes in Pacific white shrimp Litopenaeus vannamei and Atlantic white shrimp Litopenaeus setiferus [6], kuruma prawn Marsupenaeus japonicus [15], black tiger shrimp Penaeus monodon [16], Chinese shrimp Fenneropenaeus chinensis [17] and Chinese mitten crab Eriocheir sinensis [9]. In addition, sex-related genes have also been discovered in ovaries of P. monodon by ESTs analysis [18,19]. However, very little is known about immune-related genes in E. carinicauda, except conformation of “milky shrimp” disease caused by Hematodinium and identification of HSP 70, HSP 90 and Ferritin genes in E. carinicauda [1,20-22]. Until now, there were no ESTs of ridgetail white prawn in GenBank. The objective of this study was to discover immune-related genes and understand the immune system of E. carinicauda at the molecular level through constructing a hemocytes cDNA library and ESTs analysis.

Materials and methods

Animal materials

Thirty healthy adult individuals of E. carinicauda, averaging weight 1.18 ± 0.35 g, were collected from a commercial farm in Qingdao, China. They were cultured in filtered aerated seawater (salinity 20‰, pH 8.2) at 18 ± 0.5 °C for 7 days before processing. Hemocytes were collected with syringe which contained an equal volume of anti-coagulant buffer (0.45 M NaCl, 0.1 M glucose, 30 mM sodium citrate, 26 mM citric acid, 10 mM EDTA, pH 4.6) [23], and centrifuged at 800 rpm, 4 °C for 15 min. The hemocytes were preserved in liquid nitrogen immediately, up to RNA extraction.

Total RNA extraction and mRNA purification

The hemocytes were treated with Trizol reagent (Invitrogen, USA), according to the manufacturer's protocol. The RNA samples were determined with 1% agarose gel electrophoresis and spectrophotometer, all OD260/OD280 were between 1.8 and 2.0. The mRNA were isolated and purified with PolyATract mRNA Isolation System (Promega, USA), stored in −70 °C for the next step.

cDNA library construction

The cDNA library was constructed by using SMART cDNA Library Construction Kit (Clontech, USA), according to the protocol of the manufacturer. The first strand cDNA was synthesized using SMART Scribe™ MMLV reverse transcriptase with SMART IV oligonucleotide, CDS III/3′ PCR primer, DTT and dNTPs provided in the kits. The double-stranded cDNA (ds-cDNA) was synthesized by LD-PCR (long distance PCR), using 5′PCR primer, CDS III/3′ PCR primer and Advantage 2 PCR kit (Clontech, USA). The ds-cDNA was digested by proteinase K (20 μg/μL) at 45 °C for 20 min immediately, and then was digested by SfiI enzyme at 50 °C for 2 h. Following, the cDNA size fractionated by CHROMA SPIN-400 (Clontech, USA), and the first three fractions containing cDNA were collected and pooled in a clean tube for packaging. The ligation of cDNA to λTriplEx2 vector had parallel ligation reactions which had different vector-to-cDNA ratios 1:0.5, 1:1 and 1:1.5, ensuring to obtain a high volume library. Each ligation was packaged with the Packagene Lambda DNA Packaging Extracts (Promega, USA). The titre of unamplified library was detected with Escherichia coli XL1-Blue on LB/MgSO4 plates; the percentage of recombinant clones was determined by blue/white screening approach. The unamplified library was amplified and the titre of amplified library was detected as above.

Sequencing

A small quantity of E. carinicauda cDNA library was converted to E. coli BM 25.8 strain. Single colonies were picked randomly; bacterial plasmids were extracted by conventional method and sequenced at the 5′ end of each cDNA using the 5′pTriPLEx2 primer (5′-CTC CGA GAT CTG GAC GAG C-3′) and 3730 XL automatic sequencing machine (ABI).

Data analysis and annotation

Vector sequences and primer sequences were removed by using Cross-match software [24,25]. Short sequences (<100bp in length) or poor quality sequences were eliminated. High quality ESTs were analyzed with Phrap program [26]. They were divided into contiguous sequences (contigs) and singlets. All of the unique sequences were subjected to the non-redundant (nr) protein and nucleotide database with BLASTx and BLASTn program searching for functional annotation [27], and sequence homology was accepted as E-value < 1e-5. Unique sequences were clustered into categories using Cluster of Orthologous Groups of proteins (COG). Gene Ontology (GO) annotions were assigned using program Gopipe [28] according to its uniprot accession number [29]. In addition, the unique sequences associated with the innate immune response were identified by sequence alignment according to the previously public report.

Results

mRNA and cDNA library quality

The A260/A280 ratio of the RNA was 1.91. The integrity of the total RNA was verified by agarose gel electrophoresis (Fig. 1). The total amount of RNA (5 mg) was used for mRNA isolation. Purified mRNA of the hemocytes for cDNA library construction was 0.5 μg. The cDNA synthesis produced cDNA fragments in a range of sizes, but only the fragments from 500 bp to 1000 bp were extracted and used for the cDNA library construction. The cDNA library was of high quality and contained about 2.8 × 106 pfu/mL and the recombinant efficiency of cDNA library was over 99% by blue/white screening.

Fig. 1

Total RNA from the hemocytes of E. carinicauda for the cDNA library construction.

EST sequencing and analyzing

In this study, sequencing of 3411 clones produced 2853 high quality EST sequences with an average read length of 436 bp, and the length distributions were shown in Fig. 2. The sequences of 2853 ESTs have been submitted to GenBank (GenBank accession number: JK995494–JK998346).

Fig. 2

Distribution of the lengths of quality ESTs included in assembly.

The 2853 high quality ESTs were assembled by phrap program to organize the redundant ESTs into overlapping contigs. The assembly analysis of 2853 ESTs resulted in the identification of 724 singletons and the remaining 2129 sequences were assembled into 329 contigs. In total 1053 unique sequences were represented (Table 1). The length of the contigs ranged from 100 to 8722 bp, with an average length of 886 bp (Fig. 3). The length of the singlets ranged from 101 to 1209 bp, with an average length of 581 bp (Fig. 3).

Table 1

Summary of ESTs derived from E. carinicauda.

Items	Number
High quality ESTs	2853
Contigs (number of ESTs included in contigs)	329
Singlets	724
Unique genes	1053
Known genes	593
Unknown genes (hypothetical proteins)	22
No homology genes	438
Average EST length (bp)	436 bp

Fig. 3

Distribution of the length of the contigs and singlets.

Functional annotation and classification of ESTs

All of the 1053 unique sequences were compared to the non-redundant (nr) protein and nucleotide (nt) database using BLASTx and BLASTn program, respectively. Of the 1053 unique sequences, 593 (56.3%) showed high homology with known sequences in public databases, 22 (2.1%) showed homology with unknown genes (hypothetical proteins), and remaining 438 (41.6%) had no significant homology to any sequences, suggesting E. carinicauda possess a large number of unique genes that remain to be clarified [30].

The abundance of ESTs was a rough reflection of the mRNA population [31], which represented the genes that were the most abundant express in the hemocytes of E. carinicauda. Of the 20 most abundant ESTs, 13 contigs were identified to homologous to known proteins, whereas 7 were not similar to any known proteins (Table 2). In the 13 contigs that homologous to known proteins, the first was trp repressor binding protein (8.34%), the second was cytochrome c oxidase subunit I (5.08%), the third was thioredoxin (4.94%). In addition, only 2 contigs were homologous to immune genes, thioredoxin (4.94%) and calmodulin (0.49%).

Table 2

The top 20 most abundant ESTs found in the cDNA library of E. carinicauda.

Name	Annotation	Length (bp)	ESTs number	Percentage of total (%)
Contig 313	Trp repressor binding protein	2109	238	8.34
Contig3 26	Cytochrome c oxidase subunit I	1163	145	5.08
Contig 271	Thioredoxin	1079	141	4.94
Contig 309	Hypothetical protein	1321	115	4.03
Contig 244	No hits	1242	133	4.66
Contig 243	No hits	1868	96	3.36
Contig 140	Hypothetical protein	1449	81	2.84
Contig 139	Nascent polypeptide	2332	67	2.35
Contig 289	Glucose transporter	349	61	2.14
Contig 298	Kruppel-like factor 5	998	57	2.00
Contig 237	No hits	1205	54	1.89
Contig 195	Ring finger protein 39	720	54	1.89
Contig 217	MX1 predicted protein	676	47	1.65
Contig 36	Beta-galactosidase	2064	40	1.40
Contig 197	Beta-actin	1035	39	1.37
Contig 134	No hits	1632	27	0.95
Contig 33	No hits	1481	22	0.77
Contig 288	Cytochrome c oxidase subunit I	586	15	0.53
Contig 133	Cytochrome b	433	15	0.53
Contig 327	Calmodulin	2906	14	0.49

The 593 matched unique sequences were clustered into fourteen categories (Fig. 4), using Cluster of Orthologous Groups of proteins (COG). The proteins in the same COG categories were assumed to be from of a same ancestor protein, or paralogs, or orthologs. The largest categories were translation and ribosomal structure as well as biogenesis, which revealed 41.0%; the second categories were posttranslational modification, protein turnover and chaperones (11.0%); the third categories were inorganic ion transport and metabolism, as well as general function prediction only, they both revealed 9.0%.

Fig. 4

COG molecular functional classification of E. carinicauda hemocytes 593 matched unique sequences.

With the Gene Ontology (GO) classification, the 593 matched unique sequences were classified to three functional categories: molecular function (Fig. 5), biological process (Fig. 6) and cellular component (Fig. 7). In the molecular function, these matched unique sequences were clustered into nineteen classifications. The largest subcategory of the molecular function was the “nucleotide binding” (18.6%), the second was “protein binding” (14.8%). In terms of biological process, these unique sequences were classed into eighteen classifications. The most represented biological processes were “cellular physiological process” (25.0%) and “metabolic process” (15.0%). According to cellular component, these unique sequences were divided into twelve classifications. The top represented cellular components were “cell part” (25.0%) and “cell” (25.0%).

Fig. 5

Gene ontology molecular function classification of annotated known genes from the hemocytes cDNA library of E. carinicauda.

Fig. 6

Gene ontology biological process classification of annotated known genes from the haemocytes cDNA library of E. carinicauda.

Fig. 7

Gene ontology cellular component classification of annotated known genes from the haemocytes cDNA library of E. carinicauda.

Identification and classification of immune-related genes

Of all the 593 matched unique sequences, 153 unique sequences (28 contigs and 125 singlets) were homologous to genes related to immunity in other organisms, and this was the first report of these immune genes in E. carinicauda. These genes were categorized into “antimicrobial peptides” (2 contigs and 9 singlets), “prophenoloxidase activating system” (5 contigs and 18 singlets), “antioxidant defense systems” (4 contigs and 23 singlets), “chaperone proteins” (2 contigs and 13 singlets), “clottable proteins” (1 contigs and 3 singlets), “patten recognition receptors” (12 singlets) and “other immune-related genes” (14 contigs and 47 singlets). Of the 367 immune genes, 39.24% of the ESTs were thioredoxin, 4.63% of these were 14-3-3 protein, 3.81% of these were calmodulin (Table 3).

Table 3

Immune related genes identified in the hemocyte cDNA library of E. carinicauda.

Annotation	Matching organism	E-value	Number of ESTs
Antimicrobial peptides
Anti-lipopolysaccharide factor 2	Macrobrachium rosenbergii	5.00E-37	1
Antimicrobial peptide precursor	Rana pleuraden	2.00E-09	6
Amolopin-p1 antimicrobial peptide precursor	Amolops loloensis	2.00E-10	1
Odorranain-L1 antimicrobial peptide precursor	Odorrana grahami	4.00E-09	1
Antimicrobial peptide defensin 3	Anopheles gambiae	1.00E-09	1
Hemocyanin subunit L	M. japonicus	1.00E-156	2
Hemocyanin	F. chinensis	1.00E-140	1
Crustin	Litopenaeus schmitti	3.00E-11	1
NHP2 non-histone chromosome protein 2-like 1	Xenopus laevis	1.00E-06	1
Histone deacetylase 8	Bos taurus	3.00E-09	1
Histone H2A	Strongylocentrotus purpuratus	2.00E-13	1
Prophenoloxidase activating system
Prophenoloxidase	M.rosenbergii	5.00E-102	2
Prophenoloxidase-activating enzyme 1a	P. monodon	1.00E-52	1
Serine proteinase	Tenebrio molitor	2.00E-08	1
Serine protease 34	Mamestra configurata	6.00E-10	1
Clip domain serine proteinase 1	P. trituberculatus	2.00E-08	1
Encystation-mediating serine proteinase	Acanthamoeba healyi	4.00E-08	1
HtrA serine peptidase 3	Mus musculus	5.00E-09	1
Serine protease 2	M. musculus	4.00E-08	1
Serine (or cysteine) peptidase inhibitor, clade B	M. musculus	3.00E-07	2
Serpin peptidase inhibitor, clade E	Danio rerio	1.00E-05	2
Serpin peptidase inhibitor, clade C	Sus scrofa	3.00E-07	1
Serpin peptidase inhibitor, clade A	Xenopus tropicalis	6.00E-09	1
Serpin-2	Spodoptera exigua	6.00E-10	1
Salivary secreted serine protease inhibitor	Anopheles stephensi	7.00E-07	1
Kazal-type proteinase inhibitor protein	Haliotis discus discus	4.00E-31	1
Peptidase inhibitor 16	Homo sapiens	1.00E-08	1
Salivary protease inhibitor	Culicoides nubeculosus	2.00E-10	1
Chymotrypsin BII	Rimicaris exoculata	1.00E-31	1
Chymotrypsin-like proteinase	Heliothis virescens	2.00E-11	1
Chymotrypsinogen B1	D. rerio	9.00E-13	2
Chymotrypsin-like elastase family, member 3B	X. laevis	9.00E-09	1
Alpha-2-macroglobulin	M. rosenbergii	3.00E-74	2
MUC3A gene for intestinal mucin	H. sapiens	2.00E-14	1
Antioxidant defense systems
Manganese superoxide dismutase	M. rosenbergii	1.00E-141	1
CuZn-superoxide dismutase	Marchantia paleacea var. diptera	2.00E-07	1
Extracellular superoxide dismutase (SOD3)	M. musculus	7.00E-06	1
Cytosolic manganese superoxide dismutase	Macrobrachium nipponense	9.00E-119	1
Glutathione peroxidase 1	X. tropicalis	3.00E-11	1
Glutathione peroxidase	M. musculus	2.00E-23	3
Selenium-dependent glutathione peroxidase	M. rosenbergii	0.00E+00	1
Thioredoxin peroxidase 2	H. discus discus	3.00E-10	1
Selenoprotein M2	Chlamydomonas reinhardtii	5.00E-07	1
Selenoprotein K	Rattus norvegicus	2.00E-06	1
15 kDa selenoprotein	Tribolium castaneum	2.00E-42	1
Glutathione S-transferase	E. sinensis	2.00E-75	2
Glutathione S-transferase m	Laodelphax striatellus	9.00E-31	1
Glutathione S-transferase kappa 1	X. tropicalis	1.00E-05	2
Glutathione S-transferase pi 1	H. sapiens	9.00E-07	1
Peroxiredoxin 5	Nasonia vitripennis	2.00E-36	1
Peroxiredoxin 6	H. sapiens	9.00E-10	1
1-cys-peroxiredoxin	Brassica napus	5.00E-10	1
Peroxiredoxin 4	R. norvegicus	9.00E-10	1
Catalase B	Dictyostelium discoideum	1.00E-11	1
Catalase 1	Jatropha curcas	6.00E-07	1
Thioredoxin	Sebastes schlegelii	1.00E-07	1
Thioredoxin-related transmembrane protein 1	M. musculus	2.00E-10	141
Thioredoxin interacting protein	B. taurus	8.00E-06	1
Thioredoxin-like 5	Zea mays	2.00E-06	1
Alternative oxidase	Urechis unicinctus	1.00E-07	1
Myo-inositol oxygenase	X. laevis	2.00E-07	1
Chaperone proteins
Heat shock protein 70	Poecilia reticulata	3.00E-11	2
Heat shock protein 90 alpha	Anas platyrhynchos	1.00E-06	1
Heat shock protein 70 kDa family, member 13	H. sapiens	2.00E-07	1
HSP70/HSP90-organizing protein	X. laevis	5.00E-20	1
Heat shock 70kD protein 12B	H. sapiens	1.00E-12	1
DnaJ (Hsp40) homolog	R. norvegicus	4.00E-06	1
Calreticulin precursor	F. chinensis	0.00E+00	3
Chaperonin zeta subunit	Scylla paramamosain	2.00E-131	1
Co-chaperonin	Aedes aegypti	6.00E-06	1
T complex chaperonin, putative	Perkinsus marinus ATCC 50983	1.00E-119	1
Chaperonin containing TCP1, subunit 2	S. scrofa	1.00E-09	1
Chaperonin containing TCP1, subunit 4	H. sapiens	9.00E-11	1
Chaperonin containing TCP1, subunit 5	X. tropicalis	2.00E-07	1
Chaperonin containing TCP1, subunit 6a	M. musculus	8.00E-14	1
Chaperonin containing TCP1, subunit 8	H. sapiens	3.00E-06	1
Clottable proteins
Integrin, beta 5	Gallus gallus	3.00E-12	9
Integrin, alpha 11	H. sapiens	2.00E-10	1
A disintegrin and metallopeptidase domain 17	M. musculus	7.00E-08	1
Coagulation factor VII	D. rerio	3.00E-09	1
Patten recognition receptors
C-type lectin	L. vannamei	2.00E-37	1
C-type lectin 2	L. vannamei	8.00E-07	1
C-type lectin domain family 4, member a2	M. musculus	4.00E-09	1
Lectin B isoform 1	M. japonicus	2.00E-27	1
Osteoclast inhibitory lectin	R. norvegicus	2.00E-09	1
Toll-like receptor adapter molecule 1	M. musculus	1.00E-06	1
Molossinus toll-like receptor 5	M. musculus	2.00E-07	1
Toll-like receptor 5	X. laevis	2.00E-07	1
Ficolin	R. norvegicus	2.00E-10	1
Ficolin-like protein 1	P. leniusculussculus	4.00E-69	1
Ficolin (collagen/fibrinogen domain containing) 1	X. laevis	2.00E-11	1
Hemopexin	Ginglymostoma cirratum	5.00E-09	1
Other immune-related genes
Lysozyme	M. rosenbergii	9.00E-15	1
Lysozyme g	Ictalurus punctatus	4.00E-10	1
Lysozyme-like 6	H. sapiens	3.00E-06	1
Ferritin	M. rosenbergii	0.00E+00	2
14-3-3 protein	P. monodon	1.00E-44	9
14-3-3 protein beta/alpha-2	Salmo salar	5.00E-08	1
14-3-3 zeta gene	Harpegnathos saltator	6.00E-11	7
Cathepsin L	Palaemonetes varians	0.00E+00	5
Cathepsin D	P. monodon	2.00E-146	1
Cathepsin L2	Penaeus vannamei	2.00E-11	1
Ubiquitin-like 3	B. taurus	2.00E-08	1
Ubiquitin specific peptidase 8	M. musculus	5.00E-13	1
Ubiquitin specific peptidase 25	X. laevis	9.00E-07	1
Ubiquitin specific peptidase 12a	D. rerio	7.00E-08	1
Small ubiquitin-like protein	Pfiesteria piscicida	9.00E-23	1
Interleukin-10	Ctenopharyngodon idella	3.00E-23	2
Interleukin-1 beta	Conger myriaster	1.00E-08	1
Interleukin 8	B. taurus	9.00E-06	1
Hemolin	Samia cynthia ricini	2.00E-14	1
Phospholipase A2-1b	M. musculus	7.00E-13	1
Phospholipase D1	M. musculus	4.00E-10	1
Phospholipase A2, group VII	H. sapiens	4.00E-06	1
Phospholipase A2	Laticauda colubrina	7.00E-12	1
Patatin-like phospholipase domain containing 2	S. scrofa	3.00E-06	1
Arrestin, beta 2	H. sapiens	3.00E-10	8
Arrestin 1	Culex pipiens pallens	5.00E-07	1
Arrestin domain containing 3	S. scrofa	5.00E-12	1
CD200 antigen	M. musculus	3.00E-18	2
CD63 antigen	D. rerio	1.00E-09	1
Proteasome assembly chaperone 2	B. taurus	3.00E-11	1
Proteasome 26S subunit, non-ATPase,11	M. musculus	2.00E-08	2
Proteasome 26S subunit, non-ATPase, 1	Anolis carolinensis	2.00E-42	1
Proteasome subunit, beta type, 7	X. laevis	6.00E-09	1
Fibronectin 1	M. musculus	9.00E-11	1
Fibronectin type III domain containing 1	H. sapiens	2.00E-08	1
Calpain 1, (mu/I) large subunit	R. norvegicus	7.00E-07	1
Cyclophilin 2	Oryza sativa Japonica Group	5.00E-07	1
Prothymosin,alpha	H. sapiens	9.00E-06	1
Profillin	T. castaneum	1.00E-29	2
Fibrinogen C domain containing 1	X. laevis	4.00E-09	1
Defender against apoptotic death	P. monodon	7.00E-85	1
Subfamily 7a metallothionein 1	Tetrahymena hegewischi	3.00E-09	1
Matrix metalloproteinase 7	X. laevis	9.00E-17	1
Crustapain	P. borealis	1.00E-100	2
Oncoprotein nm23	L. vannamei	4.00E-132	1
Microneme protein	Toxoplasma gondii ME49	7.00E-57	2
Fibrillin 1	Nematostella vectensis	3.00E-26	1
Calmodulin	Chiloscyllium plagiosum	4.00E-10	14
Laminin,alpha 5	M. musculus	2.00E-08	1
Scavenger receptor class F, member 1	H. sapiens	4.00E-11	1
Selenium binding protein	Medicago sativa	8.00E-11	3
GST-like hemolymph protein	Corcyra cephalonica	3.00E-09	1
Suppressor of cytokine signaling 7	D. rerio	4.00E-08	1
Transferrin	S. scrofa	2.00E-07	2
Salivary secreted peptide	A. aegypti	1.00E-31	1
Valosin-containing protein	H. sapiens	9.00E-08	1
Immunity-related GTPase family, M	R. norvegicus	1.00E-06	1
Sperm associated antigen 6	B. taurus	2.00E-09	1
Antigen p97	M. musculus	1.00E-06	1
Beta-2-microglobulin	Sigmodon hispidus	4.00E-11	1
Cysteine-rich protein 1	Apis mellifera	9.00E-26	1

Discussion

The known genes which identified in our study have increased the current molecular knowledge for E. carinicauda. 41.6% of unknown sequences were founded in our library, due to the low sequence homology to existing known sequences in the public database [31]. The findings were also existed in Portunus trituberculatus and E. sinensis, which indicated that crustaceans possess a large number of unique genes [30]. For example, in the cDNA library of the swimming crab P. trituberculatus, unknown unique sequences were accounted for 64.6% of the total [30]. From the study of Chinese mitten crab E. sinensis, the unique sequences without homology were 58% [9] and 53% [31].

Through EST analysis, 14.5% of all 1053 unique sequences were homologous with genes reported as immune-related in other studies. This percentage was higher than that of previously reported cDNA library Chinese mitten crab E. sinensis [9,31]. In addition, 2.7% and 7.5% of total ESTs were isolated to be immune-related in the cDNA library of Panulirus japonicas [15] and F. chinensis [17]. According to the functions in the crustacean immune system, these immune-related genes found in our library were classified into seven categories, such as antimicrobial peptides, prophenoloxidase activating system, antioxidant defense systems, chaperone proteins, clottable proteins, pattern recognition receptors and other immune-related genes.

Antimicrobial peptides

Antimicrobial peptides (AMPs) are a micromolecule alkaline peptide, existing widely in all of the biospheres. These are important components of the innate immune systems of living organisms, playing an important role of defense to bacteria, fungi and virus. In our study, 17 ESTs (of 368) were related to antimicrobial peptides, such as anti-lipopolysaccharide factor (ALFs) (1 ESTs), antimicrobial peptide precursor (8 ESTs), crustin (1 EST), defensin (1 EST), hemocyanin (3 ESTs) and histone (3 ESTs).

ALFs, as an antimicrobial peptides, play an important role in the innate immune systems and can be binded to lipopolysaccharide (LPS). ALFs were initially founded in the hemocytes of P. monodon [32] and L. setiferus [6] through analysis of EST. At present, ALF homologous were identified in other shrimp species P. monodon [32], F. chinensis [33], M. japonicas [34] and L. vannamei [35].

Crustins are an 11.5 kDa cationic and hydroponic proteins, and have antimicrobial activity against Gram-positive bacteria. The crustins were isolated from a variety of shrimps, including L. vannamei [36], L. setiferus [36], Panulirus argus [37], M. japonicas [38], P. monodon [39], F. chinensis [40], Homarus gammarus [41] and Pacifastacus leniusculus [42]. However, there were no reports on crustin gene in E. carinicauda.

Defensins, as cationic atimicrobial peptides, have broad-spectrum antimicrobial activities. Denfensins are the most widespread family of invertebrate atimicrobial peptides, having been founded in various species. About over 300 defensins sequences have been reported. Defensins were first identified from P. japonicas [43], and two new isoforms of defensin PJD1 and PJD2 were founded in hemocytes of Penaeus japonicus. In our library, one transcript with similarity to defensin gene was founded. This finding suggested that defensin might be involved in immunity of E. carinicauda. In addition, three ESTs were identified as hemocyanin, these are presumptive factor of immune function in the shrimps.

Hemocyanin is a multi-subunit protein complex, which makes up the major part in the hemolymph of arthropods [44]. The function of hemocyanin is to carry oxygen, but it has recently been found to be implicated in the immune response [45] and may be a novel important type of non-specific innate immune defense molecule [44,46,47]. Histone is in the structure of chromatin, and has five types: H1, H2A, H2B, H3 and H4. Many public reports have proved that histones have an important role in anti-microbial defenses in a range of species, and they also constitute the innate immune system of crustaceans [48-52].

Prophenoloxidase activating system

The prophenoloxidase (proPO) activating system is one of the major innate immune system in crustacean, which defense reactions are accompanied by melanization [53]. The proPO system included prophenoloxidase (proPO), phenoloxidase (PO), serine proteinases (SP), pattern recognition receptor (PPRs) and proteinase inhibitors. The proPO system was controlled and regulated by many proteins such as prophenoloxidase activating enzymes and proteinase inhibitors [31]. The proPO system was first identified in Penaeus californiensis [54-56]. In this study, 28 ESTs were identified as proPO system, including prophenoloxidase (2 ESTs), serine proteinases (8 ESTs), serine proteinase inhibitors (18 ESTs).

Prophenoloxidase is an inactive form of phenoloxidase, and can be activated by the serine proteinase cascade and different chemical and microbial elicitors. ProPO is first founded in P. leniusculu [57]. In some shrimps, proPO was confirmed in P. monodon [58], L. vannamei [59], Macrobrachium rosenbergii [60] and F. chinensis [61]. In our study, two ESTs were discovered as prophenoloxidase in E. carinicauda.

Serine proteinases are an important composition in the activation of the prophenoloxidase system, and belong to the trypsin proteinase family. Serine proteinases include chymotrypsin, subtilisin, lactoferrin [62], mucin [62] and other enzymes. In our library, four forms of serine proteinases were discovered such as HtrA serine peptidase 3, musin, chymotrypsin and prophenoloxidase activating enzyme. Chymotrypsin is synthesised in the inactive precursor form of chymotrypsinogen and was identified in L. vannamei [63,64] and F. chinensis [65]. Previously, only two prophenoloxidase activating enzymes genes have been identified in P. leniusculus [66] and P. monodon [67]. Mucin is a macromolecular glycoprotein and used as antibacterial material for inhibiting pathogenic bacteria's growth and proliferation. In crustacean, Mucin gene was firstly identified in E. carinicauda.

Serine proteinase inhibitors are an important composition of host defense system, involving blood coagulation and pathogen digestion. Serine proteinase inhibitors family contains four types: kazal-type proteinase inhibitor, kunitz-type proteinase inhibitor, serpin and α-2-macroglobulin [68]. In our library, α-2-macroglobulin, kazal and serpin were all founded, but kunitz were not founded. In arthropod, kazal, serpin and α-2-macroglobulin were reported in shrimps, such as F. chinensis [69], P. monodon [70], Farfantepenaeus paulensis [71].

Antioxidant defense systems

Reactive oxygen species (ROS), including hydrogen peroxide, hydroxyl radical, superoxide anion radical, singleton oxygen, and peroxyl radical, etc, can through react with DNA, unsaturated fatty acids and proteins, causing oxdative damages to body. However, crustaceans have a set of antioxidant defense systems in body, which can protect themselves from damages of ROS. The antioxidant defense systems major include enzymatic antioxidant systems and non-enzymatic antioxidant systems, such as: (1) enzymatic antioxidant systems: GSH-peroxidase, catalase, superoxide dismutase, haem-oxygenase, etc; (2) non-enzymatic antioxidant systems: thioredoxin, glutathione, a-tocopherol, ascorbate, β-carotene, melanins, etc [72].

In our library, 171 ESTs with high homology to antioxidant defense systems were identified as superoxide dismutase (4 ESTs), glutathione peroxidase (4 ESTs), thioredoxin peroxidase (1 EST), glutathione S-transferase (6 ESTs), peroxiredoxin (2 ESTs), catalase (2 ESTs), thioredoxin (145 ESTs) (Table 3). In addition, 3 ESTs were identified as oxidation, such as selenoprotein (3 ESTs); Alternative oxidase (1 EST); Inositol oxygenase (1 EST). These genes are affirmative modulators of immune system in E. carinicauda.

Chaperone proteins

Chaperone proteins, as a family of protein molecules, have important physiological functions and highly conservative structure, interacting and keeping stability with other proteins. Chaperone proteins are usually synthesized under conditions of cellular stress, such as infection, heat, denaturant and oxidation. Some chaperone proteins, such as heat shock protein (HSP), calreticulin and chaperonin, were identified in our library (Table 3). Three kinds of HSPs (HSP 40, HSP 70, HSP 90) and Hsp 70/Hsp 90-organizing protein were found in library, which would contribute to research on immunity of E. carinicauda. HSP 70 and HSP 90 genes of E. carinicauda have been reported recently [20,21].

Coagulation system

Hemolymph coagulation is one part of the innate immune response, and can prevent leakage of hemolymph from sites of injury and dissemination of bacteria. Integrin and coagulation factor have been founded in the hemocytes cDNA library of E. sinensis, and have been reported to be involved in the blood coagulation system of crustacean [9]. In our library, integrin and coagulation factor were both founded (Table 3), which provided useful information for further study of coagulation and immunity system in E. carinicauda.

Pattern recognition receptors

Pattern recognition receptors (PRRs) can recognize non-self and danger signals in innate system of invertebrates. PRPs also can recognize pathogen associated molecule patterns (PAMPs), such as bacterial and fungal glycoproteins and lipopolysaccharides and intracellular components released through injury or infection [73]. PRRs include lectins, collectins, ficolins, pentraxins, scavenger receptors, β-1, 3-glucan binding proteins, toll-like receptors, complement receptors, lipopolysaccharide binding proteins, peptidoglycan binding proteins, etc [74].

Three recognition receptors, including lectins, toll-like receptors and ficolins, were founded in our cDNA library. Lectins play important roles in animal innate immune responses by serving as PRRs, opsonins, or effector molecules [73]. Because of the importance of lectins, cloning of lectins have been studied in shrimps [73,75]. In our library, we found two types of lectins: C-type lectin and lectin B. This finding will contribute to the study of pattern recognition receptors in crustacean. Toll-like receptors are type-1 transmembrane proteins, which ectodomains containing interspersed leucine-rich repeat motifs involved in the recognition of a growing variety of PAMPs [74]. Ficolins have fibrinogen-like and collagen-like domains that are involved in the first line of host defense against pathogens [76]. It has been proved that ficolins serves as a phagocytic receptor for microorganism recognition, and plays an important role in innate immunity by acting as an opsonin [77].

Other immune-related genes

Hemocytes play key roles in invertebrate immunity. In hemocyte cDNA library, many immune-related genes could be discovered. Besides the genes discussed above, some other immune-related genes were also identified, such as ferritin, lysozyme, proteasome, profillin, cyclophilin and 14-3-3 protein.

In summary, a hemocyte cDNA library was constructed to generate an EST collection of E. carinicauda, providing a reference for discovery of immune-related genes as well as a primary view of the transcript properties of this species. The library generated 2853 valid ESTs sequences, and assembly resulted in the identification of 1053 unique genes. Of all the 593 homology unique sequences, 153 unique sequences (28 contigs and 125 singlets) were identified as immune-related genes, and many of these immune genes were first reported in E. carinicauda. The sequence information will permit more detailed genomic research on E. carinicauda. Furthermore, the discovery of genes that related to immune defense enables us to investigate the mechanism of innate immunity in E. carinicauda.