Marine Biotechnology

, Volume 11, Issue 1, pp 24–44

Analysis of Genes Isolated from Plated Hemocytes of the Pacific Oyster, Crassostreas gigas

Authors

  • Steven Roberts
    • School of Aquatic and Fishery SciencesUniversity of Washington—Seattle
  • Giles Goetz
    • Great Lakes WATER InstituteUniversity of Wisconsin—Milwaukee
  • Samuel White
    • School of Aquatic and Fishery SciencesUniversity of Washington—Seattle
    • Great Lakes WATER InstituteUniversity of Wisconsin—Milwaukee
Original Article

DOI: 10.1007/s10126-008-9117-6

Cite this article as:
Roberts, S., Goetz, G., White, S. et al. Mar Biotechnol (2009) 11: 24. doi:10.1007/s10126-008-9117-6

Abstract

A complementary deoxyribonucleic acid library was constructed from hemocytes of Crassostrea gigas that had been plated on poly-lysine plates for 24 h. From this library, 2,198 expressed sequence tags (ESTs) of greater than or equal to 100 bp were generated and analyzed. A large number of genes that potentially could be involved in the physiology of the oyster hemocyte were uncovered. They included proteins involved in cytoskeleton rearrangement, proteases and antiproteases, regulators of transcription and translation, cell death regulators, receptors and their associated protein factors, lectins, signal transduction proteins, and enzymes involved in eicosanoid and steroid synthesis and xenobiotic metabolism. Based on their relationship with innate immunity, the expression of selected genes was analyzed by quantitative polymerase chain reaction in gills from bacterial-challenged oysters. Several genes observed in the library were significantly upregulated by bacterial challenge including interleukin 17, astacin, cystatin B, the EP4 receptor for prostaglandin E, the ectodysplasin receptor, c-jun, and the p100 subunit of nuclear factor-kB. Using a similar approach, we have been analyzing the genes expressed in trout macrophages. While there are significant differences between the types of genes present in vertebrate macrophages compared with oyster hemocytes, there are some striking similarities including proteins involved in cytoskeletal rearrangement, proteases and antiproteases, and genes involved in certain signal transduction pathways underlying immune processes such as phagocytosis. Finally, C. virginica homologs of some of the C. gigas genes uncovered in the ESTs were obtained by aligning the ESTs reported here, against the assembled C. virginica ESTs at the National Center for Biotechnology Information.

Keywords

HemocytesESTsPacific oysterC. gigasInnate immunityBacterial challenge

Introduction

In oysters, hemocytes are responsible for cell-mediated defense. Bivalve hemocytes are composed of several subclasses of cells that can be discriminated on the basis of microscopy, flow cytometry, and even functionality (Cheng 1996). The two major classes of oyster hemocytes that are generally recognized include granulocytes (containing cytoplasmic granules) and hyalinocytes (lacking granules).

In oysters, hemocytes are involved in nutrient digestion and transport, wound and shell repair, and internal defense against pathogens. A major defense exhibited by hemocytes involves the direct phagocytosis of antigens. During phagocytosis, the hemocyte may recognize or bind to an antigen by the presence of specific lectins either in the hemolymph or in the membrane of the hemocyte (Cheng 1996; Ford and Tripp 1996). Upon contact with antigens, hemocytes may also produce reactive oxygen species (ROS) that are believed to be important cytotoxic, antimicrobial factors (Adema et al. 1991). It is assumed that oyster hemocytes possess the biochemical components necessary for ROS production, and this is based on the detection of luminol or lucigenin-derived chemiluminescence when hemocytes are stimulated with agents such as zymosan or Perkinsus (Anderson 1999). While a number of studies have concentrated on the ability of oyster hemocytes to aggregate, encapsulate, and phagocytose antigens, very few investigations have looked at the biochemical products actually produced by hemocytes. It has been demonstrated that oyster hemocytes produce lysozyme (Yoshino and Cheng 1976), and several other enzymes have been shown to be associated with oyster hemocytes including lipase, β-glucuronidase, acid phosphatase, and aminopeptidase (Cheng and Rodrick 1975).

Recombinant deoxyribonucleic acid (DNA) approaches have been used to look at the gene products produced by various oyster tissues including hemocytes. A review of recent advancements in the field of bivalve genomics overall is provided by Saavedra and Bachere (2006). Specifically related to oysters, expressed sequence tag (EST) libraries have been made for eastern oyster (Crassostrea virginica) hemocytes (Jenny et al. 2002), for hemocytes obtained from Pacific oysters (Crassostrea gigas) challenged with bacteria (Gueguen et al. 2003), and for C. virginica and C. gigas challenged with Perkinsus marinus (Tanguy et al. 2004). A mixed tissue (including hemocytes) library and ESTs have also been produced for C. gigas (Tanguy et al. 2008). In the study on hemocytes from bacterial-challenged Pacific oysters, a highly expressed gene was a tissue inhibitor of metalloproteinase (TIMP) that has been shown to be very responsive to pathogen stimulation and wounding (Montagnani et al. 2001; Montagnani et al. 2001). Using a targeted gene approach, oysters have also been shown to produce defensin (Gueguen et al. 2003), transforming growth factor β (TGFβ; Lelong et al. 2007), and chitinase-like proteins (Badariotti et al. 2007). Some of these genes have been reported to participate in the immune response of the oyster to pathogen challenge. While there have been additional targeted gene isolations originating from the oyster hemocyte ESTs (Gonzalez et al. 2005; Gueguen et al. 2003), few immune-related genes have been reported in oysters from these or other libraries (Saavedra and Bachere 2006).

It appears that in past studies using oyster hemocytes for gene discovery, cells have been isolated from hemolymph by centrifugation and then used immediately for ribonucleic acid (RNA) isolation. Over the past several years, we have developed primary cell culture techniques to obtain trout macrophages from head kidneys (Mackenzie et al. 2003). These primary cell cultures have been used for EST isolation (Goetz et al. 2004a) and also to investigate pathogen recognition in fish (Iliev et al. 2005, 2006). The technique involves plating cells isolated from the head kidney on poly-lysine plates for incubation. Thus, only adherent cells are used for experimentation. In the present study, we used a similar protocol for Pacific oyster hemocytes to obtain adherent cells for the construction of a C. gigas hemocyte library. Based on the genes obtained so far, this library appears to have high gene complexity and a large number of genes that have the potential to be involved in immunity. For example, two clones of an interleukin 17 homolog were isolated from this library, and the C. gigas IL17 was subsequently shown to be highly regulated in hemocytes by bacterial stimulation (Roberts et al. 2008). However, in analyzing the ESTs obtained in this study, we also realized that there are, in fact, a number of potential genes in the existing oyster ESTs that have not been characterized or reported and that also appeared in the current hemocyte library. By assembling all of the ESTs currently available for C. virginica and using the contigs obtained as a searchable database, we were also able to obtain complementary DNAs (cDNAs) for the eastern oyster homologs of some of the C. gigas ESTs described here. These could be valuable for comparative studies of pathogen infection between the two species.

Materials and Methods

Animals and Hemocyte Plating

Pacific oysters (6–8 in.) were purchased from Taylor Shellfish Farms (Seattle, WA, USA) and held at 10°C in seawater until use. Shells were opened, and the hemolymph (8–10 ml) was obtained directly from the heart using a syringe. The hemolymph was placed in a plastic tube on ice and an additional volume of cold, sterile-filtered (22 μm) seawater containing 100 U/ml penicillin, and 100 μg/ml streptomycin was added (2 ml seawater/5 ml hemolymph). After gentle mixing, 6–7 ml of the hemolymph/seawater solution was spread onto 60-mm culture plates coated with poly-lysine (Becton Dickinson). Hemolymph was obtained from nine oysters, and each oyster provided enough hemolymph for two plates. All plates were incubated at 12°C for 24 h.

RNA Extraction and Library Construction

After 24 h, the plated hemolymph/saltwater solution was decanted and replaced with 1 ml of Tri Reagent (Molecular Research Center) per plate. Total RNA was extracted from the Tri Reagent according to the manufacturer’s protocol (Chomcynski 1993; Chomcynski and Sacchi 1987), and polyA+ RNA was isolated using the Poly-Atract messenger RNA (mRNA) isolation system (Promega). The mRNA obtained from the hemocytes was used to construct a cDNA library in Zap Express (Stratagene). cDNA produced for library construction was size-fractionated using sephacryl SF500, and the two largest cDNA size classes were ligated together with the Zap Express vector. After packaging and titering, the library was mass-excised to pBK-CMV phagemids and plated at low density. Individual colonies were randomly picked, and plasmid preparations were made using the RevPrep Orbit (GeneMachines). Plasmid preparations were sequenced from the 5′ end using the dideoxy chain termination method with “Big Dye Terminator” (Applied Biosystems) and the BK reverse vector primer. The reactions were precipitated and resuspended in “Hi-Di Formamide with EDTA” (Applied Biosystems) and run on an ABI Prism 3730 automated sequencer (Applied Biosystems).

Sequence Data Analysis

Sequence chromatogram files were trimmed for quality using phred (http://www.phrap.org/phrap.docs/phred.html), vector-screened using cross match (http://www.phrap.org/phrap.docs/phrap.html), and analyzed locally using (1) Blastx against the National Center for Biotechnology Information (NCBI) nonredundant protein database, (2) Blastn against the NCBI nucleotide database, and (3) Blastn against the existing C. gigas ESTs at NCBI. Sequences were analyzed for redundancy using CAP3 (Huang and Madan 1999) and were also annotated locally using the Gene Ontology Database (version GO.200801). Simple sequence repeats were identified using the simple sequence identification tool (http://www.gramene.org/db/searches/ssrtool; Temnykh et al. 2001).

All ESTs for C. virginica were downloaded from NCBI to our local cluster and were assembled using CAP3. The ESTs described in the current paper (Table 4) were aligned (Blastn) against the contigs generated from these assembled ESTs to obtain possible homologs for C. virginica. Since NCBI does not allow the third-party submission of sequences that do not have direct “wet bench” data to support their annotation, it was not possible to submit these assembled C. viriginica sequences to NCBI for accession numbers. Instead, we provide these assembled sequences, their translated amino acid products, and the ESTs that went into creating them in an Online Appendix to the paper.

Animals, Tissue Collection, and Bacterial Challenges

Pacific oysters (C. gigas) were obtained from Taylor Shellfish Farms and kept in the University of Washington holding facilities in 15°C seawater, until experimentation. In preparation for bacterial challenges, 20 oysters (ten control, ten challenge) were transferred to smaller tanks, both containing 3 l of 15°C seawater with air stones for circulation. The oysters were allowed to acclimate to ~20°C for 24 h prior to exposure. Challenges were conducted with two species of live bacteria: Vibrio vulnificus and Vibrio parahaemolyticus. Starter cultures of each species were grown separately in 100 ml Luria Bertani broth (LB) overnight at 37°C with shaking at 230 rpm. The following day, the starter cultures were combined and used to inoculate 1 l of LB. This large culture was grown at 37°C with shaking at 230 rpm until the optical density at 550 nm (OD550) = 0.410. One OD550 unit is equivalent to 5 × 108 bacteria per milliliter (Gueguen et al. 2003), thus resulting in ~2.05 × 1011 bacteria. The culture was centrifuged at 4,000 rpm for 30 min at 4°C to pellet the bacteria. The supernatant was removed, and bacteria were resuspended in 100 ml of ~20°C seawater. This suspension was added to one tank containing ten oysters. The other tank of ten oysters received 100 ml of ~20°C seawater. After 24 h of exposure, the oysters were removed from their tanks, and gill tissue was collected from all oysters and immediately frozen on dry ice. The samples were stored at −80°C until RNA extraction.

Quantitative PCR Analysis

Frozen gill tissue (50 mg) was homogenized in Tri Reagent (1 ml), and total RNA was extracted according to the manufacturer’s protocol (Chomcynski 1993; Chomcynski and Sacchi 1987). Total RNA was treated with TURBO DNA-free (Ambion) according to the manufacturer’s protocol to remove any possible genomic DNA carryover. The treated RNA samples were quantified and all samples diluted to 0.122 μg/μl. Removal of genomic DNA from the treated RNAs was verified via real-time polymerase chain reaction (PCR) using primers known to amplify genomic DNA (data not shown). First-strand cDNA synthesis was performed with avian myeloblastosis virus reverse transcriptase (Promega) according to the manufacturer’s protocol, utilizing oligo dT primers. The reverse transcription reactions each contained 0.61 μg of total RNA.

All real-time PCR reactions were created as master mixes, and individual reactions contained the following: 0.5 μL cDNA, 0.04 μM forward/reverse primers (Table 1), 2 μM SYTO-13 (Invitrogen), and 1× Immomix Master Mix (Bioline). Cycling and fluorescence measurements were carried out in an Opticon 2 System (Bio-Rad) with the following cycling parameters: one cycle of 95°C for 10 min, 40 cycles of 95°C for 30 s, 55°C for 1 min, and 72°C for 30 s. Fluorescence readings were taken at the end of each cycle. Immediately after cycling, a melting curve protocol was run. Temperature was increased from 55°C to 95°C at a rate of 0.2°C/s with fluorescence readings every 0.5°C increase, followed by an incubation of 21°C for 10 min. Negative controls containing water instead of cDNA template were run for each primer set.
Table 1

Primer sequences used for expression analysis of selected ESTs (Table 4) in gill tissue of bacterial challenged Pacific oysters

Gene

Accession number

Primer names

Primer sequences

NF-kappa-B p100 subunit

EW779317

NFkB p100 5′

ATCTGTGCCAGTTCCAATCC

NFkB p100 3′

TTCTTTTGCTCCATGGCTCT

Focal adhesion kinase

EW777781

FAK 5′

ATCACCAAGGCAACCTGAAC

FAK 3′

AATGGTGCTGGGAGTGTAGG

c-type lectin

EW778826

C-type lectin 5′

TTCAGCGCAAAAATGAATTG

C-type lectin 3′

AGCGCGTTTTCTGAAATTGT

c-jun protein

EW778895

C-jun 5′

CAAGGCAGTGTGAAGATGGA

C-jun 3′

CCAACACATGGGACTGTTTG

Astacin-like protein

EW779002

Astacin-like 5′

ACGCCCTAGTTGGATGAATG

Astacin-like 3′

ACTTGGTCTGGGGTTGTTTG

Tissue inhibitor of metalloprotease

EW777598

TIMP 5′

AACATCCGGTTTTGTTTCCA

TIMP 3′

GTCAGTGACGGCGAGTTCTT

Preprocathepsin C

EW778914

Preprocathep 5′

GCTGGAGGTTGCGAACTAAC

Preprocathep 3′

AAGGTCCGTTCTTCACCAAA

Serine-protease inhibitor

EW778389

SPI 5′

ATGGCCGATTGTTATGGTGT

SPI 3′

ACTGCAATTAGCCTGGCACT

Cystatin B

EW778247

Cystatin B 5′

GAGATTCCCCCTCACTCCTC

Cystatin B 3′

TGCTGAAAGCCTCCAAATCT

Prostaglandin E2 receptor, EP4 subtype

EW777722

PGER EP4 5′

ACCGAGAGTGCTGAGTGGTT

PGER EP4 3′

GGCAAACTGTAAGCCAGGAG

Ectodysplasin A2 receptor

EW778669

Ecdysplasin E2 5′

TGTTGATGTGGACCCAGTGT

Ecdysplasin E2 3′

CCGATTCCTGACCATTCTGT

CD45-like

EW777914

CD45-like 5′

ACAACAAGCCAAGGAACAGG

CD45-like 3′

TGATGTCTCCATGCGTCACT

Integrin beta-PS precursor

EW778853

Integrin 5′

GACGATTTGCTCCAACCATT

Integrin 3′

ACACTGGCACAAACCCTTTC

Tumor necrosis factor receptor-associated factor 3

EW779535

TNFRAF3 5′

CAAGCAACGAAAACAAAGCA

TNFRAF3 3′

AGGCTGGTGTTCAACCATTC

High-mobility group protein

EW778188

HMGP 5′

CAAGAAAGCCAAACCTCAGC

HMGP 3′

CTGGGAACCAATGCACTTTT

High-mobility group protein 1

EW778687

HMGP1 5′

TCATCAAAATGGCTGGTGAA

HMGP1 3′

ATGGACTGGCTTTGTTTTGG

Raw data were processed with Real-time PCR Miner (Zhao and Fernald 2005). Quantification was performed by calculating the relative mRNA concentration (R0) for each gene for each individual. Briefly, this was calculated using the following equation: \({{R0 = 1} \mathord{\left/ {\vphantom {{R0 = 1} {\left( {1 + E} \right)}}} \right. \kern-\nulldelimiterspace} {\left( {1 + E} \right)}}^{{\text{Ct}}} \), where E is the average gene efficiency and Ct is the cycle number at threshold. The R0 for each gene was normalized to a control (elongation factor 1) R0 from each individual. Using the normalized R0, fold increase over the minimum R0 value for each gene was calculated for all individuals (n = 20). All data were analyzed using one-way analysis of variance with the SPSS 13 software (SPSS). Data with a significance value less than or equal to 0.05 was considered to be statistically different.

Results and Discussion

General

Table 2 summarizes the characteristics of the C. gigas hemocyte cDNA library and the pertinent aspects of the ESTs that were derived from it. A total of 2,646 clones were sequenced, and from these, 2,198 ESTs greater than or equal to 100 bp were analyzed and submitted to NCBI (EW777381–EW779578). Following assembly with CAP3, there were 275 contigs of greater than or equal to two sequences and 987 singletons resulting in a redundancy of 55%. When aligned to the existing C. gigas ESTs at NCBI, there were 554 sequences (one of four of the total) that had a blastn E score of greater than or equal to 10−3. Based on this score, we consider these sequences new ESTs for C. gigas, and this is a very conservative estimate given our cutoff. There have been several other bivalve hemocyte libraries reported in the literature on various species including the oyster (Gueguen et al. 2003; Jenny et al. 2002; Quilang et al. 2007—mixed with other tissues), Manila clam (Kang et al. 2006), carpet-shell clam (Gestal et al. 2007), and mussel (Pallavicini et al. 2008). These libraries have reported redundancy rates from as low as 15.4% (Quilang et al. 2007) to highs of 72.7% (Gestal et al. 2007) and 78.5% (Pallavicini et al. 2008). Some of the studies (Jenny et al. 2002) did not report the percent redundancy, and strict comparisons of the rate of novel gene discovery and redundancy between studies may not always be appropriate since some ESTs were generated directly from sequencing cDNA libraries (Kang et al. 2006) while others sequenced suppression subtractive hybridization libraries (Gestal et al. 2007; Pallavicini et al. 2008) that certainly will have different rates of redundancy and sequence size that would affect assembly. Further, some libraries are normalized, and this obviously reduces the redundancy as shown in the libraries reported by Quilang et al. (2007; 15.4%) and by Tanguy et al. (2008; average 7%) for various bivalve species and tissues. The redundancy rate of the current study (55%) falls in the middle compared to other hemocyte investigations, but regardless, the rate of new gene discovery was high.
Table 2

Characteristics of C. gigas hemocyte library and ESTs

Characteristic

Value

Average library insert size

1,058 bp

Total number of ESTs sequenced

2,646

Total number of ESTs ≥100 bp

2,198

Phred quality score

13 (95%)

Average sequence length (all ESTs)

708 bp

Number of cDNA contigs

275

Number of cDNA singletons

987

Percent redundancy

55

Number of sequences with no Blastx hit (≥10−3)

876

Number of sequences with no significant blastn hit against C. gigas ESTs with a cutoff at 10−3

554

ESTs containing microsatellites with di-, tri-, and one tetranucleotide repeat were observed (Table 3). Several of these ESTs also had significant Blastx hits, including one EST (EW778460) that we highlight later in the paper (Table 4).
Table 3

ESTs with microsatellites including pertinent characteristics of the repeat region

EST accession number

Repeat sequence

Number of repeats

Start bp

Stop bp

Length of EST

Existing EST at NCBIa

Blastx hit [Species]

Identity score

EW778058

aac

6

410

427

671

 

Cathepsin Z [Sus scrofa]

130/172 (75%)

EW778818

aag

11

705

737

904

CU684921

No hit

 

EW778624

aag

11

690

722

886

CU684921

No hit

 

EW778956

ag

18

141

176

910

AM855122

No hit

 

EW779136

ag

14

190

217

873

 

No hit

 

EW777687

at

9

109

126

759

BQ426523

No hit

 

EW778246

ca

9

95

112

788

 

No hit

 

EW779536

caa

19

402

458

813

 

No hit

 

EW778460

caa

6

319

336

787

AM855218

Mnk [Aplysia californica]

90/121 (74%)

EW777448

ct

9

86

103

830

 

No hit

 

EW777476

ct

17

580

613

804

 

No hit

 

EW778377

ct

24

453

500

680

 

No hit

 

EW778359

ga

25

102

151

588

 

No hit

 

EW779138

gga

6

96

113

915

 

Nucleosome assembly protein [Danio rerio]

148/237 (62%)

EW778453

ggc

9

386

412

897

 

No hit

 

EW778928

ggc

9

582

608

835

 

RNA polymerase I [Gallus gallus]

58/163 (35%)

EW778400

ggc

9

339

365

823

 

No hit

 

EW778607

ggc

9

380

406

806

 

No hit

 

EW778589

tac

6

701

718

894

 

No hit

 

EW777557

tc

13

348

373

862

 

No hit

 

EW779289

tgga

6

725

748

797

AM858458

No hit

 

aExisting C. gigas ESTs at NCBI with identical sequences

Table 4

Selected genes from the C. gigas hemocyte cDNA library including accession numbers, size, and the most similar Blastx comparison to the Gene Ontology Database

Putative name/function

Partial cDNA accession number

C. virginica contiga

Size (bp)

Number in library

Blastx identities

Species most similar to

Accession number of similar protein

GO terms

GO description

Cell structure and motility

Cyclase-associated protein-1

EW778004

 

851

1

110/210 (52%)

Zebrafish

Q6YBS2

GO:0003785

Actin monomer binding

Phospholipid scramblase 1

EW778680

1,347

813

1

120/204 (58%)

Human

O15162

GO:0017121

Phospholipids scrambling

GO:0030168

Platelet activation

Cofilin (actin-depolymerizing factor 1)

EW777946

229

722

2

52/133 (39%)

Y. lipolytica

Q6C0Y0

GO:0003779

Actin binding

ARP2 (actin-related protein 2)

EW777744

 

784

2

215/253 (84%)

Zebrafish

NP_998664

GO:0005515

Protein binding

Proteases/antiproteases

Astacin-like protein

EW779002

 

915

7

147/281 (52%)

Pearl oyster

Q2VU37

GO:0008237

Metallopeptidase activity

Tissue inhibitor of metalloproteinase (TIMP)

EW777598

1,081

807

3

115/115 (100%)

Pacific oyster

Q9GPJ2

GO:0008191

Metalloendopeptidase inhibitor activity

Cathepsin

EW777605

1,402

791

1

33/41 (80%)

Lancelet

Q7YT27

GO:0004197

Cysteine-type endopeptidase activity

GO:0008233

Peptidase activity

Thimet oligopeptidase 1

EW77774

1,340

811

1

143/232 (61%)

Sea urchin

XP_799208

GO:0004222

Metalloendopeptidase activity

GO:0008191

Metalloendopeptidase inhibitor activity

GO:0008237

Metallopeptidase activity

Blastula protease 10 precursor

EW777902

 

864

1

95/231 (41%)

Common urchin

P42674

GO:0008233

Peptidase activity

GO:0008237

Metallopeptidase activity

Prepro-cathepsin C

EW778914

 

856

2

155/287 (54%)

Rainbow trout

Q64HY0

GO:0004197

Cysteine-type endopeptidase activity

GO:0008234

Cysteine-type peptidase activity

Cathepsin Z

EW778927

1,365

911

1

195/273 (71%)

X. tropicalis

Q5EAM1

GO:0004197

Cysteine-type endopeptidase activity

GO:0008234

Cysteine-type peptidase activity

Cathepsin-L-like cysteine peptidase

EW779434

1,402

760

1

143/257 (55%)

Mealworm

Q7YXL4

GO:0004197

Cysteine-type endopeptidase activity

ADAM metalloproteinase

EW778307

 

803

1

52/131 (39%)

A. aegypti

Q17E69

GO:0004222

Metalloendopeptidase activity

Matrix metalloproteinase

EW778009

 

739

1

44/173 (25%)

Sea urchin

NP_001028823

GO:0004222

Metalloendopeptidase activity

Serine protease inhibitor

EW778389

 

823

1

73/208 (35%)

Scallop

Q32TF4

GO:0004867

Serine endopeptidase inhibitor activity

Bone morphogenetic protein 1

EW778952

 

856

2

82/266 (30%)

X. tropicalis

Q28C16

GO:0005509

Calcium ion binding

GO:0008237

Metallopeptidase activity

Cystatin B like

EW778247

896

877

1

49/98 (50%)

Zebrafish

Q7ZUH6

GO:0004866

Endopeptidase inhibitor activity

Cytokines/lytic proteins

Interleukin 17 isoform D.

EW779217

 

840

2

56/173 (32%)

Rainbow trout

Q70I20

GO:0005125

Cytokine activity

Macrophage expressed gene

EW778608

 

894

1

135/243 (55%)

Abalone

ABP96718

Unknown

 

Receptors and associated proteins

Prostaglandin E2 receptor, EP4 subtype

EW777722

 

876

1

63/166 (37%)

Human

P35408

GO:0007186

G-protein-coupled receptor signaling

GO:0007188

G-protein signaling, coupled to cAMP

Receptor protein tyrosine phosphatase (CD45-like)

EW777914

 

909

1

70/198 (35%)

Catfish

Q6UNF4

GO:0004725

Protein tyrosine phosphatase activity

GO:0016787

Hydrolase activity

Scavenger receptor class F member 2 precursor

EW777983

 

846

1

29/99 (29%)

Mouse

P59222

GO:0005044

Scavenger receptor activity

Putative neuropeptide receptor

EW778447

 

905

1

29/89 (32%)

Flatworm

Q964E5

GO:0001584

Rhodopsin-like receptor activity

GO:0004930

G-protein-coupled receptor activity

GO:0004983

Neuropeptide Y receptor activity

Similar to tachykinin receptor

EW778555

 

907

1

51/210 (24%)

Urchin

XP_783390

GO:0001584

Rhodopsin-like receptor activity

GO:0004930

G-protein-coupled receptor activity

Ectodysplasin A2 receptor

EW778669

 

774

1

35/97 (36%)

Human

Q5VYX9

GO:0005031

Tumor necrosis factor receptor activity

Integrin beta-PS precursor (position-specific antigen beta chain)

EW778853

 

843

2

48/137 (35%)

Fly

P11584

GO:0004872

Receptor activity

GO:0050839

Cell adhesion molecule binding

Syntenin-1 (Syndecan-binding protein 1)

EW779127

1,409

804

1

132/239 (55%)

Rat

Q9JI92

GO:0005137

Interleukin-5 receptor binding

GO:0008093

Cytoskeletal adaptor activity

TNF receptor-associated factor 3

EW779535

 

809

1

122/238 (51%)

Lancelet

A2TK68

GO:0004872

Receptor activity

TNF receptor-associated protein 1

EW778409

 

626

1

134/210 (63%)

Chicken

NP_001006175

GO:0000166

Nucleotide binding

GO:0005524

ATP binding

Epidermal growth factor-like receptor

EW778003

 

816

1

157/217 (72%)

Mosquito

CAC35008

GO:0004714

Receptor protein tyrosine kinase activity

Lectins/immunoglobulins

Galectin 4-like protein

EW777685

 

765

2

90/144 (62%)

Abalone

A3FKF6

GO:0005529

Sugar binding

C-type lectin 1

EW778826

 

737

2

24/66 (36%)

Eel

Q8AXR7

GO:0005529

Sugar binding

Ficolin 3

EW778094

 

779

1

66/186 (35%)

Sea squirt

Q95P98

GO:0005102

Receptor binding

Lachesin 1

EW778576

 

815

1

79/260 (30%)

Grasshopper

Q26474

GO:0005515

Protein binding

Lachesin 2

EW778214

 

771

1

38/131 (29%)

Fly

Q24372

GO:0007156

Homophilic cell adhesion

GO:0007165

Signal transduction

Transcription/translation/cell cycle

Inhibitor of growth protein 3

EW779472

 

754

3

103/183 (56%)

Zebrafish

Q6TEM2

GO:0003677

DNA binding

GO:0005515

Protein binding

GO:0008270

Zinc ion binding

cAMP-responsive element binding protein 2

EW777507

 

841

1

42/71 (59%)

Sea hare

Q16946

GO:0003677

DNA binding

GO:0003700

Transcription factor activity

cAMP responsive element binding protein 3-like 2

EW777815

1,385

573

1

89/155 (57%)

Cow

NP_001096003

  

Erythroid differentiation-related factor 1

EW777674

 

750

1

79/215 (36%)

P. pygmaeus

Q5R9R1

GO:0006355

Regulation of transcription

Myocyte enhancer factor 2

EW777688

 

796

1

53/60 (88%)

Marine jellyfish

Q8T363

GO:0003677

DNA binding

GO:0003700

Transcription factor activity

GO:0043565

Sequence-specific DNA binding

High-mobility group protein

EW778188

1,309

783

1

92/116 (79%)

Pacific oyster

Q70ML6

GO:0003677

DNA binding

High-mobility group protein 1

EW778687

 

940

1

96/169 (56%

Snail

Q8ITG9

GO:0003677

DNA binding

Septin 11

EW778483

 

816

1

93/161 (57%)

X. laevis

Q66J62

GO:0000166

Nucleotide binding

GO:0005525

GTP binding

Mps One Binder kinase activator-like

EW778538

 

783

1

192/211 (90%)

Fly

Q95RA8

GO:0019207

Kinase regulator activity

Signal transduction proteins

RAB18, member RAS oncogene family

EW777589

 

799

3

160/205 (78%)

X. tropicalis

Q28D30

GO:0005524

ATP binding

GO:0005525

GTP binding

GO:0008134

Transcription factor binding

Rab GDP-dissociation inhibitor

EW778125

525

806

1

186/259 (71%)

A. aegypti

Q16KQ6

GO:0005093

Rab GDP-dissociation inhibitor activity

3′,5′-cyclic nucleotide phosphodiesterase-like

EW778398

 

880

1

47/104 (45%)

A. aegypti

Q16HU8

GO:0003824

Catalytic activity

GO:0004114

Cyclic AMP phosphodiesterase activity

Dual specificity protein phosphatase 10 (mitogen-activated protein kinase phosphatase)

EW778617

 

834

1

145/288 (50%)

Human

Q9Y6W6

GO:0004725

Protein tyrosine phosphatase activity

GO:0017017

MAP kinase phosphatase activity

MAP kinase-interacting serine/threonine kinase 1

EW778460

 

787

1

90/121 (74%)

A. californica

Q27SZ8

GO:0004674

Protein serine/threonine kinase activity

14-3-3 protein gamma (protein kinase C inhibitor protein 1)

EW778905

899

878

1

142/236 (60%)

Cow

P68252

GO:0003779

Actin binding

GO:0008426

Protein kinase C inhibitor activity

GO:0005159

Insulin-like growth factor receptor binding

cAMP responsive element binding protein-like

EW779405

 

872

2

138/154 (89%)

Pacific oyster

Q5Y1E2

GO:0043565

Sequence-specific DNA binding

GO:0003700

Transcription factor activity

GO:0046983

Protein dimerization activity

cAMP-dependent protein kinase regulatory subunit (N4 subunit of protein kinase A)

EW779098

 

886

1

221/260 (85%)

A. californica

P31319

GO:0000166

Nucleotide binding

GO:0008603

cAMP-dependent pK regulator

G protein beta subunit.

EW779259

 

818

1

265/272 (97%)

Pearl oyster

Q5GIS3

GO:0004871

Signal transducer activity

G protein alpha S subunit

EW779391

 

816

1

134/265 (50%)

Silkworm

NP_001093292

GO:0004871

Signal transducer activity

GO:0005525

GTP binding

Nuclear factor NF-kappa-B p100 subunit

EW779317

 

848

1

31/83 (37%)

Chicken

P98150

GO:0003700

Transcription factor activity

GO:0005515

Protein binding

Rho coiled-coil associated kinase alpha

EW779189

 

844

1

118/279 (42%)

Zebrafish

Q90Y37

GO:0000166

Nucleotide binding

GO:0004674

Protein serine/threonine kinase activity

Rho family small GTP-binding protein cdc42.

EW779284

504

797

1

171/191 (89%)

Aphid

Q6PW11

GO:0000166

Nucleotide binding

Focal adhesion kinase 1

EW777968

 

530

1

130/176 (73%)

X. laevis

AAA99456

GO:0004713

Protein-tyrosine kinase activity

Focal adhesion kinase

EW777781

 

899

1

73/128 (57%)

Mosquito

EAT43915

GO:0004713

Protein-tyrosine kinase activity

Growth factor receptor bound protein 2

EW778129

 

762

1

131/217 (60%)

Chicken

ABM91436

GO:0007242

Signaling

Mitogen-activated protein kinase-activated protein kinase 2

EW778383

 

838

2

156/212 (73%)

Human

EAW93531

GO:0000166

Nucleotide binding

GO:0004674

Protein serine/threonine kinase activity

Apoptosis

Seven in absentia

EW778477

 

861

2

214/228 (93%)

Eastern oyster

ABC95994

GO:0005515

Protein binding

GO:0008270

Zinc ion binding

Anamorsin (cytokine-induced apoptosis inhibitor 1)

EW779007

 

886

1

80/200 (40%)

Human

Q6FI81

GO:0006915

Apoptosis

GO:0006916

Antiapoptosis

BNIP-2

EW779117

346

875

1

133/259 (51%)

Mouse

Q52KR3

GO:0006915

Apoptosis

C-Jun protein

EW778895

1,449

862

1

86/246 (34%)

Fugu

Q800B5

GO:0043565

Sequence-specific DNA binding

GO:0003700

Transcription factor activity

Oxidative stress induced growth inhibitor 2

EW779296

 

641

1

90/146 (61%)

Human

Q9Y236

  

Phosducin-like protein 3.

EW779247

 

792

1

107/192 (55%)

Cow

Q0VCW8

GO:0005515

Protein binding

Inhibitor of apoptosis protein

EW777636

 

808

1

85/246 (34%)

Xenopus

NP_001082290

GO:0005515

Protein binding

Steroidogenic/eicosanoic/metabolic enzymes

Lipoxygenase 1

EW778068

 

779

1

96/262 (36%)

Human

O15296

GO:0016165

Lipoxygenase activity

GO:0016491

Oxidoreductase activity

Lipoxygnease 2

EW779143

 

876

2

98/299 (32%)

G. fruticosa

Q2N410

GO:0016165

Lipoxygenase activity

GO:0016491

Oxidoreductase activity

Lipoxygenase 3

EW778932

 

930

1

93/318 (29%)

Rat

P12527

GO:0016165

Lipoxygenase activity

GO:0016491

Oxidoreductase activity

15-hydroxyprostaglandin dehydrogenase

EW777901

 

824

2

96/248 (38%)

Chicken

XP_420526

GO:0016491

Oxidoreductase activity

Sterol 12-alpha-hydroxylase (CYP8B1)

EW778166

 

831

2

66/220 (30%)

Mouse

O88962

GO:0004497

Monooxygenase activity

GO:0016491

Oxidoreductase activity

GO:0020037

Heme binding

Cytochrome P450: 1A1-like

EW778340

 

903

1

73/276 (26%)

X. tropicalis

Q5FVX6

GO:0004497

Monooxygenase activity

GO:0016491

Oxidoreductase activity

Cytochrome P450: 2D28-like

EW779033

 

816

2

243/249 (97%)

Pacific oyster

ABO38814

GO:0004497

Monooxygenase activity

GO:0005506

Iron ion binding

GO:0016491

Oxidoreductase activity

Cytochrome P450: 17A1-like

EW779361

 

797

1

82/242 (33%)

Catfish

O73853

GO:0004497

Monooxygenase activity

GO:0004508

Steroid 17-alpha-monooxygenase activity

GO:0016491

Oxidoreductase activity

Cytochrome P450: 3A16-like

EW779213

 

348

1

41/84 (48%)

Mouse

Q64481

GO:0004497

Monooxygenase activity

GO:0016491

Oxidoreductase activity

Cytochrome P450: family 4 peptide-like

EW779105

1,150

842

1

101/288 (35%)

Zebrafish

Q6PH32

GO:0004497

Monooxygenase activity

GO:00016491

Oxidoreductase activity

Cytochrome P450

EW779500

 

721

1

60/230 (26%)

A. aegypti

Q16Y74

GO:0004497

Monooxygenase activity

GO:0005506

Iron ion binding

GO:0020037

Heme binding

Cytochrome P450 2C20 (CYPIIC20)

EW777670

 

804

1

54/122 (44%

Macaque

AAB24950

GO:0004497

Monooxygenase activity

GO:0005506

Iron ion binding

GO:0016491

Oxidoreductase activity

Cytochrome P450, family 4, subfamily f, polypeptide 16

EW777481

 

803

1

122/266 (45%)

Mouse

NP_077762

GO:0004497

Monooxygenase activity

GO:0005506

Iron ion binding

GO:0016491

Oxidoreductase activity

Insulin-induced gene 2 protein (INSIG-2)

EW779367

 

726

4

119/195 (61%)

Mouse

Q91WG1

GO:0006629

Lipid metabolic process

GO:0008202

Steroid metabolic process

GO:0008203

Cholesterol metabolic process

Others

Heat shock protein 25

EW777519

1,405

836

1

32/83 (38%)

Zebrafish

Q645R1

GO:0009408

Response to heat

Heat shock protein 12B

EW777988

 

887

2

55/181 (30%)

Zebrafish

Q0R4G9

GO:0002040

Sprouting angiogenesis

GO:0048514

Blood vessel morphogenesis

Heat shock protein 90

EW777936

377

678

2

199/199 (100%)

Pacific oyster

ABS18268

GO:0005524

ATP binding

GO:0051082

Unfolded protein binding

Heat shock protein 70

EW778010

1,395

924

9

291/303 (96%)

Pacific oyster

Q9XZJ2

GO:0000166

Nucleotide binding

GO:0005524

ATP binding

Oxidative stress protein

EW778471

 

798

1

83/165 (50%)

Moon jelly

Q5EN85

GO:0008270

Zinc ion binding

Cavortin

EW778149

 

728

1

158/192 (82%)

Pacific oyster

Q5QGY9

GO:0004785

Copper/zinc superoxide dismutase activity

GO:0005507

Copper ion binding

GO:0008270

Zinc ion binding

Dual oxidase 1.

EW778215

 

853

1

193/282 (68%)

Urchin

Q5XMJ0

GO:0004601

Peroxidase activity

GO:0005506

Iron ion binding

GO:0005509

Calcium ion binding

GO:0016174

NAD(P)H oxidase activity

Histone 3

EW778115

 

659

1

136/136 (100%)

Human

NP_002098

GO:0003677

DNA binding

Lysosomal phospholipase

EW779151

 

805

1

103/222 (46%)

Human

BAD96510

GO:0004622

Lysophospholipase activity

GO:0005543

Binding

aC. viginica homologs in Appendix 1

Following annotation, the greatest proportion of the ESTs was in protein metabolism and synthesis and metabolism in general (Fig. 1). Included in these categories were ribosomal proteins that, not surprisingly, made up a large proportion of the ESTs. Actin (e.g., EW778950) was the most frequently observed structural EST followed by collagen (e.g., EW778483), tubulin (e.g., EW779030), and a calponin/transgelin-like cDNA (e.g., EW779129). ESTs involved in metabolism or biosynthesis that were frequently observed included ferritin (e.g., EW778702), NADH dehydrogenase 5 (e.g., EW779040), cytochrome b (e.g., EW778907), and elongation factor (e.g., EW777741). A total of 876 ESTs had no annotation when analyzed by Blastx against the Gene Ontology (GO) Database. These ESTs were not included in the GO analysis shown in Fig. 1.
https://static-content.springer.com/image/art%3A10.1007%2Fs10126-008-9117-6/MediaObjects/10126_2008_9117_Fig1_HTML.gif
Fig. 1

Categorization of ESTs into cellular processes derived from Blastx comparisons to the Gene Ontology Database (version GO.200801). ESTs without annotation were not included in this analysis

A subset of ESTs are provided in Table 4 including size, accession number, sequence similarity at the protein level, GO annotation, and where available, C. virginica homologs that are provided in the Online Appendix. The ESTs presented in Table 4 were chosen based on their possible relationship with important hemocyte functions.

Cell Structure and Motility

Locomotion, phagocytosis, and the regulation of cell shape are crucial elements of hemocyte function in oysters (Cheng 1975; Fisher 1986). Therefore, the actin cytoskeleton and its reorganization are fundamental to the function of the hemocyte as they are to vertebrate granulocytes. Sequences were observed that had high identity to genes involved in actin cytoskeletal organization. A cDNA (EW777744) was observed that was similar to Arp3, a protein that, together with Arp2, form a complex that is involved with actin polymerization (Welch et al. 1997b). Arps are found in a diverse group of eukaryotes (Welch et al. 1997a), and the Arp2/3 complex regulates the assembly of new actin filament networks at the leading front of moving cells that drives cell motility and chemotaxis (Jones 2000). An EST (EW777946) was also found that had sequence similarity to cofilin, an actin-binding protein that controls actin assembly and has also been reported in Manila clam hemocytes (Kang et al. 2006). Finally, an EST (EW778680) was observed that had high sequence similarity to phospholipid scrambalase. Scrambalase is a palmitoylated, lipid raft-associated endofacial plasma membrane protein that accelerates bidirectional movement of plasma membrane phospholipids during conditions of elevated calcium (Zhou et al. 1997). Scrambalase is stimulated by cytokines such as stem cell factor and granulocyte colony-stimulating factor, suggesting that it functionally contributes to cytokine-regulated cell proliferation and differentiation during myelopoiesis (Zhou et al. 2002).

Proteases and Antiproteases

A number of ESTs were obtained that had sequence similarities to known proteases and protease inhibitors. Proteases are specifically classified into groupings based primarily on the activity of their active site, their preferred substrate, and the pH range of their proteolytic activity. As such, four classes are observed in mammalian leukocytes including serine, metalloproteinases, cysteine, and aspartic proteinases (Owen and Campbell 1999). The ESTs observed in the C. gigas hemocyte library covered nearly all of these protease classifications. Several cathepsins were observed in the library including cathepsin C (EW778914), Z (EW778927), and L (EW779434). Cathepsin C (dipeptidyl peptidase I) is recognized as a multifunctional protease that is essential for the activation of other enzymes (Turk et al. 2001). Cathepsin L is involved in the degradation of the invariant chain of the major histocompatibility class II complexes (Nakagawa et al. 1998) but is also capable of generating kinin from kininogen and, therefore, may act as a kininogenase at inflammatory sites (Desmazes et al. 2003). Cathepsin Z (also called cathepsin Y) is a cysteine endopeptidase originally isolated from rat blood, which produces bradykinin-potentiating peptide from rat plasma and thereby has a dramatic effect on the action of kinins (Nakazono et al. 2002; Sakamoto et al. 1999).

In invertebrates, there have been investigations on the function of several cathepsins. However, these have been primarily related to digestion (Cristofoletti et al. 2005) or other very specific functions such as the role of cathepsin L in early development in brine shrimp (Liu and Warner 2006) and cathepsins B and D in oyster development (Donald et al. 2003). To our knowledge, there have not been reports linking cathepsins to immune functions in bivalves, though past EST investigations on oyster hemocytes have reported the presence of cathepsin B (Jenny et al. 2002), cathepsin Z (Jenny et al. 2002), and cathepsin L (Gueguen et al. 2003). Interestingly, one of the most highly upregulated genes in mussels exposed to metal or organic mixtures is cathepsin L (Venier et al. 2006).

Besides cathepsins, ESTs for two members of the astacin metalloproteinase family were observed. This included an EST (EW779002) that was most similar to a recently cloned astacin metalloproteinase that was upregulated by lipopolysaccharide (LPS) challenge in the pearl oyster (Pinctada fucata; Xiong et al. 2006). This protease was also significantly upregulated by bacterial challenge in oysters in the current study (Fig. 2). The astacin family also includes proteases such as bone morphogenetic protein 1 (BMP-1), and an EST (EW778952) for BMP-1 was also found in the C. gigas hemocyte library. BMP-1 has a number of functions in vertebrates including the formation of the extracellular matrix, the activation of latent complexes of certain TGFβ superfamily members, and the cleavage of specific proteins (e.g., prolactin) to form angiogenic factors (Ge et al. 2007). ESTs for several metalloproteinases were also observed including a matrix metalloproteinase (EW778009) and a cDNA (EW778307) for an ADAM (“a metalloprotease and disintegrin”) metalloproteinase. The ADAM metalloproteinases are particularly intriguing since they are one of the major factors involved in the proteolytic release of extracellular domains from membrane-bound precursors such as cytokines, receptors, and growth factors (Huovila et al. 2005).
https://static-content.springer.com/image/art%3A10.1007%2Fs10126-008-9117-6/MediaObjects/10126_2008_9117_Fig2a_HTML.gifhttps://static-content.springer.com/image/art%3A10.1007%2Fs10126-008-9117-6/MediaObjects/10126_2008_9117_Fig2b_HTML.gif
Fig. 2

Messenger RNA expression levels of selected ESTs in gill tissue of Pacific oysters challenged with bacteria. Bars represent the means of ten replicate oysters ± standard error. Asterisks indicate genes for which there was a significant difference (p < 0.05) between control and bacterial-challenged oysters

ESTs for several protease inhibitors were observed, including the TIMP (EW777598). This TIMP has been extensively studied in the Pacific oyster and found to be highly and rapidly upregulated following bacterial challenge or shell damage (Montagnani et al. 2001), and it was also upregulated in the oysters challenged by bacteria in the present study (Fig. 2). In addition, an EST (EW778389) was observed for a protease inhibitor with highest sequence identity to a novel serine protease inhibitor recently described in the bay scallop (Argopecten irradians; Zhu et al. 2006). In the bay scallop protein, the inhibitor exhibits structural domains characteristic of Kazal-type serine protease inhibitors, and the expression of the inhibitor is upregulated by bacterial challenge and injury. Finally, an EST (EW778247) for a cysteine protease inhibitor most similar to cystatin B was observed. Cystatin B has been observed previously in oyster (Jenny et al. 2002) and Manila clam (Kang et al. 2006) hemocytes, and this gene was significantly upregulated by bacterial challenge in the oysters in the present study (Fig. 2).

When we aligned the ESTs for proteases and protease inhibitors against the assembled C. virginica contigs, we found homologs for thimet oligopeptidase, cathepsin Z, cathepsin L, and cystatin B (see Table 4). However, the most interesting observation was a C. virginica contig (1081—Online Appendix) that showed high identity to the C. gigas TIMP.

Cytokines/Lytic Proteins

While there are comparative immunohistochemical and experimental data to suggest that cytokines are present in invertebrates (Raftos and Nair 2004), several large-scale genomic analyses of sequenced genomes (S. purpuratus, D. melanogaster, C. intestinalis) have generally not observed chemokines (Devries et al. 2006) or helical cytokines (Huising et al. 2006). The exception is the presence of tumor necrosis factor (TNF) homologs (Robertson et al. 2006) and a number of potential interleukin 17 (IL17) homologs (Hibino et al. 2006). In the C. gigas hemocyte library, we found two ESTs (EW779217; EW779442) that aligned to vertebrate IL17. In mammals, six forms of IL17s, labeled A–F, have been described (Moseley et al. 2003). The C. gigas IL17 was most similar to the IL17D form and was found to be upregulated in hemocytes rapidly following exposure to bacteria (Roberts et al. 2008).

An EST (EW778608) was obtained for a cDNA with high identity to the macrophage expressed gene (MEG), recently identified in several species of abalone (Mah et al. 2004; Wang et al. 2008). The proteins encoded by MEG in mammals and in abalone have sequence similarity to perforin (Spilsbury et al. 1995; Wang et al. 2008) and, therefore, may be involved in direct cell killing. The possible involvement of MEG in the immune response in invertebrates is further supported by the observation that it is upregulated by bacterial challenge in gastropods (Wang et al. 2008).

Receptors and Associated Proteins

TNF is a well-characterized proinflammatory cytokine in vertebrates that has been demonstrated to affect the growth, differentiation, and survival of immune and nonimmune cells (Goetz et al. 2004b). It is a member of the “TNF ligand superfamily” that includes a number of ligands in addition to TNF. These ligands interact with specific receptors, for example, TNF with TNF receptor 1 (TNFR1) and TNF receptor 2. TNFR1 binds intracellularly with TNF receptor-associated protein 1 (TRAP1; Song et al. 1995), and an EST (EW778409) for TRAP1 was identified in the C. gigas hemocyte library. Gene models for TNF and TNF receptors have been identified in sequenced genomes of invertebrates (Robertson et al. 2006).

Besides TNFR1 and 2, a large number of other receptors have been characterized for the TNF ligand superfamily members including, for example, the ectodysplasin receptor (Ware 2003) for which an EST (EW778669) was observed in the present library and was significantly upregulated in oyster gills by bacterial stimulation (Fig. 2). TNF ligand receptors contain one or more TNF-associated factor (TRAF) motifs in their cytoplasmic domains that interact with TRAF proteins in producing a cellular effect (Dempsey et al. 2003). In mammals, there are six different TRAFs (TRAF1–6), and some appear to be evolutionarily conserved across vertebrates and invertebrates. In the hemocyte ESTs, a cDNA (EW779535) was observed that was most similar to TRAF3, and this gene was significantly elevated by bacterial challenge (Fig. 2). TRAF3 is one of the evolutionary-conserved TRAFs (Dempsey et al. 2003), and interestingly, it has been demonstrated that TRAF3 is essential for the stimulation of type I interferon by all viral cellular recognition systems in mammals (Saha and Cheng 2006).

Perhaps one the most interesting ESTs observed in the hemocyte library was a cDNA (EW777722) similar to the EP4 subtype prostaglandin E2 receptor. Prostaglandin E2 is both proinflammatory and anti-inflammatory in mammals, and it is believed that the anti-inflammatory effects of PGE2 are mediated by the EP4 receptor subtype (Minami et al. 2008; Takayama et al. 2006). In mammalian macrophages, LPS upregulates the EP2 receptor subtype but downregulates the EP4 PGE2 receptor (Ikegami et al. 2001). The C. gigas EP4 receptor mRNA was very strongly upregulated in oyster gills by bacterial stimulation (Fig. 2).

The library contained an EST (EW777983) with similarity to scavenger receptors and the Drosophila draper, a gene believed to be involved in the phagocytosis of apoptotic cells in hemocytes (Manaka et al. 2004). An EST (EW777914) with identity to the vertebrate CD45 gene was observed. CD45 is expressed on all hematopoietic cells in vertebrates and is a large protein of the “receptor-type protein tyrosine phosphatase” family. While CD45 has been observed in jawed and jawless vertebrates (Uinuk-Ool et al. 2002), to our knowledge, this would be the first occurrence in an invertebrate.

Lectins/Immunoglobulins

Several ESTs were observed that are similar to lectins. Lectins are proteins that bind specific sugar moieties, and in immune systems, they frequently bind carbohydrate moieties on pathogens. For example, a lectin was recently reported that was responsible for recognizing the parasite, P. marinus, by the Eastern oyster hemocyte (Tasumi and Vasta 2007). A cDNA (EW777685) for galectin 4 was observed in the C. gigas hemocyte library. This EST was most similar to an abalone galectin but also was very similar to recently described galectins from the freshwater snail, Biophalaria glabrata (Yoshino et al. 2008), and the argasid tick, Ornithodoros moubata (Huang et al. 2007). Both of those are tandem repeat galectins. The C. gigas galectin 4 EST did not have similarity to the galectin described by Tasumi and Vasta (2007); however, another EST (EW779109) from the library did align at the amino acid level to the C. virginica galectin and could be the C. gigas homolog.

A cDNA (EW778094) was also observed for a ficolin that most closely resembled the ficolin characterized from the ascidian, Halocynthia roretzi (Kenjo et al. 2001). Ficolins are lectin proteins that contain a collagen-like domain and a fibrinogen-like domain and have been implicated in pathogen recognition for phagocytosis (Matsushita et al. 1996). An EST (EW778826) most similar to a galactose-binding C-type lectin in the eel (Mistry et al. 2001) was also observed. Besides lectins, two different lachesin-like cDNAs (EW778576, EW778214) were observed in the library. Lachesins are interesting molecules since they contain immunoglobulin-like domains. The lachesin ESTs described here contained two conserved immunoglobulin domains also found in neural cell adhesion molecules, fasciclin II, and the insect immune protein hemolin (Karlstrom et al. 1993).

Transcription/Translation/Cell Cycle

Several putative transcription factors and molecules involved in regulating the cell cycle were identified in the ESTs. A cDNA (EW779472) with sequence similarity to inhibitor of growth 3 (ING3) was obtained. ING3 is a member of the ING family of tumor suppressors that regulate the cell cycle, apoptosis, and DNA repair. Two ESTs (EW777507, EW777815) with homology to cyclic adenosine monophosphate (cAMP)-responsive element-binding (CREB) proteins were identified. CREB proteins bind to the cAMP response element and are involved in regulating transcription.

An EST (EW777674) for a transcription factor, erythroid differentiation-related factor 1 (EDRF1), was identified in the library. This gene was reported to be involved in erythroid differentiation and the upregulation of the globin gene in mammals (Wang et al. 2002). Interestingly, respiratory proteins have recently been shown to be involved in the antimicrobial defense in humans and horseshoe crabs (Jiang et al. 2007). Jiang et al. (2007) found that respiratory proteins were activated by microbial proteases to produce ROS. Thus, a gene like EDRF1 could be involved in regulating hemocyte numbers and/or composition and, in the process, antimicrobial activity.

A cDNA (EW777688) for myocyte enhancer factor (MEF2) was in the C. gigas library. MEF2 is well conserved across vertebrates and invertebrates (Shiomi et al. 2005), and while it has generally been associated with the control of early muscle development (Black and Olson 1998), it is produced by other cells as well. For example, LPS (Gram-negative bacterial LPS) increased the activation of MEF2C and c-jun in mammalian monocytes, suggesting that it may have an important roll in inflammation (Han et al. 1997).

Finally, ESTs (EW778188, EW778687) with sequence similarity to high-mobility group (HMG) proteins, were obtained from the C. gigas library. One of the HMGs was most similar to a previously submitted C. gigas HMG sequence (accession no. Q70ML6); however, the EST reported here is not the homolog since the two are only 79% identical at the nucleotide level. Proteins within this family bind the minor groove of DNA and are involved in responding to pathogens and removing damaged cells (Dumitriu et al. 2005).

Signal Transduction

When macrophages encounter foreign molecules, extracellular receptors are activated which initiate a highly regulated signaling process involving an array of intracellular proteins that are coordinated to transmit this information to the nucleus resulting in a specific cellular response. In phagocytosis, an assortment of signaling takes place to coordinate membrane and cytoskeleton rearrangement. G-protein-coupled receptors are activated which initiate the phosphorylation of numerous proteins, including the Rab family and focal adhesion kinase (FAK). The Rab family of proteins plays key roles in regulating the reorganization of membrane properties via lipid rafts (Hashim et al. 2000). FAK, controlled by the Arp2/3, regulates actin assembly during phagocytosis (Serrels et al. 2007). Additionally, FAK has been shown to interact with Rho GTPases, possibly activating these molecules (Zhai et al. 2003). Not only did we find an EST for Arp2 (discussed above), we also found two ESTs (EW777968 and EW777781) with similarity to FAKs. One of these kinases was slightly elevated upon bacterial challenge (Fig. 2). We also obtained an EST (EW779284) for a Rho family small guanosine triphosphate (GTP)-binding protein, and of the Rab family of proteins, we found a cDNA (EW777589) for Rab18. Rab18 is involved in reorganizing cell membranes in Salmonella (Hashim et al. 2000).

The nuclear factor-kB (NF-kB) protein complex is involved in cellular responses to stimuli such as stress, cytokines, free radicals, and bacterial or viral antigens (Gilmore 1999). These proteins are translocated from the cytoplasm to the nucleus where they act as transcription factors for a variety of genes involved in cell proliferation and inflammation (Legarda-Addison and Ting 2007). In the NF-kB family, we observed a cDNA (EW779317) of the p100 subunit homolog that was significantly upregulated upon bacterial challenge (Fig. 2).

The mitogen-activated protein kinase (MAPK) signaling pathway is involved in phagocytosis and the prophenoloxidase cascade in invertebrates (Lamprou et al. 2007). In the oyster library, we identified ESTs for genes involved in the MAPK signaling pathway including MAPK-activated protein kinase 2 (EW778383), MAPK-interacting serine/threonine kinase 1 (EW778460), and dual specificity protein phosphatase 10 (EW778617), a factor that acts on p38, JNK, and ERK and appears to be involved in enhancing the innate immune response (Pulido and Van Huijsduijnen 2008). Finally, an EST (EW778398) was found with homology to 3′,5′-cyclic nucleotide phosphodiesterase, a key enzyme that degrades cAMP and, thus, regulates cAMP levels.

Apoptosis

Apoptosis or programmed cell death is important for the homeostatic restoration of the number of immune cells at the termination of an immune response (Droin et al. 2003).

In the sequencing of the C. gigas hemocyte cDNA library, several homologs of genes involved in apoptosis were identified. ESTs (EW778477) were observed for seven in absentia (SIAH) that induces cell growth arrest. In humans, SIAH is a downstream effector of p53 which functions to suppress cell growth (Matsuzawa et al. 1998). An EST (EW779007) for cytokine-induced apoptosis inhibitor (also known as anamorsin) was also found in the library. Anamorsin inhibits programmed cell death following stimulation by cytokines such as IL3 (Shibayama et al. 2004). A Bcl-2/adenovirus E1B 19 kDa interacting protein (BNIP) domain containing cDNA (EW779117) was sequenced. BNIPs are a proapoptotic subgroup of the Bcl-2 family and have previously been found in several taxa including C. elegans (Zhang et al. 2003). Another Bcl-2 family member, BAD, interacts with the 14-3-3 proteins in the regulation of apoptosis (Fu et al. 2000). 14-3-3 proteins are highly conserved and comprise a complex family that contains seven distinct isoforms in vertebrates (Aitken et al. 1995). An EST (EW778905) for the 14-3-3 protein gamma was observed in the hemocyte library. Finally, a cDNA (EW778895) for c-jun was observed, and it was significantly upregulated by bacterial challenge (Fig. 2). C-jun is a downstream target of the JNK signaling pathway activated by mitogen-activated kinases (Weston and Davis 2002). This pathway is stimulated by cytokines and exposure to environmental stress, and JNK activation has been observed, for example, in Mytilus galloprovincialis under elevated holding temperatures (Anestis et al. 2007). A c-jun EST was also reported in hemocytes of Manila clams (Kang et al. 2006).

Steroidogenic/Eicosanoic/Metabolizing Enzymes

A large number of ESTs were observed in the C. gigas library that had sequence similarity to cytochrome P450 enzymes. Cytochrome P450 is a large superfamily of enzymes that catalyze many reactions involved in the metabolism of xenobiotics and the synthesis of cholesterol, steroids, and other lipids. Several P450 genes have been previously observed in bivalve ESTs (Tanguy et al. 2004), but as previously noted, these enzymes have not received very much attention in invertebrate studies (Tanguy et al. 2008).

Following sequence assembly of the ESTs from C. gigas hemocytes, there were at least nine unique cytochrome P450 gene products shown in Table 4. The nomenclature system for genes in the cytochrome P450 superfamily involves the use of “CYP” followed by a number indicating the gene family, a letter indicating the subfamily, and an additional number indicating the individual gene. From the first family, one cDNA (EW778340) was identified, as being most similar to CYP1A1. CYP1A1 is also known as aryl hydrocarbon hydroxylase and is involved in the activation of aromatic hydrocarbons. CYP1A1 is regulated in response to exposure to aromatic hydrocarbons and is therefore often used as a biomarker in aquatic organisms (Chaty et al. 2004; Mcclain et al. 2003). A second EST (EW777670) was most similar to the CYP2C subfamily. Proteins in this subfamily are primarily involved in xenobiotic and steroid metabolism. The only EST (EW779033) with significant homology to an existing C. gigas P450 cDNA also belongs in the CYP2 family. There was one EST (EW779213) that was most similar to the CYP3 family and two ESTs (EW779105; EW777481) that likely belong to family 4. Finally, another cytochrome P450 cDNA (EW778166) had high sequence similarity with both the CYP7 and CYP8 families. Interestingly, members of these families are involved in bile acid biosynthesis. While there is limited information on a relationship between bile acid and pathogens outside of nonmammalian systems, bile is a primary stressor for pathogens and has been shown to have antimicrobial activity (Begley et al. 2005).

CYP17A1 is an enzyme which acts on pregnenolone and progesterone in mammalian systems to convert pregnenolone and progesterone to their 17α-hydroxylated products and subsequently to dehydroepiandrosterone and androstenedione, catalyzing both the 17α-hydroxylation and the 17,20-lyase reaction. An EST (EW779361) for this gene was observed in the library. Recently, a CYP17 homolog was discovered in the amphioxus (Mizuta and Kubokawa 2007), and investigators have identified components of a sex steroid pathway in C. gigas (Matsumoto et al. 2003, 2007). These data suggest CYP17A1 could have similar steroidogenic function across taxa including oysters.

ESTs for several enzymes possibly involved in the eicosanoid synthetic pathway were observed in the library. These included three separate lipoxygenase cDNAs (EW778068, EW779143, EW778932) that did not assemble with CAP3, but all had similarity to several lipoxygenases depending on the species comparison, including arachidonic 5, 8, and 15 lipoxygenases. Lipoxygenases are enzymes that catalyze the oxygenation of polyunsaturated fatty acids to corresponding hydroperoxy derivatives (Brash 1999). Depending on the particular conditions and related enzymes in a cell, lipoxygenase activity can result in the ultimate production of a diverse number and type of molecules including leukotrienes, lipoxins, hepoxilins, and hydroperoxy fatty acids (Kuhn and O’Donnell 2006). Some of these mediators are produced by mammalian leukocytes and play a role in inflammation and other immune activities (Kuhn and O’Donnell 2006). Another very interesting EST (EW777901) observed was a cDNA with high similarity to mammalian 15-hydroxyprostaglandin dehydrogenase. This enzyme is key in the inactivation of prostaglandins by oxidizing the 15-hydroxyl group to a corresponding 15-keto-metabolite (Ensor and Tai 1995). The EST as sequenced contained the complete open reading frame of the enzyme (accession no. EU622636).

Others

A number of additional genes of interest were found that did not fit easily into the other categories. A set of heat shock proteins (Hsp), including a C. gigas homologue (EW777988) of HspA12B, as well as the C. gigas Hsp70 (EW778010) and Hsp90 (EW777936), were all found in the library. The Hsps70 and 90 are interesting because of the shear number of cellular processes in which they are involved. Most Hsp70s are constitutively expressed as they are a primary component of folding newly assembled proteins, and not surprisingly, we found a large number of Hsp70 copies in the library (Table 4). Additionally, Hsp70 is involved in refolding denatured or damaged proteins, transporting these proteins to organelles for degradation, and in protection of cellular components in response to varying types of stresses. Danio rerio Hsp12B is distantly related to other Hsp70s. It is also constitutively expressed and possesses a putative adenosine triphosphate (ATP)-binding domain like other Hsp70s. However, unlike most hsp70s, HspA12B is not ubiquitously expressed throughout all cell types and is only expressed in endothelial cells (Durr et al. 2004; Hu et al. 2006; Steagall et al. 2006). It also plays a critical role in regulating angiogenesis in zebrafish during development (Hu et al. 2006). Virtually, no research has examined what effect various cellular stresses may have on the expression of HSPA12B. A thorough examination of the human hsp70 family and its evolution was unable to find homologs of human Hsp12AB in any invertebrate (Brocchieri et al. 2008). As such, the data presented here provide the first evidence for the existence of an invertebrate HspA12B homolog.

Another intriguing EST (EW778149) that was found in the library was cavortin. Cavortin is interesting because it is the major protein in oyster hemolymph, yet its function(s) remains unknown. Early research describes it as a natural hemmaglutinin (Acton et al. 1969). Further research described it as a carbonic anhydrase and also a relative of copper/zinc superoxide dismutases (Cu/Zn SOD; Gonzalez et al. 2005). However, recent research by Scotti et al. (2001) shows that cavortin cannot bind a sufficient number of copper/zinc atoms to serve as a Cu/Zn SOD. Although the actual function of cavortin in oysters is still unknown, the expression of cavortin has been shown to increase during infection and in oysters which exhibit resistance to summer mortality (Huvet et al. 2004).

Conclusions

A number of genes that might be involved in the function of the oyster hemocyte based on their annotation and, in some cases, their expression following bacterial challenge were observed in the current EST analysis. While some of the genes have been reported in other hemocyte EST studies, there were many that have not. Some of the new genes (e.g., IL17) may have been expressed in the oyster hemocytes as a result of their being incubated overnight on poly-lysine plates. The physical process of plating and the adhering of the cells could act as a stimulus for transcription. We have been analyzing the genes expressed in fish macrophages using similar plating and EST approaches (Goetz et al. 2004a). There are some interesting similarities between the genes observed in macrophages and hemocytes that would support a role for some genes in innate immunity extending across vertebrates and invertebrates. For example, we found genes that are presumably involved in cytoskeleton rearrangement including Arps and cofilin in both hemocytes and macrophage cDNA libraries. We also found many of the same proteases in macrophages and hemocytes including matrix metalloproteinase, cathepsins L, C, and Z (Y), and protease inhibitors such as cystatin and the TIMP. While some of these proteases my be involved in tissue reorganization at sites of inflammation, the relationship of some of the cathepsins with the kinin–kininogen pathway (Desmazes et al. 2003) is intriguing and may be occurring in both vertebrates and invertebrates given the apparent ubiquity of the kinin–kininogen system (Torfs et al. 1999; Zhou et al. 2006).

It is well known that vertebrate macrophages can produce a number of different eicosanoids (Sorrell et al. 1989), and ESTs for trout macrophages contained several important genes involved in eicosanoid synthesis (Goetz et al. 2004a). In the oyster hemocytes, we also found cDNAs for several enzymes involved in eicosanoid synthesis and prostaglandin recognition. Interestingly, we found the key enzyme, 15-hydroxyprostaglandin dehydrogenase, that is involved in the initial inactivation of prostaglandins like PGE. This same conversion is accomplished by the dual activity enzyme, LTB4 12-hydroxydehydrogenase/prostaglandin15-keto-reductase, found in trout macrophages (Goetz et al. 2004a). Eicosanoid synthesis and the possible effects of these mediators on immune function in hemocytes have been explored in insect hemocytes (Gadelhak et al. 1995; Stanley 2006). However, there is relatively little information concerning the role of eicosanoids in bivalve immunity and hemocyte function. Experiments using prostaglandin endoperoxide synthase and lipoxygenase inhibitors suggest that eicosanoids are involved in the response to bacteria in Mytilus hemocytes (Canesi et al. 2002). In addition, arachidonic acid supplementation in the Pacific oyster resulted in significant effects on hemocyte numbers and cellular activity, again suggesting that an eicosanoid may be involved in hemocyte function (Delaporte et al. 2006). However, to our knowledge, the synthesis and direct effects of eicosanoids have not been investigated in bivalve hemocytes. Some of the ESTs and their expression observed here would strongly suggest the involvement of a PGE eicosanoid.

Of course, there are also major differences observed between hemocytes and macrophages, particularly in the occurrence in macrophages of many cytokines/chemokines and their receptors and also the presence of specific cell membrane antigens such as the major histocompatibility complex class proteins and clusters of differentiation. It is possible that many cytokines and chemokines evolved within the vertebrate lineage and, therefore, are not present in invertebrates. Further, some of the antigens present on the trout macrophages are related to the adaptive immune system that is not present in invertebrates.

Until recently, virtually no research had been done on immune cell signaling in invertebrates aside from models such as Drosophila. Recent work examined medfly hemocytes and the signaling process involved during phagocytosis (Lamprou et al. 2007). By looking at FAK activation and FAK-interacting molecules, the basic signaling pathways that are activated during phagocytosis in medfly hemocytes were found to be nearly identical to those in vertebrates (Lamprou et al. 2007). Thus, some pathways may have remained relatively unchanged from insects to mammals. The cell-signaling genes uncovered in the present EST analysis suggest that similar signal transduction pathways are present in C. gigas hemocytes.

Finally, there are currently 26,820 ESTs for C. gigas and 14,560 ESTs for C. virginica. While that is not necessarily a large number relative to some vertebrate species, there are already a number of genes present in those ESTs that could be very useful in understanding the immune system in bivalves. This became clear when we aligned the C. gigas ESTs in the present study with the assembled ESTs from C. virginica. A number of the homologs were already present, and some of these may hopefully be used to design primers for comparative studies between the species.

Acknowledgments

This research was supported in part by the Cooperative State Research Education and Extension Service, US Department of Agriculture, under Agreement no. 2003-38500-13505 (SR).

Supplementary material

10126_2008_9117_MOESM1_ESM.rtf (73 kb)
Online AppendixCrassostrea virginica-assembled homologs of selected C. gigas ESTs. Note that the translated products of some of the C. virginica contigs may have stop codons that do not necessarily reflect the true end of the ORF but may be a result of sequencing errors (RTF 73.1 KB)

Copyright information

© Springer Science+Business Media, LLC 2008