TRBP Homolog Interacts with Eukaryotic Initiation Factor 6 (eIF6) in Fenneropenaeus chinensis¹

The HIV transactivating response RNA-binding protein (TRBP)³ was originally characterized and isolated from a HeLa cell expression library. TRBP was the first cellular protein identified that strongly binds the HIV-1 encoded leader RNA, a trans-activation response (TAR) protein, and enhances virus expression (1, 2, 3, 4, 5).

Three TRBP isoforms, TRBP1–3, have been reported in humans. TRBP1 was the first of these proteins to be identified. TRBP2 contains a splice variant that adds 21 additional amino acids to its N terminus (6). Compared with TRBP1, TRBP3 is missing the C-terminal dsRNA-binding domain (7). TRBPs belong to a family of dsRNA-binding proteins, and both TRBP1 and TRBP2 contain three dsRNA-binding domains (7). The third C-terminal domain mediates binding to Dicer (7, 8) and protein kinase R (PKR) (9), instead of binding to RNA.

TRBP has recently been identified as an important Dicer-interacting protein in mammals (7, 10), and Dicer has been shown to interact with the Argonaute 2 (Ago2) protein, a key component of the RNA-induced silencing complex (RISC). Experiments have shown that knockdown of TRBP results in a general loss of RNA interference (RNAi)-mediated silencing of other target mRNAs, which is consistent with the hypothesis that TRBP interacts with Dicer (8). Work by Chendrimada et al. further supported a role for TRBP in forming a siRNA-TRBP-Dicer-Ago2 complex, in which TRBP functions as a bridge between dsRNA and Dicer for Ago2 recruitment (10). MacRae et al. have shown that the RISC-loading complex can spontaneously assemble in vitro from purified Ago2, Dicer, and TRBP, and subsequently display dicing, slicing, guide-strand selection, and Ago2-loading activities (11).

TRBP has been shown to modulate HIV-1 gene expression through its association with TAR (4, 5), regulate translation through interactions with structured RNA motifs (4), regulate translation during spermatogenesis in testis (12), regulate cell growth (13, 14), and bind to and inhibit IFN-induced PKR in vertebrates (15, 16, 17).

Two recently published studies reported TRBP to be an integral component of a Dicer-containing complex. Knockdown of TRBP was shown to inhibit microRNA (miRNA) biogenesis (10). Eukaryotic initiation factor 6 (eIF6) has been shown to have an important role in the RNAi pathway. Depletion of eIF6 in human cells specifically impairs miRNA-mediated regulation of target protein and mRNA levels (18). Similarly, depletion of eIF6 in Caenorhabditis elegans diminishes lin-4 miRNA-mediated repression of endogenous LIN-14 and LIN-28 protein and mRNA levels. These activities associated with eIF6 are consistent with its proposed abilities to bind the free 60S ribosome subunit and prevent productive assembly of the 80S ribosome (19). eIF6 has also been shown to be a component of the human RISC, a multiprotein complex that also includes MOV10, the homolog of Drosophila translational repressor Armitage, and 60S ribosome subunit proteins (18).

Until now, TRBP and eIF6 have not been identified in shrimp. In this study, we describe the cloning of three full-length cDNAs that encode a TRBP2 homolog and an eIF6 protein from Chinese shrimp, Fenneropenaeus chinensis. The genes from which these cDNAs were derived are designated Fc-TRBP1–3 and Fc-eIF6, respectively. Interactions between Fc-TRBP and Fc-eIF6 were identified in T7 phage library panning experiments and confirmed by pull-down assays and Far Western blots. We also analyzed the expression patterns of TRBPs and eIF6 in response to infection by the shrimp pathogen Vibrio anguillarum and the white spot syndrome virus (WSSV).

F. chinensis (∼10–20 g each) were bought from a shrimp market in Qingdao, Shangdong Province, China. The shrimp were kept in tanks containing aerated seawater.

The WSSV inoculum was prepared according to a previously described method (20). WSSV was extracted from the gills of naturally infected F. chinensis and stored at −80°C. Gill tissue (1 g) was homogenized in PBS (10 ml; 140 mM NaCl, 2.7 mM KCl, 10 mM Na₂HPO₄, and 1.8 mM KH₂PO₄). After centrifugation at 5000 × g for 10 min at 4°C, the supernatant was passed through a 450-nm membrane filter and then used as inoculum. Viral quantification was performed according to a previously published method (21). Briefly, a 619-bp WSSV DNA fragment was amplified from the viral DNA extracted from the gills of WSSV-infected shrimp by PCR using primers vF1 and vR1 (Table I). The fragment (denoted as v1) was cloned into the pBluescript II SK⁺ plasmid and a series of dilutions of this recombinant plasmid (pBluescript II-v1) were made and quantified by use of a spectrophotometer (GeneQuant; Amersham Biosciences). The primers used for real-time PCR assays were vF2 (Table I) and vR1. The real-time PCR program was 94°C for 5 min, followed by 40 cycles of 94°C for 15 s and 62°C for 50 s. Melting curve analysis of the amplification products was performed at the end of each PCR to confirm that only one fragment was amplified. Quantitative real-time PCR data were analyzed by Opticon Monitor 2, and the baseline was automatically set by the software. The WSSV standard curve was derived from the cycle threshold (C_T) and the quantity of the template used. The number of virus copies in an infected shrimp was calculated by use of the standard curve.

For the immune challenge experiments, we injected V. anguillarum (∼3 × 10⁷ cells/shrimp) or WSSV (3.2 × 10⁷ copies/shrimp) into the abdominal segment of F. chinensis. Total RNA from the challenged shrimp was isolated from a variety of tissues (hemocytes, heart, hepatopancreas, gills, stomach, intestine, ovary, and spermary) using Unizol (Biostar). The RNA was isolated at 2, 6, 12, and 24 h postinfection; RNA from the unchallenged shrimp was also isolated as a control. To reverse transcribe the isolated RNA, a SMART cDNA kit (BD Biosciences) was used. The first cDNA strand was reverse transcribed from 5 μg of total RNA by use of the oligo anchor R and SMART F primers (Table I) in combination with a first-strand cDNA synthesis kit (Sangon Biological Engineering Technology and Services).

The methods for constructing and panning the T7 phage display library have been previously described (22). Briefly, the T7 phage display library was constructed according to T7Select System procedures (Novagen). The mRNA was isolated from hemocytes of V. anguillarum-challenged shrimp using the QuickPrep Micro mRNA purification kit (Amersham Biosciences). After reverse transcription, the double-stranded cDNA was prepared and the EcoRI/HindIII linkers were added to the two ends of the double-stranded cDNA. The cDNA was cleaved and ligated to T7 selected vector arms. The vector arms were packed with T7 packaging extracts and the library was amplified after calculation of the phage titer.

Formaldehyde-fixed Micrococcus luteus were used to pan the T7 phage display library. M. luteus was collected, washed twice with TBS (50 mM Tris-HCl, 0.5 M NaCl (pH 7.5)), and fixed with 3.5% paraformaldehyde. The T7 phage display library was also washed with TBS before the addition of a 100-μl aliquot of the library to 100 μl of bacteria (OD₆₀₀ = 2.0). After gentle mixing, the sample was kept overnight at 4°C. The sample was then centrifuged at 6000 × g for 5 min and the bacteria-bound phages were collected and washed three times with TBST (50 mM Tris-HCl, 0.5 M NaCl, 0.02% Tween 20 (pH 7.5)). The bacteria-bound phages were resuspended in 200 μl of 1% SDS for 10 min before the phages were eluted and collected by centrifugation at 6000 × g for 5 min. Ten microliters of supernatant was then added to 1 ml of BLT5403 host cells and agitated at 37°C for 1 h. After centrifugation, 200 μl of phage-containing supernatant was amplified on an agarose plate containing BLT5403 host cells at 37°C for 3 h. Five milliliters of extraction buffer (100 mM NaCl, 6 mM MgSO₄, 20 mM Tris-HCl (pH 8.0)) was added to the plates, which were further incubated at 4°C overnight. Extracted phages were then collected for a second round of panning. After three rounds, 10 plaques were randomly selected for PCR amplification and sequencing. A 1000-bp fragment was obtained and identified as the 3′ end of an Fc-TRBP cDNA. The 5′ end of Fc-TRBP, including the 5′ untranslated region (UTR), was amplified from hemocyte cDNA with a reverse gene-specific primer, TRBPR4, and the 5′ PCR primer (Table I). The PCR procedure included 1 cycle at 94°C for 2 min and 35 cycles at 94°C for 30 s, 55°C for 45 s, and 72°C for 45 s, followed by 1 cycle at 72°C for 10 min. Two isoforms of Fc-TRBP was cloned with primers TRBPexF and TRBPexR (Table I).

The open reading frame (ORF) of Fc-TRBP1 (1032 bp) was amplified from hemocyte cDNA using the primers TRBPexF and TRBPexR (Table I) with EcoRI and XhoI restriction sites inserted at the beginning and end of the ORF, respectively. The PCR conditions were as follows: 1 cycle at 94°C for 3 min; 35 cycles at 94°C for 30 s, 55°C for 45 s, and 72°C for 45 s; and 1 cycle at 72°C for 10 min. The PCR products were then cloned into the EcoRI and XhoI restriction sites of pET-30a. The recombinant pET30a-Fc-TRBP plasmid was transformed into Escherichia coli BL21 (DE3) cells, which were then cultured in Luria-Bertani medium supplemented with 25 μg/ml ampicillin. Isopropyl β-D-thiogalactoside (IPTG; 0.1 mM) was added when the OD₆₀₀ of the culture reached 0.5. After 3 h of culture, the cells were collected by centrifugation at 6000 rpm for 10 min and resuspended in PBS containing 0.2% Triton X-100. Following cell sonication and centrifugation, Fc-TRBP was purified using His-Bind resin (Novagen) according to the manufacturer’s instructions. The purified protein was analyzed by 12.5% SDS-PAGE and stained by Coomassie brilliant blue G250.

Approximately 40 μg of purified Fc-TRBP was added to an enzyme immunoassay plate and incubated at 4°C overnight for panning of the T7 phage display library. After three rounds of panning, 10 plaques were selected for PCR amplification and sequencing. A 454-bp fragment was obtained and identified as Fc-eIF6. The 3′ end of Fc-eIF6 was amplified from hepatopancreas cDNA with gene-specific primers eIF6F1 and the 3′ anchor R primer (Table I). The 5′ end of the gene was amplified by use of a reverse gene-specific primer, eIF6R2, and a 5′ PCR primer. The PCR procedure was as follows: 1 cycle at 94°C for 2 min and 35 cycles at 94°C for 30 s, 55°C for 45 s, and 72°C for 45 s, followed by 1 cycle at 72°C for 10 min.

The ORF of Fc-eIF6 (738 bp) was amplified from hepatopancreas cDNA using the primers eIF6exF and eIF6exR (Table I) with EcoRI and XhoI restriction sites inserted at the beginning and end of the ORF, respectively. Fc-eIF6 was cloned into the EcoRI and XhoI restriction sites of pET-30a and pGEX-4T-1. The recombinant pET30a-Fc-eIF6 and pGEX-4T-1-Fc-eIF6 plasmids were transformed into E. coli BL21 (DE3) and E. coli BL21, respectively, which were then cultured in Luria-Bertani medium supplemented with 25 μg/ml ampicillin. IPTG (0.1 mM) was added when the OD₆₀₀ of the culture reached 0.5. After 3 h of culture, the cells were collected by centrifugation at 6000 rpm for 10 min. After cell sonication and centrifugation (10,000 rpm, 10 min at 4°C), His-tagged Fc-eIF6 and GST-tagged Fc-eIF6 were purified using His-Bind resin (Novagen) and glutathione Sepharose 4B (Amersham Biosciences), respectively, according to the manufacturers’ instructions. Purified proteins were analyzed by 12.5% SDS-PAGE.

A search for sequences matching those of Fc-TRBP and Fc-eIF6 was performed using BLASTX (www.ncbi.nlm.nih.gov). Characteristics of each protein were calculated using the Expert Protein Analysis System (ExPASy; au.expasy.org), and domain predictions were performed using the Simple Modular Architecture Research Tool (SMART; smart. embl-heidelberg.de). Sequence alignments were assembled using DNAMAN software, and a phylogenetic tree was constructed using Molecular Evolutionary Genetics Analysis (MEGA) 3.1.

The Fc-TRBP sequence was inserted into the pGEM-T-Easy vector in reverse orientation (Promega), and the plasmid was linearized by EcoRI digestion. Digoxigenin (Dig)-tagged antisense probes were transcribed using T7 RNA polymerase according to the manufacturer’s instructions (Roche Applied Sciences). Northern blot analysis was performed according to a previously described method (23). Briefly, ∼10 μg of total RNA was denatured and electrophoresed on an agarose gel containing formaldehyde and transferred onto a nylon membrane. The target mRNA was hybridized to a Dig-labeled antisense RNA probe (100 ng/ml) in 50% formamide at 68°C. Following stringent washing at 68°C, the anti-Dig-phosphatase Ab (AB) was used to detect the probe. Ab binding was visualized by incubation of the membrane with 5-bromo-4-chloro-3-indolyl phosphate and NBT chloride. The intensities of the resulting bands were quantified using Quantity One (Bio-Rad). The ratios of Fc-TRBP to 18S rRNA were calculated.

Five micrograms of DNase-treated total RNA (Takara RNase-free DNase I; Takara Bio) was used as template for reverse transcription of the first cDNA strand. PCR-based amplification of Fc-TRBP was performed using the TRBPF1 and TRBPexR primers (Table I). PCR-based amplification of Fc-eIF6 was performed using the eIF6RTF and eIF6RTR primers (Table I). The PCR procedure included: 1 cycle at 94°C for 2 min and 27 cycles of 94°C for 30 s, 53°C for 45 s, and 72°C for 45 s, followed by a final cycle at 72°C for 10 min. A β-actin fragment was amplified as a reference using actin F and actin R primers (Table I). The PCR procedure for β-actin included: 1 cycle at 94°C for 2 min and 23 cycles of 94°C for 30 s, 53°C for 45 s, and 72°C for 45 s, followed by a final cycle at 72°C for 10 min. RT-PCR assays were performed in triplicate, and the ratios of Fc-TRBP to β-actin and Fc-eIF6 were calculated using Quantity One (Bio-Rad).

Fc-TRBP1 with an N-terminal His tag was expressed in E. coli BL21 (DE3). After induction with 0.1 mM IPTG at 37°C for 2 h, 200 ml of cells was collected and lysed by sonication in 20 ml of PBS containing 0.2% Triton X-100. Ten milliliters of lysed cells was subsequently incubated with His-binding resin (1 ml) for 5 min, then washed with 10 ml of binding buffer (0.5 M NaCl, 20 mM Tris-Cl (pH 7.9), 5 mM imidazole) and then 6 ml of wash buffer (0.5 M NaCl, 20 mM Tris-Cl (pH 7.9), 60 mM imidazole). Pull-down assays were performed by incubation of ∼200 μg of purified Fc-eIF6-GST protein or Fc-eIF6 (after removal of the N-terminal His tag by incubation with thrombin) with His-tagged Fc-TRBP1 and His-Bind resin for 10 min at 4°C. GST protein was used as a control for the pull-down analysis. After three washes with 10 ml of wash buffer, the proteins were eluted with eluting buffer (0.5 M NaCl, 20 mM Tris-Cl (pH 7.9), 1 M imidazole). The eluted proteins were analyzed by 12.5% SDS-PAGE.

The method used for antiserum preparation has been previously described (22). In brief, purified recombinant Fc-TRBP1 (100 μg) was diluted to 1 ml with saline, mixed with CFA (1 ml), and injected into rabbits once per week for 3 wk. Subsequently, the rabbits were given booster injections (100 μg) of the Ag without adjuvant once per week for 3 wk. Blood samples were collected and the antiserum titer was determined by double immunodiffusion using 1% agar in 0.02 M Tris-HCl (pH 8.0) containing 0.15 M NaCl and 0.01% sodium azide.

The method used for Western blot analysis has been previously described (24). In brief, protein was extracted from WSSV-challenged shrimp tissues at 6, 12, and 24 h postinfection. Protein was similarly extracted from unchallenged shrimp as a control. Approximately 200 μg of total protein was submitted to SDS-PAGE and then transferred to a nitrocellulose (NC) membrane. The NC membrane was incubated with Fc-TRBP1-specific antiserum (1/100). Following stringent washing with TBST (100 mM NaCl, 10 mM Tris-HCl, 0.02% Tween), goat anti-rabbit IgG peroxidase conjugates were used to detect the TRBP polyclonal Ab. Ab binding was visualized by incubation of the NC membrane with 4-chloro-1-naphthol (4-CIN) and H₂O₂.

Approximately 2 μg of purified rFc-eIF6 (after removal of the N-terminal His tag by incubation with thrombin) was electrophoresed by SDS-PAGE and then transferred to a NC membrane. The membrane was incubated with 5 μg/ml purified Fc-TRBP1 for 5 h at room temperature. After three stringent washes with TBST, the NC membrane was incubated with Fc-TRBP1 antiserum (1/100) to detect Fc-TRBP1 bound to Fc-eIF6 on the NC membrane. Goat anti-rabbit IgG peroxidase conjugates were used to detect the Fc-TRBP1 polyclonal Ab. Ab binding was visualized by incubation of the NC membrane with 4-CIN and H₂O₂. Purified Fc-TRBP1 and BSA were used as positive and negative controls, respectively.

The shrimp were kept in tanks containing aerated seawater. We injected WSSV (3.2 × 10⁷ copies/shrimp) and 6 μg of Fc-TRBP1 into the abdominal segment of F. chinensis. In another group, WSSV (3.2 × 10⁷ copies/shrimp) and 30 μl of TRBP antiserum were similarly injected. The control group was injected with WSSV (3.2 × 10⁷ copies/shrimp) and 6 μg of BSA. At 24 h postinjection, the genomic DNA was extracted from the gills of the shrimp and semiquantitative RT-PCR and quantitative real-time PCR were performed to quantify the WSSV.

Formaldehyde-fixed M. luteus was used to pan a T7 phage display library, and an 873-bp fragment containing poly(A) tail was obtained. A GenBank search was performed using the amino acid sequence of the fragment. This search identified a sequence that is highly homologous to a TRBP from another species. Consequently, the fragment isolated by panning was termed Fc-TRBP. Using the sequence of Fc-TRBP, the 5′ end of the TRBP, including the 5′-UTR, was cloned from hemocyte cDNA. The complete cDNA of Fc-TRBP was determined to be 1526 bp in length and contain an ORF of 1032 bp that encodes a 343-aa protein. Furthermore, Fc-TRBP mRNA was found to contain a 225-bp 5′-UTR and a 269-bp 3′-UTR. Fc-TRBP protein has a theoretical molecular mass of 36.8 kDa and a predicted isoelectric point of 7.11, and it contains three dsRNA-binding motif (DSRM) domains between residues 27 and 92, 130 and 196, and 272 and 338 (supplemental Fig. 1).⁴ The cDNA sequence of Fc-TRBP (designated Fc-TRBP1) was deposited in GenBank (www.ncbi.nlm.nih.gov/Genbank/index.html) (accession no. EU679001). Two additional isoforms of Fc-TRBP1 were amplified by PCR using the Fc-TRBP expression primers. Fc-TRBP isoform 2 (GenBank no. FJ573167) and isoform 3 (GenBank no. FJ573168) contain ORFs of 972 and 891 bp, which encode 323- and 296-aa proteins, respectively (supplemental Figs. 2 and 3). Both Fc-TRBP2 and Fc-TRBP3 are identical to the Fc-TRBP1, except that they lack 60 and 141 bp between DSRM2 and DSRM3, respectively (Fig. 1 A). The isoforms result from alternative splicing of the same transcript.

A BLAST protein search revealed that Fc-TRBP is highly conserved in relationship to the TRBP2 protein from Aedes aegypti (47% identity) and to the TRBP from Drosophila melanogaster (49% identity). In contrast, homology was relatively low in relationship to vertebrate animals such as Danio rerio (39% identity), Homo sapiens (36% identity), Mus musculus (37% identity), Rattus norvegicus (36% identity), and Xenopus tropicalis (35% identity). However, the DSRM region of these proteins is highly conserved (supplemental Fig. 4). We constructed a phylogenetic tree for the TRBP family of genes (supplemental Fig. 5), and the results show that the TRBP gene family is divided into two groups. TRBP sequences from vertebrates such as D. rerio and H. sapiens form one cluster, while sequences from invertebrates such as F. chinensis and A. aegypti form another cluster.

Purified, recombinant Fc-TRBP1 was used to pan a T7 phage display library. After three rounds of panning, 10 plaques were randomly selected for PCR amplification and sequencing (Table II). A 454-bp fragment was obtained and its deduced amino acid sequence was found to be highly homologous to those of eIF6 from other species. Therefore, it was termed Fc-eIF6. Based on the sequence of the identified fragment, gene-specific primers were designed and used to amplify Fc-eIF6 from hepatopancreas cDNA. The complete cDNA of Fc-eIF6 was 1465 bp and contained a 738-bp ORF that encodes a 245-aa protein. The Fc-eIF6 mRNA was found to contain a 48-bp 5′-UTR and a 679-bp 3′-UTR. The Fc-eIF6 protein includes an eIF6 domain between residues 3 and 204. The protein has a theoretical molecular mass of 26.5 kDa and a predicted isoelectric point of 4.99 (Fig. 1 B and supplemental Fig. 6). The cDNA sequence was deposited in GenBank (www.ncbi.nlm.nih.gov/Genbank/index.html) (accession no. EU679002).

A BLAST protein search using the Fc-eIF6 sequence indicated that it is highly similar to the eIF6 sequences in other species, including A. aegypti (66% identity), Bombyx mori (66% identity), D. rerio (68% identity), H. sapiens (68% identity), M. musculus (68% identity), and X. tropicalis (69% identity). Alignment of these eIF6 sequences is shown in supplemental Fig. 7. A phylogenetic tree was constructed for the eIF6 family and shows three groups, one from vertebrates such as D. rerio and H. sapiens, one from insects such as B. mori and A. aegypti, and a third from F. chinensis (supplemental Fig. 8).

Northern blot and RT-PCR analyses were used to examine the transcription levels of Fc-TRBP and Fc-eIF6 in several tissues following challenge with V. anguillarum or WSSV. At 24 h postinfection, RNAs from challenged and unchallenged shrimp were collected. Both Fc-TRBP and Fc-eIF6 mRNAs were detected in all tested tissues obtained from the unchallenged shrimp (i.e., hemocytes, heart, hepatopancreas, stomach, gills, intestine, and ovary). In tissues from shrimp challenged with V. anguillarum, Fc-TRBP was slightly increased in hemocyte, stomach, gill, and intestinal tissues, although the increases were not statistically significant. Fc-TRBP was also up-regulated in heart and hepatopancreas tissues following V. anguillarum challenge; no obvious change was observed in Fc-TRBP levels in the ovary and spermary tissues (Figs. 2 and 3,A). Fc-eIF6 was also found to be up-regulated in hepatopancreas tissues, although there was no statistically significant change in expression detected in hemocyte, heart, stomach, gill, and intestinal tissues (Fig. 3 C).

RT-PCR assays revealed that expression of Fc-TRBP was increased in several tissues, including heart, gills, and intestinal tissues, in response to challenge by WSSV infection. In contrast, Fc-TRBP expression was decreased in hemocyte, hepatopancreas, and stomach tissues. No change in expression was detected in ovary tissues (Fig. 3,B). Fc-eIF6 expression was also found to be decreased in hemocyte and stomach tissues, and there was a sharp increase in intestinal tissue. The Fc-eIF6 level seemed to be slightly increased in heart, hepatopancreas, gill, and spermary tissue, although the increases were not statistically significant (Fig. 3 D).

Time-course RT-PCR analysis was performed to evaluate the expression of Fc-TRBP in hepatopancreas and hemocyte tissues. TRBP expression in hepatopancreas tissues was found to decrease between 2 and 6 h after WSSV infection; it returned to normal levels at 24 h postinfection. Similarly, Fc-TRBP expression decreased in hemocytes and began to recover at 24 h postinfection (Fig. 4).

Western blot analysis was performed with Fc-TRBP1-specific antiserum to confirm the expression pattern of Fc-TRBP detected by RT-PCR. Three immunoreactive protein bands were detected, a result that is consistent with the three isoforms of TRBP (Fig. 5). Proteins were extracted from hepatopancreas and gill tissues of the shrimp at 6, 12, and 24 h following WSSV challenge. Proteins obtained from uninfected shrimp were used as a negative control. Expression of TRBPs in the hepatopancreas was found to decrease at 6 h postinfection and return to normal levels at 12 h postinfection. In contrast, expression of TRBPs increased in gill tissue at 6 h after WSSV challenge. The expression pattern of TRBPs in response to challenge by V. anguillarum was also analyzed. TRBP expression largely remained the same in the hepatopancreas after infection, while it slightly increased in gill tissues.

Due to the potential for false positives in phage display screening, pull-down assays were performed to confirm the interaction between rFc-TRBP and rFc-eIF6. His-tagged rFc-TRBP and GST-tagged rFc-eIF6 were purified as shown in Fig. 6. Results from the pull-down assays indicate binding between rFc-TRBP and rGST-Fc-eIF6, as well as between rFc-TRBP and rHis-Fc-eIF6 following removal of the His tag by treatment with thrombin in vitro (Fig. 7). Residual thrombin was present following cleavage of the His tag from Fc-eIF6. The residual thrombin partially removed the His tag from Fc-TRBP, resulting in visualization of two Fc-TRBP bands. As a control, we determined that GST alone was not pulled down in these assays.

The Far Western blot assay was also performed to confirm the interaction between rFc-TRBP and rFc-eIF6. rFc-eIF6 without a His tag was transferred to NC membrane and incubated with 5 μg/ml purified TRBP. TRBP bound to Fc-eIF6 was detected by immunoblotting (Fig. 8). The results from these assays support the hypothesis that Fc-TRBP directly binds Fc-eIF6.

To study the function of rFc-TRBP in vivo, WSSV and rFc-TRBP1 were injected into the shrimp. At 24 h postinjection, the replication of WSSV was evaluated by quantitative RT-PCR. WSSV replication was inhibited (∼2.3 × 10⁶ copies/g tissues) by coinjection with rFc-TRBP compared with shrimp injected with WSSV and BSA (∼1.4 × 10⁸ copies/g tissues). Injection with rFc-TRBP antiserum slightly increased the replication of WSSV (2.2 × 10⁸ copies/g tissues), although there was no statistically significant change relative to the control. Results from this experiment suggest that Fc-TRBP plays an important role in the inhibition of WSSV replication.

Increasing evidence indicates that RNAi may have an important role in antiviral immunity in invertebrates. The first direct evidence was reported in Drosophila, in which the flock house virus is both an initiator and a target of RNA silencing in Drosophila host cells. Flock house virus infection has been shown to require suppression of RNA silencing by a flock house virus-encoded protein, B2 (25).

dsRNA can induce sequence-specific antiviral silencing and nonspecific immunity in a marine shrimp, Litopenaeus vannamei (26, 27). Inhibition of viral disease by injection of shrimp with dsRNA specific to viral genes has been reported for at least three viruses: WSSV, Taura syndrome virus, and yellow-head virus (26, 28, 29). dsRNA is produced by both positive-strand RNA viruses (i.e., Taura syndrome virus and yellow-head virus) and DNA viruses (i.e., WSSV) during their infectious cycles (30). These dsRNAs are thought to be involved in the RNAi pathway, which can result in generation of an antiviral response.

The exact mechanism of RNAi in shrimp is not clear. In two previously published papers (7, 10), TRBP was shown to interact with Dicer in mammalian models and to be required for assembly and/or function of RNAi as mediated by siRNA and miNAs. Furthermore, Chendrimada et al. showed that TRBP, as part of RISC, can form a complex with Dicer and Ago2, thereby functioning as a potential bridge between dsRNA and Dicer for Ago2 recruitment (10).

In this study, three new members of the TRBP family in F. chinensis, Fc-TRBP1–3, have been characterized. Phylogenetic analysis and sequence alignments from multiple species demonstrate that Fc-TRBP is highly conserved in relationship to other members of the same family, especially in terms of the DSRM sequence. Sequence alignments indicate that Fc-TRBP belongs to the longest isoform of the TRBP family, TRBP2, which contains three DSRM domains (31, 32), two N-terminal domains that mediate dsRNA binding, and a C-terminal domain that is required for interactions with Dicer (7, 8).

The association of RISC with a multiprotein complex that includes eIF6, MOV10, and 60S ribosome subunit proteins suggests the importance of eIF6 in the RNAi pathway (18). T7 phage display panning using rFc-TRBP1 led to isolation of Fc-eIF6 protein. Pull-down assays and Far Western blot analysis were performed to confirm the interaction between the recombinant proteins rFc-TRBP and rFc-eIF6. These results support the hypothesis that the interaction between Fc-TRBP and Fc-eIF6 is mediated by the multiprotein complex RISC (Fig. 9).

The expression patterns of Fc-TRBP and Fc-eIF6 in response to challenges by V. anguillarum or WSSV were analyzed by RT-PCR. Expression of both genes was detected in all analyzed tissues and was slightly up-regulated in selected shrimp tissues at 24 h postchallenge with V. anguillarum. When the shrimp were challenged with WSSV, more complex changes were detected. Expression of Fc-eIF6 was increased in the intestine and decreased in stomach. Fc-TRBP was also up-regulated in the gills and intestine; however, no change in gene expression was detected in heart and ovary tissues. Fc-TRBP was down-regulated in hemocyte, hepatopancreas, and stomach tissues. Western blot analysis confirmed the results of the RT-PCR assays at the protein level. Three bands were identified as Fc-TRBPs by Western blot analysis. This result is consistent with the molecular cloning result: three isoforms of Fc-TRBP were cloned from the shrimp. Indeed, multiple isoforms of TRBP have been detected in mammalian species, which may result from alternative splicing of the same mRNA (7). All of the Fc-TRBP isoforms contained three DSRM domains, while human TRBP3 lacks the third dsRNA-binding domain.

Human TRBP was reported to display dual functions, with roles in HIV replication and RNA interference. TRBP enhances HIV expression and replication by inhibiting PKR and increasing translation of structured RNAs, despite its crucial function in an RNAi pathway that could inhibit HIV infection (15, 33). It is reported that the TAR RNA of HIV sequesters TRBP from RISC, resulting in inhibition of RISC activity. This inhibition is relieved by exogenous expression of TRBP (5). We could not find a sequence in the WSSV genome similar to the HIV TAR RNA; however, our results indicate that injection of rFc-TRBP1 reduces the replication of WSSV. Our results suggest that exogenous TRBP may recruit components of RISC in vivo and thereby enhance RNAi.

Weber et al. investigated the presence and localization of dsRNA in cells infected with a range of viruses. Their results indicate that significant amounts of dsRNA are present following productive infection with positive-strand ssRNA, dsRNA, or DNA viruses (30). The origin of the dsRNA structures detected following infection with DNA viruses remains to be determined. dsRNA may arise as a result of overlapping converging transcription (34, 35) or highly structured ssRNAs, such as the adenovirus virus-associated RNAs, of which up to >10⁸ copies are present in a single infected cell (36). Given the data we have presented regarding the interaction between rFc-TRBP and rFc-eIF6, as well as the association of eIF6 with RNAi signaling pathways and involvement with RISC, we propose that interactions between Fc-TRBP and Fc-eIF6 are mediated by RISC and have a role in the anti-viral immune response in shrimp (Fig. 10).

We thank Dr. Xiao-Qiang Yu, School of Biological Sciences, University of Missouri at Kansas City, for useful discussion and critical reading of the manuscript.

The authors have no financial conflicts of interest.