Complete genome sequence of a Megalocytivirus (family Iridoviridae) associated with turbot mortality in China

Background Turbot reddish body iridovirus (TRBIV) causes serious systemic diseases with high mortality in the cultured turbot, Scophthalmus maximus. We here sequenced and analyzed the complete genome of TRBIV, which was identified in Shandong province, China. Results The genome of TRBIV is a linear double-stranded DNA of 110,104 base pairs, comprising 55% G + C. Total 115 open reading frames were identified, encoding polypeptides ranging from 40 to 1168 amino acids. Amino acid sequences analysis revealed that 39 of the 115 potential gene products of TRBIV show significant homology to other iridovirus proteins. Phylogenetic analysis of conserved genes indicated that TRBIV is closely related to infectious spleen and kidney necrosis virus (ISKNV), rock bream iridovirus (RBIV), orange-spotted grouper iridovirus (OSGIV), and large yellow croaker iridovirus (LYCIV). The results indicated that TRBIV belongs to the genus Megalocytivirus (family Iridoviridae). Conclusions The determination of the genome of TRBIV will provide useful information for comparative study of Megalocytivirus and developing strategies to control outbreaks of TRBIV-induced disease.

Turbot, Scophthalmus maximus, is an important aquaculture species in coastal areas of northern China. The annual production value of farmed turbot in China achieved US$400 million in 2006. Recently, more and more epizootic diseases of farmed turbot in China occurred because of high density stocking and improper management. In 2004, a fish disease causing high mortality and severe damage to turbot cultures was reported in China [18]. The histopathology of the viral infection was characterized by cell hypertrophy in the spleen, kidney, cranial connective tissue, and endocardium. The causative agent was confirmed to be an iridovirus-like virus based on microscopic examination and transmission electron microscopy (TEM). The virus was then classified as an iridovirus and named as turbot reddish body iridovirus (TRBIV) [18]. As the disease was important to turbot culture, we have analyzed and molecularly characterized the complete genome of TRBIV. We also performed phylogenetic analysis of selected TRBIV proteins compared with those of other iridoviruses and discussed the taxonomic position of TRBIV. The determination of whole genome of TRBIV will provide useful information for comparative study of Megalocytivirus and developing strategies to control outbreaks of TRBIV-induced disease.

Results and discussion
Determination of the viral genomic DNA sequence PCR was performed using primers -MCP-irido5: 5' GGAAGCTTCAAGTGAGGAGCG TGA 3' and MCP-irido6: 5' GGGAATTCACAGGATAGGGAAGCC 3', which were designed based on the MCP of SBIV by Sudthongkong et al. [19] has and have been used to amplify the C terminal of coding region of SBIV, RSIV, GSDIV, ALIV and DGIV to detect diseased turbot. Sequencing of the PCR products revealed that the region of TRBIV showed a high degree of sequence identity to SBIV (98%), RSIV (98%), GSDIV (98%), ALIV (98%), DGIV (98%) and RBIV (97%), the identities of the amino acid sequence were also higher than those of the other iridoviruses, such as CIV, MIV, TFV, FV-3, LCDV, LCDV-1, ATV, GIV, and SGIV. Subsequently, with the completion of the sequencing of the MCP and ATPase of TRBIV, the sequence analysis results indicated that TRBIV was much more closely related to RSIV, ISKNV, RBIV, OSGIV and LYCIV than to other iridoviruses. We then designed a primer pairs based on the conserved genes of ISKNV, OSGIV, and RBIV, to amplify the TRBIV genome. The amplified PCR products were approximately 2,000 ~ 6,000 bp in length. The initial PCR products were designed to have at least 200 bp of overlapping sequence. The PCR products were cloned into vector pMD18-T and sequenced in both directions with the universal forward and reverse M13 primers using an ABI PRISM 3700 automated DNA sequencer. Finally, about 6× coverage of the TRBIV genome sequence was accomplished.
The TRBIV genome contained numerous short direct, inverted, and palindromic repetitive sequences. The highly (direct) repetitive region was identified at position 23,512 to 23,762 bp in the TRBIV genome, which was also found in ISKNV and OSGIV [14,15]. The highly (direct) repetitive region showed highly homologies to the repetitive regions of ISKNV (78%) and OSGIV (78%).

Comparison to other iridoviruses isolated from fish
TRBIV is similar to three other Megalocytiviruses (ISKNV, OSGIV and RBIV) in overall genome structure and exhibited colinearity among them. An analysis of the amino acid sequences deduced from the individual ORFs revealed that 23 of ORFs were the same as the ORFs of ISKNV, OSGIV, RBIV in size and orientation, with homologies ranging from 85% to 99% at the amino acid level. 94, 105 and 111 ORFs of TRBIV had homologies to those of RBIV (58-100%), ISKNV (66-100%) and OSGIV (66-100%) at the amino acid level [13][14][15]. Comparing TRBIV with ISKNV, 42 ORFs were the same size and the identities ranged from 84% to 99%. Comparing TRBIV and OSGIV, 37 of the 115 ORFs were the same size and their identities varied from 87% to 99%; 35 of the 115 ORFs differed in size with identities raging from 66% to 99%. Comparing TRBIV and RBIV, 30 ORFs were the same size and the identities ranged from 77% to100%.
The complete sequence of TRBIV was compared to all iridovirus sequences available in the data banks. The result showed that the putative gene products of TRBIV shared high homology to the corresponding viral proteins of other iridoviruses (Additional file 1). There are 26 core genes which can be found in all iridovirus genomes, such as Serine/threonine protein kinases, DNA polymerase (DdDP), DNA methyl transferase (DMet), SNF2-like helicase, DNA repair protein (RAD2), large subunits of DNA-dependent RNA polymerase (DdRPI and DdRPII), RNase III (RIII), ATPase and MCP. These genes are involved in virus replication, transcription, modification, and structural composition.

Sequence similarities to proteins in databases
The deduced gene products of the 115 ORFs were compared to amino acid sequences by the NCBI BLAST program http://www.ncbi.nlm.nih.gov/BLAST/. Thirty-nine ORFs showed significant homology to functionally characterized proteins from other species. These proteins included structural proteins and enzymes involved in virus replication, transcription, protein modification, and virus-host interaction (Additional file 1).
TRBIV ORF45L encoded a homolog of cytosine DNA methyltransferase. High levels of methylation of cytosine at CpG residues were identified in iridoviruses from vertebrate hosts, such as FV3, LCDV-1, EHNV, ISKNV, TFV and OSGIV [7,15,[20][21][22]. DNA methylation is carried out by a group of enzymes called DNA methyltransferases. These enzymes not only determine the DNA methylation patterns during early development, but are also responsible for copying these patterns to the strands generated from DNA replication. In vertebrate iridoviruses, methyltransferases might play a crucial role in the expression of the viral genome [7].
TRBIV ORF20R encoded a protein showing significant similarity to DNA polymerase. DNA polymerase catalyzes the polymerization of deoxyribonucleotides into a DNA strand. DNA polymerases are best known for their role in DNA replication. Like other DNA polymerases, ORF20R consisted of 3'-5'exonuclease and polymerization motifs located at the N-terminal and C-terminal portions of the protein. These highly conserved motifs are important characteristics of the enzyme [23,24].

Transcription and nucleotide metabolism
TRBIV encodes at least six enzymes that are involved in transcription and nucleotide metabolism. They are ribonucleotide reductase small subunit (ORF25R), two large subunits of the DNA-dependent RNA polymerase (ORF28L, ORF33R), transcription elongation factor IIS (TFIIS, ORF29L), mRNA capping enzyme (ORF59L), and ribonuclease III (ORF80L). These genes of TRBIV are similar to those found in ISKNV and they showed 95 9 6% amino acid sequence identities. TFIIS is ubiquitous in many organisms and plays an important role in transcript elongation [25,26]. Stage and tissue-specific gene expression is an initial and important step in establishing the fate of each cell during development. Expression of particular genes or gene families could be regulated by tissue-specific S-II at the level of transcriptional elongation [27].

Host-related functions
TRBIV encodes at least 15 proteins involved in hostrelated functions. ORF12R, 60L, 61L, 90L and 110R encode proteins of various sizes, which contain a RINGfinger-containing ubiquitin ligase domain. They are highly homologous to those of ISKNV (ORF12R, 65L, 66L, 99L, and 119R). The study on Ring finger proteins (RFPs) showed that the four RFPs of ISKNV (ORF12R, 65L, 66L, and 111L) acted as the E3 enzyme in the presence of ubiquitin activating enzyme (E1), ubiquitin, zinc ions, and a specific E2 protein (UbcH5 subfamily). The RING domain of RFP in ISKNV (ORF66L) was proved to be essential for the activity of E3 enzyme by mutational analysis [28].
ORF36L encoded a protein of 449 amino acids that showed homology to ORF37L of LYCIV, containing a biologically active RGD tri-peptide. The RGD-containing proteins have been identified to be involved in virus infection and virus-host interaction by their RGD motif interaction with the host cell surface receptors. The study on 37L of LYCIV suggested that it had an important role in virus infection [29][30][31][32][33].
ORF47R, which showed homology to the ORF48R of ISKNV, showed higher similarity to the vascular endothelial growth factor (VEGF) encoded by Megalocytivirus and Parapoxvirus than to those encoded by fish and mammals. VEGF functions as a specific mitogen for vascular endothelial cells and as a potent inducer of vascular permeability. The study on ORF48R of ISKNV demonstrated that ISKNV ORF48R functions as a potent growth factor to stimulate angiogenesis and can bind its receptor, FLK-1, to affect zebra fish early embryonic vascular development [34].

Putative membrane-associated proteins
The ORFs of TRBIV were analyzed for the presence of putative transmembrane domains (TMs). One or more putative TMs (1 ~ 11) were found in five ORFs (ORF1L, 7L, 11L, 48L, 84R, 112L), using the computer software TMHMM 2.0 [35]. ORF1L exhibited high homology to the amino acid sequences of ORF1L of ISKNV and OSGIV [14,15]. ORF1L of ISKNV, containing 10 ~ 11 putative transmembrane domains, is a membrane protein and is considered to serve as a model for analyzing the topology and roles of different hydrophobic regions in multi-transmembrane proteins [36].
DNA dot matrix analyses of TRBIV genomic DNA with itself, ISKNV, OSGIV and the RBIV genome were performed using DS GENE 1.5 (Accelrys Inc.). The results revealed that the gene order among TRBIV, ISKNV, OSGIV and RBIV was markedly conserved (Fig. 2).
There are many different genes compared with ISKNV, OSGIV and RBIV in the non-coding region and coding region of TRBIV, including gene insertions and deletions (results not shown). Interestingly, there is an obvious gene deletion of about 1070bp compared with ISKNV at position 67468 ~ 67518 of TRBIV, but hardly any difference with OSGIV and RBIV. The differences of the genes among the three viruses reflect their host specificities.

Relationship of TRBIV to other iridoviruses
Iridovirus sequences available in data banks were compared to the complete sequence of TRBIV. The result showed that the putative gene products of TRBIV shared high homology to the corresponding viral proteins of other iridoviruses (Additional file 1). There were some homologous genes in the ISKNV, OSGIV, RBIV, TFV, FV-3, ATV, SGIV, LCDV-1 and LCDV-C genomes. These included genes for DNA polymerase (DdDP), DNA methyl transferase (DMet), two large subunits of DNAdependent RNA polymerase (DdRPI and DdRPII), RNase III (RIII), ATPase and MCP, involved in virus replication, transcription, modification, and structural composition.

Phylogenetic analysis
To determine the phylogenetic relationship of TRBIV to other iridoviruses, the amino acid sequences of MCP, ATPase, cytosine DNA methyl transferase and DNA polymerase were used in alignments with other iridoviruses and non-iridoviruses from GenBank. MCP is highly conserved in iridoviruses [37,38]. In the MCP tree, the iridoviruses used in the multiple alignments were subdivided into four groups: group I, Lymphocystiviruses, including LCDV-1 and LCDV-C; group II, the insect iri-doviruses, including CIV, MIV; group III, Megalocytiviruses, including TRBIV, ISKNV, OSGIV, RBIV; and group IV, Ranaviruses, including FV3, TFV, ATV, GIV and SGIV (Fig. 3A). The results showed that TRBIV was more closely related to ISKNV, OSGIV and RBIV, than to the Ranavirus and Lymphocystiviruses. Phylogenetic analyses using the highly conserved full-length protein sequences of the ATPase, cytosine DNA methyl transferase and DNA polymerase from the iridoviruses (Figs. 3B, C, D) C.

B.
A.

D.
also supported the view that TRBIV was more closely related to ISKNV, OSGIV and RBIV.

Conclusions
In summary, the present study provided the complete genome sequence of TRBIV. The genomic organization, gene content, and amino acid composition of TRBIV are very similar to those of ISKNV, OSGIV and RBIV. Dot plot analyses and phylogenetic trees suggested that TRBIV belongs to the genus Megalocytivirus in the family Iridoviridae. The detailed viral genome analysis will provide valuable information to serve as the genetic basis for future studies and to control iridovirus disease in cultured fish.

Infected fish
TRBIV used in this study was originally isolated from cultured turbot (Scophthalmus maximus) with TRBIV disease from Shandong Province, China, in 2005. These fish had enlarged spleen and kidney cells, as verified by histopathology. The spleens and kidneys were removed from the diseased fish and the viruses were observed by light and electron microscopy. PCR was performed for further confirmation of the disease. These samples were collected and kept at -80°C for future use.

Virus and viral DNA
Total DNA was extracted from spleen and kidney of diseased turbot using high pure PCR template preparation kit (Roche, USA). The quality and concentration were determined by agarose gel electrophoresis and by a spectrophotometer.

PCR amplification and sequencing
The complete genomic DNA of TRBIV was sequenced by PCR method. As TRBIV showed high homology to ISKNV, RBIV, and OSGIV, the primers were derived from the DNA sequence of ISKNV, RBIV, and OSGIV in Gen-Bank/EMBL/DDBJ database (AF371960, AY532606, AY894343). Amplified PCR products were about 2000 6 000 bp in length, and the initial PCR products were designed to have at least 200 bp of overlapping sequence. The PCR products were purified using TaKaRa Agarose Gel DNA Purification Kit Ver.2.0 (TaKaRa). The purified PCR products were cloned into vector pMD18-T (TaKaRa), and sequencing was performed using an ABI PRISM 3700 automated DNA sequencer in both directions with the universal forward and reverse M13 primers.
ORFs were identified by the following criteria: (1) they were not less than 120 bp and (2) they were not located within larger ORFs. In addition, TRBIV ORFs that were likely to be expressed were also identified on the basis of significant identity to known protein sequences within the databases.