Introduction

The aquatic birnaviruses are the largest and most diverse group of viruses within the family Birnaviridae. They include a variety of viruses from numerous species of fish and marine invertebrates worldwide [1, 2]. Infectious pancreatic necrosis (IPN) was first described in a freshwater brook trout (Salvelinus fontinalis) facility in North America in the 1950s [3] with the subsequent isolation of the virus reported by Wolf et al. [4]. In Europe, the disease was initially reported in freshwater rainbow trout (Oncorhynchus mykiss) farms in France [5], Denmark [6], Scotland [7], and Norway [8]. By the mid-1970s, IPN had been reported in North America, Europe, and Japan [9]. In 1986, the first outbreak of IPN was recorded in hatcheries for juvenile rainbow trout in Shanxi Province, northwest China [10]. This was followed by a series of IPN outbreaks in cultured juvenile rainbow trout in various districts of China [11–13].

IPN, a significant disease of farmed fish worldwide, is caused by infectious pancreatic necrosis virus (IPNV), a member of the family Birnaviridae, a family of non-enveloped, double-stranded RNA viruses with a bi-segmented (A and B) genome encoding five or six proteins [14]. The genome segment A (3.1 kb) contains two partially overlapping open reading frames (ORFs), and the large one encodes a 106 kDa polyprotein (pVP2–VP4–VP3), which will be co-translationally cleaved by the viral non-structural protease (VP4; 29 kDa) to generate pre-VP2 (pVP2, 63 kDa) and VP3 (29–31 kDa) proteins. There is a small ORF which overlaps with the amino-terminal end of the large ORF and generates a 15 kDa (VP5) non-structural polypeptide [15–17]. Both VP2 (449 aa) and VP3 (217 aa) are major components of the virion, while VP4 (252 aa) is present in lesser amounts. Major neutralizing epitopes are carried on VP2, suggesting that it is at least partly exposed in the outer surface of the capsid, and the VP2 polypeptide is glycosylated in the virion [16, 18, 19]. The nucleotide sequence of segment B encoding the protein VP1 is found in two forms: RNA-dependent RNA polymerase (RdRp), which occurs as a free polypeptide, and a protein VPg, which is linked to the 5′ end of both genome segments [20]. The aquabirnaviruses can be divided into two distinct serogroups, A and B, based on results from cross-neutralization assays [21]. Serogroup A contains the majority of IPNV isolates, including those associated with disease in salmonids, and within this group nine serotypes exist: A1 (reference strain WB)(West Buxton), A2 (Sp)(Spajarup), A3 (Ab)(Abild), A4 (He)(Hecht), A5 (Te)(Tellina), A6 (C1)(Canada 1), A7 (C2)(Canada 2), A8 (C3)(Canada 3), and A9 (Ja)(Jasper) [22]. Serotype A1 includes USA isolates, serotypes A6–A9 are mainly detected in Canada, and serotypes A2–A5 and B1 are found in Europe and Asia [23]. Blake et al. demonstrated that IPNV strains were clustered into six genogroups (1–6) based on amino acid sequences of the VP2 [1]. Many researchers suggested that all marine birnaviruses should form a new genogroup (genogroup 7) different from the six IPNV genogroups [24–26].

In China, an increase in outbreaks of IPN in rainbow trout has been reported in recent years [12, 27]. However, no gene sequences of any Chinese IPNV strains have been submitted to the GenBank database. There is only one complete gene sequence of a Chinese marine birnavirus strain, Paralichthys olivaceus birnavirus (POBV), that is available [28]. However, the marine birnavirus is of the different genogroups and serotypes with the IPNV isolated from salmonids. In this study, a novel IPNV strain was isolated from farmed rainbow trout in Yunnan Province, southeast China, in 2013. This strain was designated as IPNV ChRtm213. To facilitate the detailed comparison of the sequence characteristics of a Chinese IPNV strain with other phylogenetically related IPNV strains isolated from around the world, the complete IPNV ChRtm213 genomic sequences were described. The sequence of IPNV ChRtm213 was the first sequence of IPNV isolates in China, and it will be used as a reference strain of IPNV in China in the future.

Materials and methods

Cell line and virus strain

Moribund rainbow trout fry samples were collected from Yunnan province of southeast China in 2013. The Chinook salmon Embryo cell line (CHSE-214) was used to isolate and propagate the IPNV isolate. The methods of virus stock, growth, and purifications refer to a previous study [17]. The IPNV isolate was passaged two times prior to sequencing.

Primers design

To design the overlapping primers, available complete genome sequences of IPNV isolates from the National Center for Biotechnology Information were aligned. Five pairs of overlapping primers that cover the entire IPNV genome were designed using Oligo 6.0 and Primer Premier Version 5 software on the basis of the full-length genome sequences of IPNV isolates in GenBank database. The 5′ rapid amplification of cDNA ends (5′ RACE) [29] represents useful PCR-based approaches for the determination of 5′ flanking cDNA sequence. The 3′ RACE by lariat-dependent nested PCR (3′ RACE LaNe) [30] is a simple and rapid method that allows “two-sided” gene-specific, fully nested PCR series for the elucidation of 3′ flanking cDNA sequence without the requirement for ligation [31]. All primers were synthesized by Shanghai Sangon Biological Engineering Technology & Services Co., Ltd. Table 1 presents the primers used in this study.

Table 1 Characteristics of the primers used to amplify genome cDNA of IPNV ChRtm213 isolate
Full size table

Amplification of the IPNV complete genome

Total RNA was extracted from the cell culture supernatant using the RNeasy RNA extraction kit (Qiagen, Germany) following manufacturer’s instructions and then stored at −80 °C. Reverse transcription polymerase chain reaction (RT-PCR) was performed in different tubes with use of the PrimeScript™ One-step RT-PCR kit (TaKaRa, Shiga, Japan). Viral RNA (1 μg in 4 μL H2O) was mixed with 0.5 μg of random hexamers in a final volume of 7.2 μL. Concentration of primers used in PCR is 1 mmol/L. Amplification reactions were carried out on polymerase chain reaction (PCR) instrument (Eppendorf AG, Germany) using the polymerase Ex Taq (TaKaRa, Shiga, Japan) with heated lid, programmed to perform 1 cycle of 50 °C for 30 min, then, 1 cycle of 94 °C for 2 min followed by 30 cycles of denaturing at 94 °C for 30 s, annealing at 55 °C for 30 s and extension at 72 °C for 2 min followed by a single extension step of 5 min at 72 °C. The amplified products were analyzed by electrophoresis on a 1% agarose gel. The gel was stained with ethidium bromide and visualized under UV light.

The sequences of two ends of segment A were determined using the 5′ RACE and 3′ RACE systems for amplification of cDNA ends Version 2.0 (Invitrogen) using both dC-tailing and dA-tailing methods according to the manufacturer’s instructions. This assay was done in duplicate to obtain the complete nucleotide sequence of segment A and segment B of the Chinese IPNV isolate ChRtm213, and the sequences were deposited in GenBank database.

Sequence alignment and analyses

All PCR products with the expected size were separated from the agarose gel and purified using an AxyPrep DNA Gel Extraction Kit (Axygen, USA). The purified products were cloned using a pMD18-T simple TA vector (TaKaRa, Dalian). For each of the overlapping fragments, at least three positive clones were randomly selected and sequenced by Shenzhen, China-BGI Tech Solutions Co., Ltd.

The contiguous region between the forward and reverse sequences was assembled and verified using DNASIS software (Hitachi Software Engineering Co., Ltd.). Lasergene software (DNAStar, Inc., USA) was used to translate and conduct multiple alignments of the sequences, as well as to determine the percentage identities and similarity scores. To analyze the percentage identities and similarity scores, both nucleotide and amino acid sequences were aligned using the default options in Clustal W. Phylogenetic trees were constructed using the neighbor-joining method using MEGA Version 5.05 software [33]. The Kimura two-parameter distance measurements were used as the substitution model. The reliability of the phylogeny groupings was evaluated using bootstrapping with 1000 replicates. Aquabirnaviruses isolates used for analysis in this study are presented in Table 2 (segment A) and Table 3 (segment B), respectively. All these isolates had clear genogroups and were commonly used as reference strains in previous studies [14, 34].

Table 2 Segment A of aquabirnaviruses isolates used in this study
Full size table
Table 3 Segment B of aquabirnaviruses used in this study
Full size table

Results

Genome organization of the IPNV strain ChRtm213

The IPNV isolate ChRtm213 was isolated and propagated in CHSE-214 cells, and was harvested when most of the cells showed obvious cytopathic effect (CPE) (Fig. 1). After two passages in CHSE-214 cells, IPNV was collected and subjected to genome sequencing. Segments A and B of IPNV isolate of ChRtm213 that covered all ORFs and flanking regions were assembled from RT-PCR amplifications. Segment A of the ChRtm213 is 3099 bp long and producing fragments of 1359, 918, 447, and 750 bp encoding VP2, VP4, VP5, and VP3 genes, respectively. Segment B of the ChRtm213 genome is 2789 bp long and encodes VP1 protein, a minor internal protein (93 kDa). The segment B of ChRtm213 begins with the conserved sequence GATTC and ends with the sequence GGCGAC, containing the ORF of VP1, which consists of 2535 nucleotides, beginning at the nucleotide 135 and ending with a TGA termination codon at nucleotide 2670. The full-length genomic nucleotide sequence of strain ChRtm213 was deposited in the GenBank database under Accession numbers KX234591 (segment A) and KX234590 (segment B). Schematic diagram of the deduced genomic organization of ChRtm213 and comparison with IPNV isolates from different genogroups are shown in Fig. 2.

Fig. 1
figure 1

Cytopathic effect (CPE) of CHSE-214 cells caused by IPNV. Healthy CHSE-214 cells were used as a negative control (a). Monolayers of CHSE-214 cultured in six-well plate were inoculated with IPNV ChRtm213 with a multiplicity of infection (MOI) of 0.1 (b)

Full size image
Fig. 2
figure 2

Schematic diagram of the deduced genomic organization of ChRtm213 and comparison with IPNV isolates from genogroup 1–7, including 2310, 578, Canada 1, Canada 2, 1146, He, and POBV, respectively. The straight lines with arrow show locations of the cDNA clones used in determining the ChRtm213 complete genome. Shaded areas indicated the coding regions, and the unshaded areas represent non-coding regions (NCR). Numbers indicated the start and end of the respective ORF

Full size image

Homology and phylogenetic analyses

Unrooted phylogenetic trees for genomic segment A and B were built by the neighbor-joining method [35–37]. The phylogenetic tree analyses of the sequences of the segment A and B of various IPNV strains are shown in Figs. 2 and 3, respectively. Aquabirnavirus strains were clustered into seven genogroups in the phylogenetic tree based on segment A (Fig. 3). The results of phylogenetic analysis of the segment A showed that ChRtm213 clustered with IPNV strains including Spain 2310, Japanese AM98, American VR299, Buhl, WB, Korean HL-1, Canadian Jasper, and Mexican Mexican within the genogroup 1 (Fig. 3), and maintained 97.3, 96.3, 96.9, 97.0, 90.4, 96.5, 96.8, and 96.7% nucleotide identity with the eight strains, respectively (Table 4). The succeeding nearest neighbor of ChRtm213 was Japanese aquabirnavirus Y-6 and had 84.6% nucleotide identity for the segment A. The results of phylogenetic analysis of the segment B were accordant with the results from the phylogenetic analysis of the segment A, and the ChRtm213 also clustered with these IPNV strains within the genogroup 1 (Fig. 4). The succeeding nearest neighbor of ChRtm213 also was Japanese aquabirnavirus Y-6 and had 85.7% nucleotide identities for the segment B. The ChRtm213 isolate had a relative close relationship with a Chinese marine birnavirus POBV isolated from P. olivaceus, and the identities of segment A and B were 84.3 and 84.5%, respectively (Table 4). The results indicated that the ChRtm213 isolate belonged to genogroup 1, and has closer relationship with reference strains 2310, Jasper, and AM-98.

Fig. 3
figure 3

Phylogenetic relationship of ChRtm213 and other reference aquabirnaviruses strains based on complete genome of segment A. Phylogenetic tree was constructed using the neighbor-joining method and bootstrap analysis (n = 1000) in MEGA 5.05. The ChRtm213 strain is highlighted in black triangle

Full size image
Table 4 Similarities of IPNV ChRtm213 with other aquabirnaviruses
Full size table
Fig. 4
figure 4

Phylogenetic relationship of ChRtm213 and other reference aquabirnaviruses strains based on complete genome of segment B. Phylogenetic tree was constructed using the neighbor-joining method and bootstrap analysis (n = 1000) in MEGA 5.05. The ChRtm213 strain is highlighted in black triangle

Full size image

Amino acid similarities of each ORF from ChRtm213 with other aquabirnaviruses from different genogroups were analyzed based on genomic sequences. Single IPNV isolates representing genogroups 1–6 and a marine birnavirus representing genogroup 7 were used in the analysis (Table 5). The results showed that the ChRtm213 had the highest identities with Spain 2310 strain of genogroup 1. The similarities of VP1, VP2, VP3, VP4, and VP5 between ChRtm213 and Spain 2310 were 99.4, 97.1, 98.2, 99.2, and 91.0%, respectively. The results showed accordance with that of the phylogenetic tree analyses, strongly supporting that the ChRtm213 belongs to genogroup 1. Among these isolates from different genogroups, an isolate which has the succeeding highest similarity with ChRtm213 was the Chinese marine birnavirus POBV belonging to genogroup 7. The similarity identities of VP1, VP2, VP3, VP4, and VP5 between ChRtm213 and POBV were 97.0, 89.0, 89.4, 88.1, and 80.6%, respectively (Table 5). Compared to other viral proteins of IPNV, VP5 is not conservative, and the similarities of VP5 between ChRtm213 and other strains were from 59.4 to 91.0%. In the ChRtm213, three extra termination codons were induced in the location of amino acid 43, 48, and 104 in the VP5, resulting in a truncated VP5 (Fig. 5).

Table 5 Similarities of ORFs between IPNV ChRtm213 and aquabirnaviruses from genogroups 1–7
Full size table
Fig. 5
figure 5

Alignment of pVP5 from ChRtm213 with other IPNV strains. Red squares indicate termination codons (Color figure online)

Full size image

Discussion

The first report of an IPNV outbreak in northwest China involved the rainbow trout fry and occurred in 1986 [10] followed by outbreaks of IPN occurred in Heilongjiang, Jilin, Shandong, Liaoning, and Gansu provinces of China at the late 1980s [10, 11, 13, 38]. In these previous reports, Chinese IPNV strains were isolated and characterized, and serological analyses showed that all the isolated IPNV strains more likely belonged to A2 (Spajarup, Sp) serotype [10, 11, 21]. After this event, no available information of IPN in China can be obtained until 2011 in which another IPNV strain was isolated from a rainbow trout farm in Liaoning province of China [12]. Although there are many reports on IPNV in China, no sequence of any Chinese IPNV isolates is available, and previous studies showed that IPN mainly occurred in the northern part of China. However, in this study, the IPNV ChRtm213 was isolated in Yunnan province, southwest China in 2013, and complete genomic sequence of the ChRtm213 was sequenced and deposited in the GenBank database.

Cutrín et al. proposed placing aquabirnaviruses into six genogroups, four of which can be subdivided into two genotypes based on a two-step restriction analysis, and the genotyping corresponds with serotyping as follows: genogroup I includes two genotypes corresponding to serotypes A9 (genotype I.1) and A1 (genotype I.2); genogroup II corresponds to serotype A3; genogroup III includes genotypes III.1 (serotype A2) and III.2 (serotype B1); genogroups IV and V include two genotypes, each corresponding to serotypes A5, A6, A7, and A8 (genotypes IV.1, IV.2, V.1, and V.2, respectively); and genogroup VI corresponds to serotype A4 [39, 40]. In this study, the phylogenetic trees of aquabirnaviruses genomic nucleotide sequences of segment A showed that aquabirnaviruses were clustered into seven genogroups (genogroup 1–7), with six (genogroup 1–6) for IPNV and one (genogroup 7) for marine birnaviruses. The ChRtm213 belongs to the genogroup 1, which included the Spain 2310, Japanese AM98, American VR299, Buhl, WB, Korean HL-1, Canadian Jasper, and Mexican Mexican. The genogroup 1 corresponded to the genogroup I in Cutrín’s study [39]. The genogroup I consisted of genotype I.1 and genotype I.2, and the ChRtm213 was clustered into genotype I.1 along with the Japanese AM98, Canadian Jasper, and Korean HL-1. The ChRtm213 was supposed to share the same serotype with those isolates. Therefore, the ChRtm213 more likely belonged to serotype A9 (Ja). Due to the high similarities between the ChRtm213 with each strain within the genogroup 1, the source of the ChRtm213 was still uncertain. Although phylogenetic analysis alone is insufficient to conclude from which country IPNV was introduced into China, the historical record demonstrates that 4,000,000 rainbow trout eggs from Japan were imported to a fish farm in northeast China from 1986 to 1987, and IPNV and IHNV (Infectious hematopoietic necrosis virus) were isolated from rainbow trout fry in northeast China after the importation of eggs from Japan [11]. IHNV isolates in China have been proved to be imported from Japan after 1980s by comprehensive phylogenetic analyses based on a large number of Chinese IHNV isolates [11]. However, there is no gene sequence of Chinese IPNV isolates available; therefore, we were not able to perform extensive sequence analyses on Chinese IPNV isolates. Although the ChRtm213 had high similarities on genomic sequences with Japanese AM98 strain, there still were no data to support the source country of Chinese IPNV was Japan. Previous studies demonstrated that Chinese IPNV strains isolated in the late 1980s were of serotype A2 (Sp) with high virulence and mainly found in the northern part of China [10, 13]. The results suggest that the ChRtm213 is a novel strain that is different from the Chinese IPNV isolates described previously. For now, there is only one genomic sequence of Chinese aquabirnaviruses, and the marine birnavirus strain POBV submitted to the GenBank database [24]. The POBV was isolated from flounder P. olivaceus, and clustered to the genogroup 7 in this study. Although it is having relevant high similarities with ChRtm213, the marine birnavirus POBV still has many differences with the IPNV ChRtm213, such as the genogroup, serotype, and host. It is more meaningful to analyze ChRtm213 with other Chinese IPNV stains but the POBV. The results suggest that more sequences of Chinese IPNV isolates should be determined and submitted to GenBank database or some other databases, which will play an important role in the analyses of the source and evolution of IPNV in Chinese fish farming environment. The genomic sequence of the ChRtm213 characterized in this study was the first sequence of IPNV in China. The information of IPNV ChRtm213, such as genomic sequence, genogroup, and serotype, will provide an important reference to the research of Chinese IPNV isolates in the future.

The VP5 gene in aquabirnaviruses was not as conserved as the polyprotein or RdRp, which has been detected in virus-infected cells, but so far there is no evidence of the protein being found in purified virions. Some studies support the presence of VP5 [41], and a weak VP5 ORF-expressed product in CHSE-214 cells infected by the Jasper strain of IPNV has been detected [42]. Additionally, a previous study demonstrated that the VP5 is not essential for virulence or for the establishment of persistent IPNV infection. The role of VP5 in persistent infection was investigated by challenging Atlantic salmon fry with the recovered viruses, as well as with the low-virulence field strain Sp103 and a naturally occurring VP5-deficient mutant of Sp103. The results showed that VP5 is not required for viral replication in vivo, and its absence did not alter the virulence characteristics of the virus or the establishment of persistent IPNV infection [43–45]. In this study, a truncated VP5 was observed and confirmed in the ChRtm213 strain. However, it did not affect propagation of IPNV in CHSE-214 cells either. Thus, we conjectured that the VP5 is not essential for viral replication in cell line.