Introduction

Infectious bursal disease (IBD) is a highly contagious immunosuppressive disease of chickens, which mainly attacks the bursa of Fabricius with massive destruction of B-lymphocytes in these lymphoid organs, and continues to pose a great threat to the poultry industry worldwide (Berg 2000; Müller et al. 2003; Mahgoub et al. 2012). In recent years, many efforts have been made to protect chickens against the causative agent, infectious bursal disease virus (IBDV). However, challenges still remain serious due to the emergence and rapid spread of new variant strains, especially those very virulent strains of IBDV (vvIBDV) which can break through a high level of maternally derived antibodies and cause high rates of morbidity and mortality (Lasher and Davis 1997; Müller et al. 2003; Martinez-Torrecuadrada et al. 2003). Current protection of chickens against IBDV is mainly achieved by vaccinating breeding hens with conventional inactivated or live-attenuated vaccines (Rong et al. 2005), however, some disadvantages have emerged in those vaccines, including less efficiency or inefficiency in protecting chickens from vvIBDV, potential of attenuated vaccines reverting to virulence by the recombination of RNA segments, and elaborate and costly production process (Berg et al. 1996; He et al. 2009). It is, therefore, urgent to develop new vaccines such as subunit vaccines with higher efficacy and safety to protect against IBDV.

Among the major structure proteins of IBDV, the VP2 is a unique component of the icosahedral capsid that is responsible for virulence, cell tropism, and antigenic variation, and is the best target for use in developing subunit vaccines (Rong et al. 2005; Coulibaly et al. 2005; Saugar et al. 2005). Recently, the recombinant VP2 (rVP2) has been experimentally produced using several expression systems, including Saccharomyces cerevisiae (Arnold et al. 2012), Pichia pastoris (Pitcovski et al. 2003), Escherichia coli (Rong et al. 2005), Aspergillus niger (Azizi et al. 2013), Arabidopsis thaliana (Wu et al. 2004), and Nicotiana benthamiana (Gómez et al. 2013) etc. Most of them were immunogenic and the rVP2-based subunit vaccine could confer the tested animals with full or partial protection, indicating that the development of subunit vaccines using rVP2 is a hopeful strategy to protect chickens against infection of IBDV.

The silkworm, Bombyx mori, is a well-known economic insect that has excellent abilities to synthesize vast amounts of silk protein in its silk gland and secrete them into the cocoon (Tomita 2011). Since the stable germline transformation in B. mori was achieved using piggyBac, a transposon derived from the lepidopteran Trichoplusia ni (Tamura et al. 2000), production of recombinant proteins in the silk gland of transgenic silkworm has attracted significant interest in recent years. Several expression systems such as sericin1 (Ser1)-expression system, fibroin heavy chain (FibH)-expression system, and fibroin light chain (FibL)-expression system, have been constructed to express recombinant protein in the middle or posterior silk gland. The Ser1-expression system is often preferred due to the hydrophilic performance of sericin, which makes it easy to extract target proteins from cocoons. Some recombinant proteins with biological activity have been experimentally produced in the middle silk gland of transgenic silkworm, including human serum albumin (Ogawa et al. 2007), human granulocyte–macrophage colony-stimulating factor receptor α (Urano et al. 2010), human collagen (Adachi et al. 2010), and mouse monoclonal antibody (Iizuka et al. 2009), which demonstrated the capacity of silkworm as a practical tool to produce valuable proteins for use in pharmaceutical and biomedical applications.

The objective of this study is to investigate the feasibility of producing rVP2 in the middle silk gland of transgenic silkworm. By using an optimized Ser1-expression system developed previously (Wang et al. 2013), a total of 16 transgenic silkworm lines were generated to express a codon-optimized VP2 gene of vvIBDV in its middle silk gland. Analysis of the expression, characterization and immunogenicity indicated that the rVP2 was efficiently synthesized in the middle silk gland of transgenic silkworm and displayed immunogenic activity. This study is the first to report the overexpression of recombinant animal disease virus antigen using transgenic silkworm.

Materials and methods

Animals

The non-diapaused silkworm strain D9L of B. mori, was maintained in our laboratory and used for germline transformation. Eggs were maintained at 25 °C with 95 % humidity until hatching, and larvae were reared on fresh mulberry leaves at 25–26 °C. The specific pathogen-free KM mice were purchased from Chongqing Tengxin Biological Technology Co. Ltd, and maintained in isolators according to the institutional guidelines.

Vector construction and germline transformation

Nucleotide sequence of the VP2 of vvIBDV was obtained from GenBank (accession no. JF907703), and commercially synthesized after optimizing its codon usage for the B. mori with an extra 6 × His-tag fused at its C-term. The VP2 fragment was digested with BamH I and Not I and inserted into the pSL1180 [hSer1sp-DsRed-Ser1PA] vector (Wang et al. 2013) to replace the DsRed gene, then the hSer1sp-VP2-Ser1PA cassette was cloned into the Asc I site of the piggyBac-containing vector pBac[3 × P3EGFPafm] (Horn and Wimmer 2000) to generate the donor vector pBac[hSer1sp-VP2-Ser1PA, 3 × P3EGFP], designated as pBhSer1VP2 (Fig.S1).

The donor plasmid pBhSer1VP2 and the piggyBac transposase-expressing plasmid pHA3PIG (Tamura et al. 2000), both at 500 μg/μL, was mixed and microinjected into pre-blastoderm eggs 1–4 h after oviposition following the methods previously described (Tamura et al. 2000). G0 moths developing from hatched eggs were mated with each other to generate G1 progeny. Day-5 to Day-7 G1 eggs were screened for EGFP expression in the ocelli using a fluorescent stereomicroscope (Olympus, Japan). EGFP-positive G1 moths were backcrossed to generate G2 offspring.

Expression and stability analysis of rVP2

The mRNA level of VP2 in the middle silk gland of each 5th instar Day-7 larva of transgenic silkworm was determined by qRT-PCR according to the manufacturer’s instructions. The rVP2 protein was extracted from transgenic cocoons following the method described previously (Wang et al. 2013), and detected by western blotting using a commercially prepared polyclonal anti-VP2 antibody (Dragon-lab Instruments, China). The content of rVP2 in each cocoon was calculated by densitometric measurement of the CBB-stained gels and immunoblot using Image J software.

The thermal stability of rVP2 was tested by dissolving the cocoon proteins in 50 mM Tris–HCl buffer containing 8 M Urea (pH 8.0) to a final concentration of 20 μg/mL, and treated with 25, 40, 60, and 80 °C for 0.5, 1, 1.5, and 2 h, respectively. The acid–base stability of rVP2 was evaluated by dissolving the cocoon proteins in same Tris–HCl buffer (pH 1.5, 2.5, 5.5, 7.0, 7.5, 8.0, 9.0, 10.0, and 12.0, respectively) to a final concentration of 20 μg/mL, and incubated at 25 °C for 12 h. All samples were subjected to SDS-PAGE and western blotting.

Extraction and purification of rVP2

Transgenic cocoons containing high level of rVP2 were pre-cooled in liquid nitrogen and crushed into powders. The powders were dissolved in 50 mM Tris–HCl (pH 8.0) containing 8 M Urea, incubated at 80 °C for 30 min and further homogenized at 4 °C overnight. The supernatants were collected by centrifugation at 18,000 rpm for 15 min. Then the extracts were purified using a Ni-affinity column and further dialyzed against urea using a HiTrap Q-HP column (GE, USA). Peptide sequence of rVP2 was determined by MS+MS/MS at Shanghai Applied Protein Technology Co. Ltd, and analyzed using the Mascot 2.2 software.

Immunogenicity evaluation of rVP2

Four-week-old mice were randomly divided into three groups, with each group containing five individuals. For subcutaneous immunization, the purified rVP2 was mixed with complete Freund’s adjuvant (CFA, Sigma) in the first application (4th week) and with incomplete Freund’s adjuvant (IFA, Sigma) in the second and third vaccinations (6th and 8th week). Each mouse was immunized with 100 μL emulsion containing 20 μg rVP2. Negative control group was immunized with 100 μL CFA. Positive control group was immunized with 100 μL mixture of IBDV vaccine (B87 strain, Shanghai Haley Bio-Pharmaceutical Co. Ltd, China) and adjuvant. Two weeks after the last immunization, blood samples were taken from eyeball. During the experiment, the mice were anesthetized or euthanized with 1.5 % isoflurane. The immunogenicity of rVP2 was determined by indirect enzyme-linked immunosorbent assay (ELISA) and agar-gel precipitation (AGP) test. PBS buffer (pH 7.4), IBD antigen and IBD positive serum (China Institute of Veterinary Drug Control) were used as negative and positive controls, respectively.

Results and discussion

The rVP2 is highly expressed in the middle silk gland of transgenic silkworms

To generate the transgenic silkworm expressing codon-optimized VP2 in its middle silk gland, the mixture of pBhSer1VP2 and pHA3PIG plasmids was microinjected into 400 silkworm eggs. A total of 110 moths were recovered and crossed, leading to 16 transgenic lines that produced EGFP-positive progeny (Fig.S2). Transcriptional analysis of VP2 in the middle silk gland of six randomly selected transgenic lines showed that the mRNA level of VP2 was lower than the endogenous Ser1, and accounted for about 13.90–66.10 % of Ser1 mRNA (Fig. 1a). Further analysis revealed that rVP2 proteins with a molecular weight of about 47 kDa were present in all transgenic cocoons. The content of rVP2 in the cocoon of each line was ranged from 0.07–16.10 % of the total soluble proteins (Fig. 1b). These results indicated that the rVP2 was successfully synthesized in the middle silk gland of transgenic silkworms and secreted into cocoons. The variation of rVP2 expression might be caused by some surrounding genome sequences on integrated transgenes (position effects).

Fig. 1
figure 1

Expression analysis of rVP2. a Transcription level of VP2 and endogenous Ser1 in the middle silk gland of wild-type (WT) and transgenic lines (L2, 6, 7, 8, 13, 14). Percentages of VP2 mRNA relative to Ser1 mRNA are shown on the top of each figure. b Protein level of rVP2 in the cocoon of transgenic lines. Proteins dissolved from the cocoon of wild-type (WT) and transgenic lines (L1–16) were separated by SDS-PAGE and immunoblotted with anti-VP2 antibodies. Arrowheads point to the band of rVP2. Percentages indicate the content of rVP2 in total soluble cocoon proteins of each line

Full size image

The rVP2 is more stable at different temperatures than at different pH values

To investigate the thermal and acid–base stability of the rVP2, cocoon proteins were dissolved and treated with different temperatures and pH values, respectively. As shown in Fig. 2, the rVP2 was detected at its expected molecular weight by western blotting at all conditions, except in the case of pH 1.5. Some small products were also observed in most cases. Increasing the temperature had less impact on the stability of rVP2, and appeared to promote its solubility. In contrast, strong acidity or basicity could result in the degradation of the rVP2. These results revealed that the rVP2 was able to tolerate high temperatures up to 80 °C, but only remained stable at neutral pH conditions.

Fig. 2
figure 2

Thermal and acid–base stability analysis of rVP2. Transgenic cocoons were dissolved and treated with different temperatures and pH values shown on the top of the figure, and detected by western blotting. WT wild-type

Full size image

The rVP2 is easily purified from the transgenic cocoons

To extract and purify the rVP2, transgenic cocoons containing high level of rVP2 were homogenized in Tris–HCl buffer containing 8 M urea, extracted by Ni-affinity column and further purified using a HiTrap Q-HP column. As a result, 3.33 mg of rVP2 with a purity >90 % was obtained from 30 g cocoon powders (111.0 μg/g cocoon powders), in which other proteins were barely visible in SDS-PAGE (Fig. 3). Determination of the rVP2 peptide sequence showed that five peptides with the length of ~30 aa matched well with the original sequence of VP2 (Fig. S3). These results demonstrated that the rVP2 was correctly synthesized in the middle silk gland of transgenic silkworms and could be easily purified from their cocoons. Despite there were some losses of target proteins during the purification process, the current yield of rVP2 was still acceptable and enough to be used for further study.

Fig. 3
figure 3

Extraction and purification of rVP2. Proteins in wild-type (WT) and transgenic cocoons (lane 1) were dissolved with Tris–HCl buffer and purified using a Ni-affinity column. The presence of rVP2 in flowthrough solution (lane 2) and rinsing solutions eluted with Tris–HCl buffer (lane 3), 10 mM imidazole (lane 4), 20 mM imidazole (lane 5), 30 mM imidazole (lane 6), and 120 mM imidazole (lane 7), respectively, was monitored by SDS-PAGE and western blotting. The rVP2 eluted with 120 mM imidazole was concentrated (lane 8) and dialyzed against urea using a HiTrap Q-HP column (lane 9). Arrowheads point to the band of rVP2

Full size image

The rVP2 displays the immunogenicity in mice

To evaluate the immunogenicity of the rVP2, serum samples were collected from mice immunized with the purified rVP2, and analyzed by indirect ELISA and AGP test. As revealed in Fig. 4, both the rVP2 and B87 vaccine induced antibodies in mice. The rVP2 was recognized by the antibodies in rVP2 serum, while the IBD antigen only interacted with the antibodies in B87 serum. In AGP test, the immunoprecipitation between the rVP2 and rVP2 serum as well as the IBD antigen and B87 serum or IBD positive serum was detected, but there was no clear immunoprecipitation line between the rVP2 and B87 serum or IBD positive serum, or IBD antigen and rVP2 serum (Fig. S4). It has been reported that the VP2 sequence is highly variable among IBDV strains and largely determines the antigenic variation of different strains, even a single mutation in the epitope region may change its immunogenic specificity (Eterradossi et al. 1997; Mundt 1999; Jackwood et al. 2011). Sequence analysis of the rVP2 and the VP2 of B87 strain showed that there were several amino acid mutations in their epitope region (Fig. S3), which might be the reason of unobvious immunoprecipitation between the rVP2 and positive controls. Briefly, these results revealed that the rVP2 had specific immunogenic property. In the future, we will further analyze the characterization of the rVP2 and evaluate its immunogenicity in chickens.

Fig. 4
figure 4

Immunization analysis of the purified rVP2. The rVP2 serum (left) and B87 vaccine serum (right, positive control) were collected from mice immunized with the purified rVP2 and B87 vaccine, respectively. The immunogenicity of rVP2 was determined by indirect ELISA. The OD value is the mean ± SD obtained from three independent assays

Full size image

Conclusions

In the present study, we generated 16 transgenic silkworm lines that synthesized rVP2 in its middle silk gland and secreted it into cocoons. The content of rVP2 in the cocoon of each line was various and ranged from 0.07–16.10 % of the total soluble proteins. The rVP2 was able to tolerate high temperatures and neutral pH conditions. The rVP2 was purified from the cocoons with a yield of 111.0 μg per gram of cocoon powders and a purity of >90 %, and displayed specific immunogenic property in mice. This study provides a new method for the production of rVP2, and demonstrates the feasibility of using silkworm to produce recombinant immunogens for use in subunit vaccines against animal diseases.