• Open Access

Oral immunization with transgenic rice seeds expressing VP2 protein of infectious bursal disease virus induces protective immune responses in chickens

Authors


* Correspondence (fax +86 571 86971498; e-mail zzhou@zju.edu.cn)

Summary

The expression of infectious bursal disease virus (IBDV) host-protective immunogen VP2 protein in rice seeds, its immunogenicity and protective capability in chickens were investigated. The VP2 cDNA of IBDV strain ZJ2000 was cloned downstream of the Gt1 promoter of the rice glutelin GluA-2 gene in the binary expression vector, pCambia1301-Gt1. Agrobacterium tumefaciens containing the recombinant vector was used to transform rice embryogenic calli, and 121 transgenic lines were obtained and grown to maturity in a greenhouse. The expression level of VP2 protein in transgenic rice seeds varied from 0.678% to 4.521% µg/mg of the total soluble seed protein. Specific pathogen-free chickens orally vaccinated with transgenic rice seeds expressing VP2 protein produced neutralizing antibodies against IBDV and were protected when challenged with a highly virulent IBDV strain, BC6/85. These results demonstrate that transgenic rice seeds expressing IBDV VP2 can be used as an effective, safe and inexpensive vaccine against IBDV.

Introduction

Infectious bursal disease virus (IBDV) is the causative agent of infectious bursal disease (IBD), a highly contagious and mortal disease in young chickens, which causes worldwide economic losses in the poultry industry. Two distinct serotypes of IBDV have been identified. Serotype I viruses are pathogenic to chickens (Winterfield and Thacker, 1978), whereas serotype II viruses, isolated from both turkeys and chickens, are apathogenic (Ismail et al., 1988). Serotype I viruses can be further divided into attenuated, classical virulent, antigenic variant, and very virulent IBDV strains based on pathotype (Wei et al., 2006). Virulent IBDV isolates causes severe immunosuppression in young chickens by destroying the immature B lymphocytes in the bursa of fabricius, which increases susceptibility to other diseases and decreases vaccination effects against infectious bronchitis, Newcastle disease, or Marek's disease (Kibenge et al., 1998; Van den Berg, 2000). IBDV is a member of the Birnaviridae family whose genome is composed of two segments (A and B) of double-stranded RNA (Azad et al., 1985; Brown and Skinner, 1996). Segment A contains two partially overlapping open reading frames (ORF). The first ORF encodes a small non-structural protein (VP5) that is dispensable for virus replication, and the second ORF encodes a 110-kDa polyprotein that is autocatalytically cleaved to generate three viral proteins, VP2 (47 kDa), VP4 (28 kDa) and VP3 (32 kDa) (Hudson et al., 1986). VP2 and VP3 are the major structural proteins of IBDV particles, while VP4 is the viral protease responsible for self-processing of the polyprotein (Mundt, 1999; Sanchez and Rodriguez, 1999; Caston et al., 2001). VP2 is the major host-protective antigen responsible for induction of neutralizing antibodies (Becht et al., 1988; Fahey et al., 1989). Segment B encodes a multifunctional protein (VP1) with RNA-dependent RNA polymerase and capping enzymatic activities (Yehuda et al., 1999; Tacken et al., 2000).

Currently, IBD is controlled by administration of an attenuated live or an inactivated IBDV vaccine in the poultry industry (Lai et al., 2004; Li et al., 2006). However, these vaccines have a number of disadvantages. The inactivated vaccine is costly, poorly effective and difficult to administer, while the live attenuated vaccine may result in the emergence of IBDV variants due to virus recombination. Thus, there is an urgent need to develop new vaccines with higher efficacy and fewer side-effects. A variety of different expression systems have been used to express VP2 for the development of subunit vaccines against IBDV, and successful VP2 expression were reported in Arabidopsis thaliana (Wu et al., 2004) and Saccharomyces cerevisiae (Pitcoviski et al., 2003) or from a baculovirus vector (Vakharia et al., 1994; Pitcovski et al., 1996; Lai et al., 2004), an avian herpesvirus vector (Darteil et al., 1995; Tsukamoto et al., 2002), an adenovirus vector (Francois et al., 2004), a fowlpox virus (Shaw and Davison, 2000) and Marek's disease virus (Tsukamoto et al., 1999).

Glutelin is the most abundant storage protein, accounting for about 80% of the total endosperm protein in rice. Glutelin is encoded by about 10 genes per haploid genome, which are divided into two subfamilies (GluA and GluB, Takaiwa et al., 1996). The gene for Glutelin and its promoter have been isolated and well characterized (Takaiwa et al., 1996; Wu et al., 1998; Hwang et al., 2002; Yang et al., 2003). Gt1 (also known as GluA-2) is a member of GluA subfamily, whose promoter (Gt1 promoter) is a very strong and endosperm-specific expression promoter (Hwang et al., 2002; Yang et al., 2003).

In this study, the VP2 cDNA of IBDV strain ZJ2000 was cloned downstream of the Gt1 promoter in the binary expression vector, pCambia1301-Gt1. Agrobacterium tumefaciens containing the recombinant vector was used to transform rice embryogenic calli, and transgenic rice lines expressing IBDV VP2 antigen in the rice endosperm were developed. Furthermore, the immunogenicity and the oral immunization efficacy of this seed-derived recombinant VP2 protein in chickens were investigated.

Results

Construction of expression vector and transformation of rice plants

The 35S promoter in pCambia1301-35S was substituted with the 1920 bp Gt1 promoter amplified from rice cultivar Xiushui 11 to produce an endosperm-specific expression vector, pCambia1301-Gt1. The 1530 bp IBDV VP2 gene was then cloned into pCambia1301-Gt1 under the control of the Gt1 promoter to produce pCambia1301-Gt1-VP2 (Figure 1). The VP2 gene was transformed into rice calli (Xiushui 11) mediated by agrobacterium. The transformation efficiency was 20%, and 121 transgenic plant lines were obtained by hygromycin selection. An about 1.5-kb DNA fragment was amplified in all 121 transgenic lines by polymerase chain reaction (PCR) using specific primers for VP2 gene. Southern blot analysis also indicated that the VP2 gene was integrated into the genomic DNA of all 121 transgenic lines and the copy number of the foreign gene was estimated to be one to four, the copy number of the foreign gene in the representative transgenic lines was shown in Figure 2.

Figure 1.

Map of recombinant binary expression vector. The VP2 gene (1530 bp) was under the control of 1920 bp Gt1 promoter from rice glutelin GluA-2 gene. NOS-T, the nopaline synthase gene terminator; LB, left border; RB, right border.

Figure 2.

Southern blot analysis of transgenic rice plants. 1, 23, 47, 53, 57, 99 and 107 are different transgenic rice plants; CK, non-transgenic rice; VP2, PCR products of IBDV VP2 gene (positive control); M, 1Kb DNA Marker.

Expression of VP2 protein in transgenic rice seeds

The expression levels of recombinant VP2 protein in the transgenic rice seeds (T0) were investigated with triple-antibody sandwich–enzyme-linked immunosorbent assay (TAS-ELISA) and Western blots. The 121 transgenic lines expressed different amounts of VP2 protein and the expression level of representative transgenic lines was shown in Figure 3. The calculated averages of VP2 protein in one grain of transgenic seed in individual lines ranged from 0.678% to 4.521% of the total soluble protein. The expression levels of IBDV VP2 protein calculated as the amount of VP2 protein per grain corresponded to 4.521% of total soluble protein in the highest expression line 99, which accumulated up to 56.12 µg per grain of VP2 protein, while the lowest expression line 53 accumulated 10.19 µg per grain of VP2 protein. Western blot analysis revealed that a band with a molecular mass of about 50 kDa reacted with anti-IBDV antiserum in extractions of transgenic rice seeds. However, the expression levels of VP2 in different transgenic rice plants varied and transgenic line 99 show higher accumulation of VP2 protein (Figure 4). In order to test VP2 expression stability, 100 grains of rice seeds of the T1 transgenic rice line 47 was homogenized and the total soluble protein was extracted as described by Takagi et al. (2005). TAS-ELISA showed the recombinant VP2 protein could stable express in T1 transgenic rice seeds and the level of expression reached an average of 40.21 µg of VP2 protein per grain.

Figure 3.

Detection of VP2 protein in transgenic rice seeds by TAS-ELISA. 1, 8, 12, 23, 47, 53, 57, 77, 99 and 107 are different transgenic rice plants. The absorbance value is the mean ± SD obtained from three independent assays. VP2 protein (VP2) from SDS-PAGE gels of purified IBDV particles (strain BC6/85) and non-transgenic rice (CK) were used as positive and negative controls, respectively.

Figure 4.

Western blot analysis of VP2 protein in transgenic rice seeds. 47, 53 and 99 are different transgenic rice plants. CK, non-transgenic rice; VP2, VP2 protein from SDS-PAGE gels of purified IBDV particles (strain BC6/85); M, middle molecular mass marker of proteins.

Protection efficacy against the highly very virulent IBDV challenge

Except for group 6, the 14-day-old specific pathogen-free (SPF) chickens were orally vaccinated with T1 transgenic rice seeds of line 47, which contained an average of 40.21 µg of VP2 protein per grain. At 21 days after the last boosting immunization, chickens in each group, except for group 6, were challenged with the virulent IBDV strain BC6/85. The representative tissue sections of the microscopic histopathological examination of the groups 2, 3, 5 and 6 are shown in Figure 5. Histopathological bursal scoring indicated that the chickens vaccinated with high doses had a protection rate of 83.33% (5/6) (Table 1). When the doses were decreased to 3 g or 1 g, the protection rate declined to 33.33% (2/6) and 16.67% (1/6), respectively. In the control group, none of the chickens (0/6) were protected by feeding on the non-transgenic rice seeds and two out of six chickens were protected in the attenuated live vaccine B87 group. The bursa/BW ratios of the chickens in group 3 vaccinated with transgenic rice seeds (5 g) were not significantly different from those of the normal control group (P > 0.05). All chickens in the unvaccinated but challenged group developed typical clinical signs of IBDV, including watery diarrhoea, dehydration and depression at 2 days post-challenge. At 5 and 6 days post-challenge, two chickens in the unvaccinated but challenged group were dead and the others in this group were still alive at 7 days post-challenge. At necropsy, muscular haemorrhages were evident and all the bursas were significantly atrophied. No bursal histopathological changes were observed in the normal control group. These results obviously indicated that dosage plays an important role in the immune protection against IBDV. The chicken protection assay results show that transgenic rice-derived VP2 protein could induce a strong immunological response in vaccinated chickens. Thus, the VP2 protein in transgenic rice seeds provides a safe, effective and inexpensive approach to prevent IBDV infection in young chickens.

Figure 5.

Sections of haematoxylin and eosin stained bursa of chicken challenged with IBDV strain BC6/85 (×100). Representative chickens with score 0 in group 3 (A), score 1 in group 2 (B), score 2 in group 5 (C) and score 3 in group 7 (D) were shown. The detailed bursal lesion scores were described in our previous report (Li et al., 2003).

Table 1.  Immune protection efficacy of transgenic rice seeds expressing VP2 protein
GroupVaccine*Bursa/BW ratioBursal lesions scoreAntigen in bursa§No. of protected/no. of challenged
0123
  • *

    Groups 1, 2, 3 and 4 were orally vaccinated four times with transgenic rice seeds or non-transgenic rice seeds; group 5 was vaccinated with IBDV live attenuated vaccine. Group 6 was unvaccinated and unchallenged (normal control) and group 7 was unvaccinated and challenged (challenge control).

  • The number expressed is the arithmetic average for six chickens. Mean bursa/BW = (mean bursa in grams × 1000)/total body weight in grams. Values within this column followed by different letters (A and B) are significantly different (P < 0.05).

  • The detailed of bursal lesion scores were described in our previous report (Li et al., 2003) and demonstrated in Figure 5.

  • §

    Chickens with antigen were evaluated based on immunofluorescence assay.

  • The protection evaluation for each challenge group was initially based on gross bursal lesions and further confirmed by bursa histopathological examination.

1Transgenic rice seeds (1 g)1.95B11225/61/6
2Transgenic rice seeds (3 g)2.68B22114/62/6
3Transgenic rice seeds (5 g)4.32A51001/65/6
4Non- transgenic rice Seeds (5 g)2.03B01236/60/6
5Live attenuated vaccine (100 × EID50)4.21A22114/62/6
6Normal control4.67A60000/66/6
7Challenge control1.96B  156/60/6

IBDV antigen detection in the bursa of chickens vaccinated with transgenic rice seeds expressing VP2 protein

To determine whether oral vaccination with transgenic rice seeds can prevent virus replication in the bursa of Fabricius, the presence of IBDV antigen in bursal sections from each vaccinated chicken was detected with an immunofluorescence assay at 7 days after challenge with the virulent IBDV. Oral vaccination with transgenic rice seeds expressing VP2 protein reduced the IBDV antigen-positive ratio (Table 1). Notably, five out of six chickens in group 3 that were vaccinated with a high dose (5 g) of transgenic seeds expressing VP2 protein were antigen-negative and the protection efficacy (5/6) in group 3 was higher than that in group 5 (2/6) vaccinated with IBDV live attenuated vaccine. However, the immune protection efficacy fell when the dose was reduced to 3 g or 1 g. All chickens in the challenge control were positive for antigen detection and no antigen was detected in any of the chickens in the normal control group. This indicates that oral vaccination with seeds expressing the VP2 protein can prevent replication of virulent IBDV in the bursa after challenge.

Induction of serum antibodies responses against IBDV in chickens vaccinated with transgenic rice seeds expressing VP2

Blood samples from each group were collected from the wing vein at day 0 and at 1 week after each immunization, and the serum samples were evaluated for serum antibodies with ELISA. Transgenic rice seeds expressing VP2 protein induced serum antibodies against IBDV. However, chickens in groups 1, 2 and 3 had different levels of antibody response during the experiment (Figure 6). At 14 days after the first vaccination, antibody response was detected in group 3 and group 5. At 28 days after the first vaccination, the geometric mean ELISA IBDV serum antibody titres in group 3 reached 2.81. The results indicated that transgenic rice seeds expressing VP2 protein could induce high titre serum antibodies against IBDV in chickens. Group 2 chickens, which received a medium dose (3 g), also had an antibody response at 14 days after the first vaccination, but the antibody levels were lower than those of group 3. Chickens in group 1, which received a low dose (1 g), exhibited a negligible antibody response even at 21 days after vaccination, and only a low-level antibody response was detected at 7 days after the last vaccination. As expected, an antibody response was not detected in the control group, which received non-transgenic rice seeds throughout the experiment. The unvaccinated chickens and the chickens in the challenge control also never developed a specific IBDV antibody response.

Figure 6.

ELISA antibody titre of the chickens from different groups. ELISA antibody titres for each group are represented as the geometric mean on a log10 scale of the highest dilution that yielded absorption value three times above the pre-immunization serum. Groups 1, 2 and 3 chickens were each orally immunized with 1 g, 3 g, 5 g seeds of transgenic rice, respectively, while group 4 chickens were each orally immunized with 5 g non-transgenic rice seeds. Group 5 chickens were immunized intranasally with a commercial attenuated vaccine strain B87 at 2-week-old and boosted at day 21. Group 6 chickens were unvaccinated and unchallenged to serve as normal control. Group 7 chickens were unvaccinated but challenged to provide a challenge control.

None of the vaccinated chickens had measurable levels of specific IBDV neutralizing antibody at 7 days after first vaccination. At 7 days after the first booster, neutralizing antibody titres increased gradually in chickens vaccinated with transgenic rice seeds expressing VP2 protein or the attenuated vaccine (Figure 7). Transgenic rice seeds expressing VP2 protein (high dose) induced a high level of neutralizing antibody response in the vaccinated chickens. Therefore, consistent with ELISA titres a significant dose-effect was observed. Chickens in the non-transgenic rice seeds group, the normal control and the challenge control groups never developed specific neutralizing antibodies against IBDV.

Figure 7.

Virus neutralization titres analysis of sera from chickens for each group. Virus neutralization titres are expressed as the geometric mean on a log10 scale of the highest dilution which completely prevented the appearance of cytopathic effect for pooled serum samples of each group. Treatments for groups were same as those described in Figure 6.

Safety evaluation of transgenic rice seeds expressing IBDV VP2 protein

A fivefold of the highest dose (25 g) was orally administered to six 2-week-old chickens each day for a month. No mortality, side-effects or signs of any disease were observed during 2 months inspection. The average body weights (3.12 kg) of the chickens fed 25 g of transgenic seeds were not significantly different from those (3.20 kg) of the control group. These results indicate that transgenic rice seeds expressing VP2 protein are safe in chickens when used as a subunit vaccine.

Discussion

In 1992, Mason and colleagues first reported that a transgenic plant expressed a recombinant viral antigen. Since then, transgenic plants have become attractive systems for production of human and animal biopharmaceutical recombinant proteins. To date, many recombinant proteins, including vaccines and antibodies have been expressed in plants and the viral or bacterial recombinant antigens expressed in transgenic plants can induce neutralizing antibodies in vaccinated animals (Mason et al., 1992; Arakawa et al., 1998; Carrillo et al., 1998; Daniell et al., 2001; Gil et al., 2001; Kong et al., 2001; Zhou et al., 2003; Judge et al., 2004; Molina et al., 2005; Takagi et al., 2005). The use of transgenic plants as bioreactors for production of pathogen antigens has many advantages. First, the cost of this system is low and the vaccine production is simple and can be easily scaled up. Second, the subunit vaccines produced by plants are safe and effective since plants have eukaryotic protein processing, modification and correct folding. Third, the oral administration method is simple. However, disadvantages are that the expression levels of foreign genes in plants are often low and in many cases, higher expression levels are required for practical and commercial use. In the present study, we report for the first time that an animal viral protein, IBDV VP2, was highly expressed and accumulated efficiently in rice seeds using an endosperm-specific glutelin Gt1 promoter. Moreover, we demonstrate that chickens orally vaccinated with transgenic rice seeds expressing recombinant VP2 protein produced high titre neutralizing antibodies against a IBDV isolate (strain HZ2) and can protect the chickens (5/6) from the infection of a virulent IBDV isolate (strain BC6/85). The protective efficacy and neutralizing antibody titres was influenced by the dose of transgenic rice seeds. These results demonstrate that transgenic rice seeds expressing IBDV VP2 can be used as an effective, safe and inexpensive vaccine against IBDV. Production of recombinant proteins in seeds of cereal crops offers several advantages. First, recombinant proteins can be stored safely in the protein bodies of the endosperm cell, thereby escaping proteolysis during maturation in the cytosol and subsequent programmed cell death during the final stages of cereal grain maturation. Thus recombinant proteins can be highly and stably accumulated in seeds. Second, very large supplies of recombinant proteins can be made available over several years by storage of the transgenic seeds at ambient temperature. Third, recombinant proteins in seeds of cereal crops are easily transported. Additionally, seed grains are edible plant organs and some animals like chickens can be vaccinated by feeding on the grains expressing the antigen protein without the need for processing, thus making the administration simple and further reducing the cost.

VP2 protein of IBDV strains is highly conservative. The VP2 protein of ZJ2000 is highly related to different representative strains HZ2, B87, JD1 (AF321055), OKYM (D49706), UK611 (AJ318896), IM (AY029166) and GLS (AY368653), with amino acid identity ranged from 97.3% to 99.4%. Our study indicates that the chickens immunized with the recombinant VP2 protein (strain ZJ2000) from rice seeds can produce cross-clade neutralizing antibodies against strain HZ2 and BC6/85. Thus, it is likely that the protective efficacy of the recombinant VP2 protein expressed in rice is applicable over a broad spectrum of different IBDV strains. Neither the mortality, side-effects or signs of any disease nor the average body weights different were observed when chickens were orally administered with the transgenic rice seeds with high dosage (25 g/day). These results indicate that transgenic rice seeds expressing IBDV VP2 immunogen can be used as an effective, safe and inexpensive recombinant subunit vaccine of IBDV.

VP2 protein expressed in transgenic rice seeds has not been investigated with respect to its resistant to gut degradation. The fact that chickens vaccinated with transgenic rice seeds develop high level neutralizing antibody titres indicates that the recombinant VP2 protein in rice seeds is sufficiently resistant to degradation in the chicken gut and elicit an immune response against IBDV. However, further work is necessary to demonstrate the correct conformation of VP2 and the retention of major neutralizing epitopes.

Taken together, our study provides evidence indicating that transgenic rice seeds expressing IBDV VP2 protein are useful vehicles for prevention of IBDV infections in young chickens. Furthermore, our research results indicate that rice seeds have considerable utility for production and delivery of peptide vaccines.

Experimental procedures

SPF animals

Specific pathogen-free (SPF) embryos and 4-week-old SPF white Leghorn chickens were purchased from the Experimental SPF Chicken Farm, Shandong Institute of Poultry Science, Shandong, China. The animals were maintained in isolators before administration. The isolators were supplied with filtered intake and exhaust air.

Virus, vaccines, rice, antibodies and hormones

A virulent IBDV strain ZJ2000 and IBDV strain HZ2 were isolated by this laboratory previously (Li et al., 2004, 2006). A national standard virulent strain BC6/85 for challenge studies was purchased from China National Institute for Supervision of Veterinary Pharmaceuticals, Beijing, China. The commercial attenuated live vaccine IBDV strain B87 was bought from the Heilongjiang Bioproducts Factory, Harbin, China. Rice (Oryza sativa L. ssp. japonica) cultivar Xiushui 11 was used for transformation. IBDV-specific rabbit antiserum and monoclonal antibodies against IBDV VP2 protein were developed by immunizing with killed virus strain BC6/85 in our laboratory (Li et al., 2006). Plants hormones (2,4-dichlorophenoxyacetic acid, 6-benzylaminopurine and α-naphthalene acetic acid) used in transforming were purchased from Sigma (Sigma-Aldrich, St Louis, MO).

Construction of the plant expression vector

The 35S-promoter of cauliflower mosaic virus in the binary vector pCambia1301-35S (kindly provided by Professor Zejian Guo, China Agricultural University) upstream of the multicloning sites between EcoRI and SacI was replaced by the Gt1 promoter (1920 bp) of rice glutelin GluA-2 gene. The GT1 promoter was cloned from the rice genomic DNA of Xiushui 11 by PCR based on the published sequence of promoter of rice glutelin GluA-2 gene (also known as Gt1) with the forward primer Gt1-F (5′-CGGAATTC TTG TTT TTC ACC CTC AAT A-3′, a EcoRI site was underlined) and the reverse primer Gt1-R (5′-ACGAGCTC CAT CGC ACA AGA GGA AC-3′, a SacI site was underlined) (seq. no.: DP000086.1, Takaiwa et al., 1987). The promoter was then substituted into pCambia1301-35S to generate an endosperm-specific binary expression vector, pCambia1301-Gt1. The 1530 bp VP2 gene of the IBDV strain ZJ2000 was isolated by PCR with the forward primer VP2-F (5′-AGGGTACC ATG ACA AAC CTG CAA GAT CAA-3′, a KpnI site was underlined) and the reverse primer VP2-R (5′-ACGTCGAC TTA AGT CAG CTG CCT TAT GCG GCC-3′, a SalI site was underlined) using pGEM-T-VP2 as the template (Li et al., 2003). The amplified cDNA fragment was then subcloned into pCambia1301-Gt1 under the control of the Gt1 promoter, and sequenced to confirm its identity (Figure 1). The expression vector contains genes for hygromycin resistance and kanamycin resistance. The recombinant binary expression vector was transferred into Agrobacterium tumefaciens EHA105 by tri-parental mating as described (Ditta et al., 1980).

Generation of transgenic rice plants

The scutella of mature embryos of seeds from Xiushui 11 were used for callus induction. Mature seeds were dehusked and sterilized with 70% ethanol for 1 min and then with 1.25% (w/v) NaClO for 30 min, washed with sterile water five times and cultured at 26 °C in darkness in NB medium (Zhang et al., 2003) for about 10–15 days until callus was observed. The calli were excised from seeds and subcultured in fresh NB medium for another 4 days. Actively growing calli were used for Agrobacterium-mediated transformation based on the procedures described by Hiei et al. (1994), with NB medium substituted for N6 as the basic culture and hygromycin as selective antibiotic. Regenerated plants (T0 generation) were transferred to soil in pots and grown to maturity in a greenhouse.

PCR and Southern blot analysis of transgenic plants

Rice genomic DNA was extracted from both the transgenic and non-transgenic leaf tissues (3 g) by the CTAB method as described by Ausubel et al. (1995), and analysed for the presence of VP2 gene by PCR with the primers VP2-F and VP2-R. Genomic DNA extracted from the PCR-positive plants (20–30 µg) was completely digested with BamHI, fractionated on a 0.8% agrose gel by electrophoresis at 1.5 V/cm for 16 h, and transferred on to a N+ nylon membrane (Amersham Biosciences, Bucks, UK). Hybridization was conducted as described by Sambrook and Russell (2001) using the 32P-labelled VP2 gene as a probe after labelling with the Prime-Gene DNA Label Kit (Promega, Madison, WI).

Detection of VP2 protein in transgenic rice seeds

A grain of rice seeds was homogenized and extracted in 300 µL of total soluble protein extraction buffer (50 mm Tris-HCl pH 6.8, 4% (w/v) SDS, 8 m urea, 5% (v/v) β-mercaptoethanol, 20% (w/v) glycerol) as described by Takagi et al. (2005). After centrifugation (10 000 g for 10 min at 4 °C), the amounts of soluble protein in the supernatants were estimated by the Bradford protein assay (Bradford, 1976), and the concentration of VP2 protein in the seed of transgenic rice was determined by TAS-ELISA as described (Zhou et al., 2003). A calibration curve is performed with the VP2 protein recovered from purified IBDV (strain BC6/85) particles as the standard sample in TAS-ELISA. Briefly, 96-well plates (NUNC, Roskilde, Denmark) were coated with IBDV specific rabbit antiserum (1 : 500) at 4 °C overnight and a 1 : 30 dilution of the supernatants was added to the plates. The VP2 protein recovered from SDS-PAGE gels of purified IBDV particles (strain BC6/85) was used as the standard sample and a non-transgenic plant supernatant was used as a negative control. The plates were incubated with a 1 : 5000 diluted monoclonal antibody against VP2, then incubated with a 1 : 8000 dilution of alkaline phosphatase-conjugated goat-anti-mouse IgG (Sigma). Finally, P-nitrophenylphosphate substrate (Sigma) was used for colour development, and the optical density values were measured in a Model 680 Microplate Reader (Bio-Rad, Hercules, CA) at 405 nm. Three grains of each VP2 transgenic line were analysed in this ELISA.

The VP2 expression in transgenic rice seeds was also evaluated by Western blot analysis. The protein extraction was equally loaded and separated by 12.5% SDS-PAGE as described by Takagi et al. (2005). After SDS-PAGE the proteins on the gel were electrophoretically transferred to nitrocellulose membranes (Amersham Biosciences). After incubation for 2 h in a blocking buffer (5% skimmed milk power, 0.05% Tween-20 in PBS, pH 7.4) at room temperature, the membrane was probed with antiserum against IBDV (1 : 2500 dilution). The membrane was then incubated with goat-anti-rabbit IgG secondary antibodies conjugated with alkaline phosphatase (1 : 8000, Sigma) and colour was developed in a substrate solution containing NBT/BCIP (Promega).

Chicken immunization, challenge and protection assay

Forty-two 2-week-old SPF chickens were divided into seven equal groups. Before immunization, the chickens were fasted overnight with access to water. Groups 1, 2 and 3 chickens were each fed (orally immunized) 1, 3, or 5 g amounts of seeds from transgenic rice, respectively, while group 4 chickens were orally immunized with 5 g of non-transgenic rice seeds to serve as a negative control. For a positive control, group 5 chickens were immunized intranasally with a commercial attenuated vaccine strain B87 at 2 weeks after hatching and boosted at day 21. As a normal control, group 6 chickens were unvaccinated and unchallenged. Group 7 chickens were unvaccinated but challenged to provide a challenge control. Groups 1, 2, 3 and 4 were fed seeds four times at days 0, 7, 14 and 21. Blood samples were collected from the wing vein at day 0 and at 1 week after each immunization and the sera were used for antibody detection with a serum neutralization and an ELISA test. At 21 days after the last immunization, the chickens, except for group 6, were challenged with 2 × 104 embryo infective dose (EID50)/mL of IBDV strain BC6/85 by the eye-drop method (Li et al., 2006). All chickens except for the death were euthanized at 7 days after challenge. Bursas were isolated and the protection rate for each challenge group was estimated based on gross bursal lesions (Li et al., 2003) and bursal histopathological examinations (Li et al., 2006). Chickens without histological lesions were considered to have been protected against the virulent IBDV challenge.

IBDV antigen detection of bursa of challenged chickens

An immunofluorescence assay was used to detect IBDV antigens in the bursa after virus challenge (Li et al., 2004). Each bursa was frozen and 6-mm thick samples were sectioned in a cryostat. Bursa sections with complete absence of fluorescence were considered to be negative.

ELISA titre detection of the serum antibodies

Antibody titres of chickens vaccinated against IBDV were determined by an ELISA, using purified IBDV particles (strain BC6/85) as a coating antigen and serial twofold dilutions of serum samples (Martinez-Torrecuadrada et al., 2003). Horseradish peroxidase-conjugated rabbit anti chickens IgG (Sigma) was used as the secondary antibody and 3,3,5,5-tetramethyl benzidine (TMB, Sigma) was used as the substrate. The serum titre was expressed as the log10 of the last dilution that yielded an absorption value three times above the pre-immunization serum. The geometric mean titre of the antibody was calculated for each group.

Virus neutralization assay of the sera

Virus neutralization assays for detection IBDV antibodies were performed as described by Li et al. (2004) using chicken embryo fibroblast (CEFs) obtained from SPF embryos as indicator cells. Briefly, serum samples obtained from vaccinated animals were inactivated for 30 min at 56 °C. Serial twofold dilutions of each serum sample were prepared in DMEM (Gibco BRL, Carlsbad, CA) containing 2% fetal calf serum. Each diluted serum was neutralized with 100 TCID50 units of IBDV strain HZ2 for 1 h at 37 °C. The mixture of serum and virus was added to monolayer primary CEFs cultured in 96-well cell plates. The cells were cultured at 37 °C, 5% CO2 until a cytopathic effect was observed (usually by 5 days). The virus neutralization titre was determined from the cytopathological effect using the system described by Li et al. (2003). The log10 of the geometric mean titre of antibody was calculated for each group.

Safety evaluation of transgenic rice seeds expressing IBDV VP2 protein

To confirm the safety of the transgenic rice seeds expressing IBDV VP2 protein, a fivefold of the highest dose (25 g) was orally administered to six 2-week-old chickens each day for a month. The control group was orally administered with non-transgenic rice seeds. The chickens were observed for any disease clinical signs during 2 months inspection.

Acknowledgements

We thank Prof. Andy Jackson (Plant & Microbial Biology Department, University of California, Berkeley) for critical review of the manuscript. This work was supported by the National Basic Research Program of China (2006CB101903) and Zhejiang Provincial Natural Science Foundation (grant no. 301303).

Ancillary