Adaptation of Very Virulent Infectious Bursal Disease Virus to Chicken Embryonic Fibroblasts by Site-Directed Mutagenesis of Residues 279 and 284 of Viral Coat Protein VP2

Abstract

The full-length RNA genomes of a chicken embryonic fibroblast (CEF)-nonpermissive, very virulent infectious bursal disease virus (IBDV) (strain HK46) were amplified into cDNAs by reverse transcription-PCR. The full-length cDNAs were sequenced and subcloned into a eukaryotic expression vector, from which point mutations were introduced into the VP2 region by site-directed mutagenesis. The wild-type and mutated plasmids were transfected directly into CEFs to examine their ability to generate CEF-permissive recombinant viruses. Substitution of amino acid residues 279 (Asp→Asn) and 284 (Ala→Thr) of the VP2 protein yielded a recombinant virus which was able to be passaged in CEFs, whereas the wild-type cDNAs and an amino acid substitution at residue 330 (Ser→Arg) of the VP2 protein alone did not yield viable virus. The results indicated that mutation of other viral proteins, including VP1, VP3, VP4, and VP5, was not required for CEF adaptation of the virus. The same approach may be used to produce CEF-adapted strains from newly evolved IBDVs or to manipulate the antigenicity of the virus.

Infectious bursal disease (IBD), mediated by infectious bursal disease virus (IBDV), causes significant losses to the poultry industry. IBDV multiplies rapidly in developing B lymphocytes in the bursa of Fabricius, leading to immunosuppression and increased susceptibility to other diseases. Two distinct serotypes of IBDV, serotype 1 and serotype 2, have been identified. Serotype 1 strains are pathogenic to chickens and vary in virulence, whereas serotype 2 strains isolated from turkeys are apathogenic for both turkeys and chickens. Serotype 1 strains can be further subdivided into four groups: classical virulent strains, attenuated strains, antigenic variant strains, and very virulent strains. Classical virulent strains cause bursal inflammation and severe lymphoid necrosis in infected chickens, resulting in immunodeficiency and moderate mortality (20 to 30% in specific-pathogen-free [SPF] chickens). Antigenic variant strains are recognized by the ability to escape cross-neutralization by antisera against classical strains. Chickens affected by the variant strains are characterized by severe atrophy of the bursa without showing the inflammation associated with infection by classical strains (13). Attenuated strains are generated by adapting the classical and variant strains to chicken embryo fibroblast (CEF) cells or other cell lines through serial passages. They do not cause diseases in chickens, and therefore some of them are being used as live vaccines. Since the late eighties, outbreaks of newly evolved, very virulent strains in Europe, Japan, and China have caused significant economic losses to the poultry industry. Very virulent strains can break through high levels of maternal antibody and cause up to 60 to 100% mortality in SPF birds. These strains cause lesions typical of IBDV and are antigenically similar to the classical strains (3).

IBDV is a member of the Birnaviridae family, as its genome consists of two segments of double-stranded RNA (dsRNA). Genome segment B (2.8 kb) encodes VP1, a 90-kDa multifunctional protein with polymerase and capping enzyme activities (6). Genome segment A (3.2 kb), encodes a polyprotein that is cleaved by autoproteolysis to form mature viral proteins VP2, VP3, and VP4, of which VP2 is the major host-protective immunogen of IBDV that contains the antigenic regions responsible for induction of neutralizing antibodies (2). A second open reading frame, preceding and partially overlapping the polyprotein gene, encodes VP5, a 17-kDa polypeptide present in IBDV-infected cells. However, the function of the polypeptide is still unknown (7, 19).

Seven IBDV strains isolated from China have recently been characterized (3), including a classical strain, CJ801; an attenuated strain, GZ911; a variant strain, GZ902; and four very virulent strains, G9201, G9302, F9502, and HK46. CJ801 has the greatest identity to the classical strains STC and 52/70, whereas GZ902 has the greatest identity to the American variant strains A, E, and GLS. Attenuated strain GZ911, like other cell culture-adapted strains, has substitutions at positions 279 (D to N) and 284 (A to T) as well as in the serine-rich heptapeptide region. Hence, these substitutions may play an important role in the reduced virulence of these strains. The four very virulent strains have the greatest identity with the European very virulent strain UK661 (2) and Japanese strain OKYM (18). They share unique amino acid residues at positions 222A, 256I, and 294I, which are not present in other, less virulent strains.

A method of generating virus by transfecting cRNA of the CEF-adapted strain D78 into Vero cells has recently been reported (9). However, there is no report on using similar technology to convert a CEF-nonpermissive IBDV strain into a CEF-permissive strain. In this study, we have cloned the full-length cDNA of a CEF-nonpermissive, very virulent IBDV strain in which certain substitutions were introduced into the cDNA by site-directed mutagenesis. By directly transfecting the modified cDNAs, instead of cRNAs, into CEFs, a CEF-permissive recombinant virus was generated from the very virulent IBDV strain.

MATERIALS AND METHODS

Viruses and cells.

CEF cells were prepared from 10-day-old SPF eggs (Beijing Merial Vital Laboratory Animal Technology) and maintained in Dulbecco modified Eagle medium with 5% (vol/vol) fetal bovine serum and 10% (wt/vol) tryptose phosphate broth (17). CEF-adapted strain D78 (ATCC VR-2047), as well as the recombinant viruses generated by reverse genetics, were propagated in secondary CEF. The classical strains CJ801 and G9219 and the very virulent strains HK46, G9201, G9303, and F9502 were propagated in SPF chickens at 5 weeks of age.

Purification of IBDV from bursa and extraction of viral RNA.

Bursas from SPF birds infected by HK46 were removed and homogenized in TNE buffer (10 mM Tris-HCl, 100 mM NaCl, 1 mM EDTA [pH 8.0]) at a ratio of 1 g of bursa to 10 ml of TNE buffer. After freezing and thawing three times, homogenates were centrifuged at 17,000 × g for 15 min at 4°C, and the supernatant was collected for virus purification. One milliliter of bursal homogenate supernatant was loaded onto 4 ml of a 40% sucrose cushion and ultracentrifuged at 55,000 × g (Beckman sw55Ti rotor, 22,000 rpm) for 2.5 h at 4°C. After removal of the cushion, pelleted IBDV particles were resuspended in proteinase K buffer (TNE plus 0.5% [wt/vol] sodium dodecyl sulfate), digested with 1 mg of proteinase K per ml for 1 h at 37°C, and extracted twice with phenol-chloroform-isoamylalcohol (25:24:1, vol/vol/vol). Finally, the viral dsRNAs were precipitated from the upper aqueous phase by ethanol.

Construction of full-length cDNA clones.

The cDNAs of IBDV segments A and B of the very virulent strain HK46 were synthesized separately. Two primer pairs (Table 1), designated A5-A3 and B5-B3, were used for synthesizing segment A and B cDNAs, respectively. Viral dsRNAs mixed with primers A5 and A3 (for segment A) or B5 and B3 (for segment B) were denatured by boiling for 5 min and cooled on ice for 2 min. First-strand cDNAs were synthesized as described previously (3). All primers used for PCR and mutagenesis are listed in Table 1. Four fragments, designated FA5 (5′ end fragment of segment A), FA3 (3′ end fragment of segment A), FB5 (5′ end fragment of segment B), and FB3 (3′ end fragment of segment B), were independently amplified by the Expand High Fidelity PCR system (Boehringer Mannheim) by using the first-strand cDNA as a template. The amplifications were performed with Robocycler Gradient 96 (Stratagene) in a program of 94°C for 3 min, 30 cycles of 94°C for 40 s and 60°C (for FA5 and FA3 amplification) or 52°C (for FB5 and FB3 amplification) for 40 s, 72°C for 2 min 30 s, and finally 72°C for 10 min. Four primer pairs were used for amplifications of the corresponding fragments, of which A5 and A3AP were used for amplification of fragment FA5, A5SP and A3 were used for fragment FA3, B5 and B3AP were used for fragment FB5, and B5SP and B3 were used for fragment FB3. Fragment FA5 digested with EcoRI and SalI was cloned into the EcoRI/SalI site of pBssK vector (Stratagene) to obtain plasmid FA5-pBssK. Subsequently, PCR fragment FA3 was subcloned into the SalI and KpnI sites of plasmid FA5-pBssK to generate FA-pBssK, which carried the full-length fragment A cDNA (Fig. 1). To obtain a cDNA clone of segment B of HK46, two cDNA fragments, FB5 and FB3, were amplified by PCR with two primer pairs, B5-B3AP and B5SP-B3, respectively, by using first-strand cDNA of segment B as a template. There was a unique BglII site in the overlapping region of fragments FB5 and FB3. Because vector pBssK lacks a BglII site, another plasmid, pBssK-Bgl, in which a BglII linker was cloned into the EcoRI site of pBssK, was employed to construct the full-length cDNA clone of fragment B. Fragment FB3 digested with BglII and XbaI was cloned into the BglII/XbaI site of plasmid pBssK-Bgl to obtain plasmid FB3-pBssK. Subsequently, PCR fragment FB5 was subcloned into the EcoRI and BglII sites of FB3-pBssK to yield a plasmid containing a full-length cDNA copy of segment B (Fig. 1). DNA sequences were determined by primer walking with a dRhodamine Terminator Cycle Sequencing Kit (Perkin-Elmer) in an automatic sequencer (A310 Genetic Analyzer; Perkin-Elmer).

TABLE 1.

Primers for genome cloning and mutagenesis of IBDV^a

Primer	Size (bp)	RE site	Direction	Location	Sequence
A5	40	EcoRI	Sense	A, 5′ end	5′ATGAATTCAGGATACGATCGGTCTGACCCCGGGGGAGTCA3′
A3	33	KpnI	Antisense	A, 3′ end	5′TAGGTACCAAGGGACCCGCGAACGGATCCAATT3′
B5	29	EcoRI	Sense	B, 5′ end	5′TTAGAATTCTAGGATACGATGGGTCTGAC3′
B3	29	XbaI	Antisense	B, 3′ end	5′ATTTCTAGATGGGGGCCCCCGCAGGCGAA3′
A3AP	27	SalI	Antisense	A, 1723–1749	5′CAGGTGAAGCARAGAATCCCGTCGACTA3′
A5SP	28	SalI	Sense	A, 1712–1739	5′CCAGAATCCYGTAGTCGACGGGATTCTT3′
B3AP	21	BglII	Antisense	B, 1842–1862	5′GATCCCRAGATCTTTGCTGTA3′
B5SP	24		Sense	B, 1751–1774	5′CCTTGCACAACCAGGGTACCTGAG3′
NTA	34		Antisense	A, 950–983	5′GGTCGTTAGCCCATTGTTTCGGGCCACAGCTCTG3"
NTS	38		Sense	A, 963–1000	5′GCAAACAATGGGCGAACGACCGGCACTGACAACCTTAT3′
330R	27		Antisense	A, 1109–1135	5′pACTGCTAGGCTCCCTCTTGCTGACCAT3′

^aPrimers A5, A3, B5, and B3 were designed based on sequences reported in reference 9. Primers A3AP, A5SP, B3AP, and B5SP were designed by alignment of multiple IBDV sequences available in GenBank. RE, restriction enzyme; R, A or G; Y, C or T. 

FIG. 1.

Construction of full-length cDNA clones. Four cDNA fragments derived from very virulent strain HK46, designated FA5 (5′ end fragment of segment A), FA3 (3′ end fragment of segment A), FB5 (5′ end fragment of segment B), and FB3 (3′ end fragment of segment B), were independently amplified by four primer pairs. Fragment FA5 digested with EcoRI and SalI was cloned into the EcoRI/SalI site of pBssK to obtain plasmid FA5-pBssK. Subsequently, fragment FA3 was subcloned into the SalI and KpnI sites of plasmid FA5-pBssK to generate FA-pBssk, which carried the full-length fragment A cDNA. Fragment FB3 digested with BglII and XbaI was cloned into the BglII/XbaI site of plasmid pBssK-Bgl to obtain plasmid FB3-pBssK. Subsequently, fragment FB5 was subcloned into the EcoRI and BglII sites of FB3-pBssK to yield a plasmid containing a full-length cDNA copy of segment B.

Site-directed mutagenesis by PCR.

Two combinations of mutations were introduced into the segment A cDNA of strain HK46. (i) In mutant NT, substitutions at nucleotide positions 966 (G→A) and 981 (G→A) resulted in amino acid substitutions at residues 279 (Asp→Asn) and 284 (Ala→Thr). (ii) In mutant R, a nucleotide substitution at position 1121 (T→A) resulted in amino acid substitution at position 330 (Ser→Arg). To construct mutant plasmid NT-FA-pBssK, two primer pairs, designated A5-NTA and NTS-A3AP, were used to amplify two DNA fragments, of 991 and 786 bp, respectively. These two DNA fragments were purified, combined, and reamplified by PCR with primers A5 and A3AP. Pfu polymerase (Stratagene) was used in PCR to enhance the fidelity. After amplification, the PCR product of 1,756 bp was then digested with EcoRI and SalI. The resulting fragment of 1,726 bp was cloned into the EcoRI/SalI site of FA-pBssK to replace the EcoRI/SalI insert of plasmid FA-pBssK to obtain plasmid NT-FA-pBssK. The mutated sequences were confirmed by cycle PCR sequencing as described above. Plasmid NT-FA-pBssK contained full-length cDNA of segment A of strain HK46 but had substitutions at nucleotide positions 966 (G→A) and 981 (G→A). In addition, the identities of the mutant plasmids were further confirmed by the disappearance of a NaeI restriction site after mutagenesis, since the wild type was NaeI restricted (Fig. 2).

FIG. 2.

Schematic presentation of pALTER expression plasmids containing genome segment A derived from IBDV strain HK46. Plasmid FA-pALTER contains the wild-type cDNA without modification. In plasmid R-FA-pALTER, a nucleotide substitution at position 1121 (T→A) resulted in an amino acid (a.a.) substitution at position 330 (Ser→Arg). In plasmid NT-FA-pALTER, substitutions at nucleotide positions 966 (G→A) and 981 (G→A) resulted in amino acid substitutions at residues 279 (Asp→Asn) and 284 (Ala→Thr). The nucleotide substitution at position 981 (G→A) also eliminated a unique restriction site on the segment A cDNA, NaeI. All plasmids contained cytomegalovirus (CMV) promoter sequences at their 5′ ends, which could drive transcription of the segment A genome after transfection into CEFs.

Subcloning of the cDNA of both segments of IBDV into vector pALTER-MAX.

Carrying the genes for ampicillin and chloramphenicol resistance and the cytomegalovirus immediate-early enhancer/promoter, the pALTER-MAX vector is used for mutagenesis and expression of genes in eukaryotic cells (Promega). To subclone wild-type segment A cDNA and NT mutant segment A cDNA into the pALTER-MAX vector, plasmids FA-pBssK and NT-FA-pBssK were digested with EcoRI and KpnI, and restriction fragments FA and NT-FA were cloned into the EcoRI/KpnI sites of pALTER-MAX to create plasmids FA-pALTER and NT-FA-pALTER, respectively (Fig. 2). These plasmids contained a full-length cDNA copy or a mutated segment A of strain HK46. To clone fragment FB into pALTER-MAX, plasmid FB-pBssK was digested with restriction enzymes EcoRI and XbaI. Subsequently, fragment FB was inserted into the EcoRI/XbaI site of pALTER-MAX to obtain plasmid FB-pALTER. This plasmid contained the full-length cDNA of segment B of strain HK46.

Mutation of the serine-rich heptapeptide region on VP2 by the Altered Sites II mutagenesis system.

Single-strand DNA of plasmid FA-pALTER was produced by the method recommended by the supplier (Promega). A primer designated 330R, phosphorylated at the 5′ end, was used for producing a mutation on plasmid FA-pALTER, in which nucleotide 1127 (T) was replaced with A. After annealing of primer 330R and the ampicillin repair oligonucleotide to the single-strand template, T4 DNA polymerase and T4 DNA ligase were added to perform mutant-strand synthesis and ligation. After incubation, the synthesis mixture was transformed into Escherichia coli ES1301 mutS competent cells by electroporation. The transformed ES1301 mutS cells were cultured at 37°C and shaken at 250 rpm overnight. Plasmid DNA was then extracted and then retransformed into E. coli JM109 competent cells. To screen for the mutant plasmid designated R-FA-pALTER, 10 colonies were screened by direct sequencing with an automatic sequencer (Fig. 2).

Generation of recombinant viruses by transfection.

CEFs were grown to 80% confluence in a 12-well tissue culture plate and washed once with phosphate-buffered saline (PBS). One milliliter of serum-free EME medium (SF-EMEM) (Life Technologies) was added to each well, and the cells were incubated at 37°C for 1 h in a 5% CO₂ incubator. Simultaneously, 100 μl of SF-EMEM was incubated with 10 μl of Lipofectin (Life Technologies) for 30 min at room temperature. After a 30-min incubation at room temperature, 100 μl of SF-EMEM with 5 μl (0.2 μg/μl) of pALTER plasmids FA-pALTER and FB-pALTER was added to the SF-EMEM–Lipofectin mixture and incubated at room temperature for 5 min. After removing SF-EMEM from the CEF monolayer and replacing it with 0.8 ml of fresh SF-EMEM, the plasmid-containing mixture was added to CEF monolayers. After 5 h of incubation at 37°C, the medium was replaced with EMEM containing 10% fetal bovine serum and the cells were further incubated at 37°C in a 5% CO₂ incubator for several days. The modified pALTER plasmids (NT-FA-pALTER or R-FA-pALTER) and/or FB-pALTER were also transfected into CEFs by the procedure described above. After 96 h, cytopathic effect was observed. The ability of the recombinant viruses to propagate in CEFs was further tested by adding 0.1 ml of virus suspension collected 96 h posttransfection into a well of a 12-well plate containing 80% confluent CEFs with 1 ml of cell culture medium. After virus propagation in CEFs, the mutant virus was harvested and the first-strand cDNA was generated as described above. The nucleotide sequence of the recombinant virus was confirmed by DNA sequencing.

Reactivity with MAbs.

Five monoclonal antibodies (MAbs) (3-1, 9-6, 17-82, 39A, and 44-18) raised against virulent strain 002/73 of IBDV by Fahey et al. (4) and three MAbs (B29, R63, and R69) raised against vaccine strain D78 by Snyder et al. (11) were used in antigen capture enzyme-linked immunosorbent assays (ELISA) to examine their immunoreactivities to several CEF-nonpermissive IBDV strains isolated from Hong Kong and China (3) and to two CEF-adapted strains, including the NT mutant derived from HK46 (HK46-NT) and strain D78 (ATCC VR-2041).

One hundred microliters of rabbit anti-IBDV immunoglobulin G (IgG) diluted in coating buffer (15 mM Na₂CO₃, 35 mM NaHCO₃, 0.05% [wt/vol] NaN₃ [pH 9.6]) at a ratio of 1:500, was added to each well of a Maxisorb microtiter plate (Nunc). After incubation at 4°C overnight, the plate was washed three times with washing buffer TBS/NT (20 mM Tris-HCl, 140 mM NaCl, 0.05% NaN₃, 0.05% (vol/vol) Tween 20 [pH 7.4]) and each well was blocked by 200 μl of blocking buffer (TBS/NT with 10 mg of bovine serum albumin) at room temperature for 1 h. After three washes of the plate with washing buffer, 100 μl of bursal homogenate (10% [wt/vol] in PBS) from infected SPF chickens or undiluted virus culture media was added in duplicate. The plate was then incubated at room temperature for 1 h and washed with washing buffer before 100 μl of MAbs diluted 1:10 in diluent (washing buffer with 2% [wt/vol] nonfat skim milk) were added to the wells in duplicate. After incubation for 1 h at room temperature, the plate was washed three times with washing buffer. Subsequently, 100 μl of rabbit anti-mouse IgG-horseradish peroxidase (Dako) diluted in diluent at a ratio of 1:1,000 was added. One hour later, the plate was washed three times with washing buffer. TMB peroxidase substrate (100 μl) (Kirkegaard & Perry Laboratories) was added, and the reaction was stopped 10 min later by the addition of 50 μl of TMB stop solution (Kirkegaard & Perry Laboratories). The result was read by an ELISA reader at the optical density at 450 nm (OD₄₅₀).

Infectivity of recombinant virus HK46-NT on CEF cells.

The titer of the amplified virus stock was determined by serial dilution and expressed as 50% tissue culture infective doses (TCID₅₀) per milliliter. CEF cells were seeded in 96-well plates at a density of 2 × 10⁴ cells/well, and 100 TCID₅₀ of D78 or HK46-NT were added to each well in a 0.1-ml volume. At daily intervals (up to 6 days), 20 μl of MTS/PES [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium/phenazine ethosulfate] solution (CellTiter 96AQ Cell proliferation assay; Promega) was added to each well and the plates were incubated at 37°C for 2 h in a humidified chamber with 5% CO₂. The OD₄₉₀ was then measured by an ELISA reader (SLT Spectra). Wells with 200 μl of culture medium only, without the addition of CEF cells, were measured as background signal.

Growth curve of IBDV.

The growth characteristics of strains HK46-NT and D78 were compared with each other. CEF cells were seeded in 96-well plates at a density of 2 × 10⁴ cells/well, and 100 TCID₅₀ of D78 or HK46-NT were added to each well in triplicate. At 24-h intervals, the culture supernatant in each well was collected and the virus titers were determined and expressed as TCID₅₀ per milliliter.

Nucleotide sequence accession numbers.

The full-length sequence of the genome of IBDV strain HK46 has been deposited in GenBank under accession no. AF092943 (segment A) and AF092944 (segment B).

RESULTS

Full-length cDNA sequence of the IBDV genome.

The full-length sequences of the segment A and B genomes of the very virulent IBDV strain HK46 are shown in Fig. 3 and 4, respectively. The segment A genome contains 3,269 bp and encodes two polypeptides in two reading frames. Nucleotides 86 to 532 encode a 149-residue protein, VP5, whereas nucleotides 132 to 3167 encode a VP2-4-3 polypeptide of 1,012 residues. The segment B genome contains 2,834 bp and encodes the RNA polymerase (VP1) of 879 residues. The nucleotide sequence of VP2 has eight mutations, at nucleotides 608 (T→C), 809 (A→G), 951 (T→C), 1016 (T→C), 1154 (T→A), 1163 (T→C), 1217 (A→G), and 1274 (T→A), compared to the sequence of the VP2 hypervariable region reported previously (3). These mutations did not result in any amino acid substitutions, indicating a mixed virus population in the infected bursa.

FIG. 3.

Full-length cDNA and amino acid sequences of the segment A genome of very virulent IBDV strain HK46. Cloning sites EcoRI (−6 to −1), SalI (1725 to 1730), and KpnI (3264 to 3269) are underlined. The stop codon is denoted by an asterisk.

FIG. 4.

Full-length cDNA and amino acid sequences of the segment B genome of very virulent IBDV strain HK46. Cloning sites EcoRI (−6 to −1), BglII (1850 to 1855), and XbaI (2829 to 2834) are underlined. The stop codon is denoted by an asterisk.

The full-length amino acid sequences of HK46 and two very virulent IBDV strains, OKYM, isolated in Japan (18), and UK661, isolated in England (2), were identical in their VP2 sequences, whereas there were certain differences in the VP1, VP3, and VP4 amino acid sequences (Table 2). The VP2-4-3 polypeptide of HK46 had three and seven amino acid substitutions compared with strains OKYM and UK661, respectively. On the other hand, the VP1 sequence of HK46 had 11 and 10 amino acid substitutions compared with strains OKYM and UK661 and appears to be an intermediate strain between them, since they had differences at 17 amino acid residues.

TABLE 2.

Differences in amino acids of three very virulent IBDV isolates

Position	Amino acid
	HK46	OKYM	UK661
Segment Aa
485	Thr	Ala	Ala
624	Asp	Asp	Val
632	Asp	Asp	Gly
651	Asn	Asn	Ser
654	Thr	Asn	Asn
849	Glu	Glu	Gly
919	Glu	Glu	Gly
951	Val	Ile	Val
Segment Bb
12	Ser	Gln	Ser
24	Val	Ala	Ala
69	Glu	Asp	Glu
91	Ala	Ala	Pro
125	Pro	Ser	Pro
330	Trp	Trp	Cys
331	Leu	Leu	Cys
337	Ile	Ile	Lys
341	Pro	Pro	Gln
343	Pro	Pro	Ser
380	Gly	Val	Gly
426	Arg	Pro	Arg
497	Gly	Ser	Ser
675	Ala	Ser	Ala
677	Glu	Val	Glu
697	Val	Ala	Val
751	Lys	Lys	Met
825	Asn	Asn	His
831	His	Arg	His

^aThere are 3, 7, and 6 amino acid residue differences between isolates HK46 and OKYM, HK46 and UK661, and OKYM and UK661, respectively, in segment A. 

^bThere are 11, 10, and 17 amino acid residue differences between isolates HK46 and OKYM, HK46 and UK661, and OKYM and UK661, respectively, in segment B. 

Generation of recombinant viruses by transfection.

The pALTER plasmid containing the wild-type genome fragment A cDNAs, or the pALTER plasmids containing the mutated fragment A cDNAs, were cotransfected with the pALTER plasmid carrying the wild-type genome of fragment B. Cytopathic effects could be seen only in wells transfected with the NT mutant plasmid after a 4-day culture, not in the wells transfected with wild-type or 330R mutant plasmids even after a 7-day culture. The culture supernatants collected at day 7 from these wells were blind passaged in secondary CEFs; however, no virus was propagated. The experiment was repeated three times to confirm the result. The titer of the NT mutant virus was 10^4.3 TCID₅₀/ml at day 7 posttransfection and reached 10⁶ TCID₅₀/ml at 3 days after the first passage. Viral RNA was recovered from the recombinant virus, and the full-length sequences of both genome segments were amplified and sequenced. The sequencing result indicated that a recombinant virus mutant with 222A, 256I, and 294I, the signature amino acid residues of very virulent IBDV strains, and 279N and 284T, the signature amino acid residues of attenuated strains was produced (Fig. 5). This recombinant virus has the conserved serine-rich heptapeptide S-W-S-A-S-G-S (sequencing data not shown).

FIG. 5.

Confirmation of 279 Asn and 284 Thr substitutions in recombinant virus HK46-NT by DNA sequencing. The VP2 hypervariable region of recombinant virus HK46-NT was recovered by reverse transcription-PCR and sequenced. The sequencing result was compared with the wild-type sequence in plasmid FA-pALTER. Arrows indicate substitutions at nucleotide positions 966 (G→A) and 981 (G→A), which resulted in amino acid substitutions at residues 279 (Asp→Asn) and 284 (Ala→Thr). The amino acid residue in red (294I) is unique to very virulent IBDV.

Reactivity of MAbs.

MAbs B29, R63, 9-6, 17-82, and 39A showed activities toward all strains examined, whereas MAb B69 recognized strain D78 but not any other local strain (Table 3). MAb 44-18 showed strong binding to the very virulent IBDV strains HK46, G9201, G9303, and F9502, as well as CEF-adapted strains D78 and the HK46-NT mutant, but had only weak activity toward classical strains CJ801 and G9212 (Table 3). MAb 3-1 recognized all very virulent strains and classical strains but not CEF-adapted vaccine strain D78 and the HK46-NT mutant (Table 3). Since these two CEF-adapted strains had common amino acid residues, 279N and 284T, which were not present in non-CEF-adapted strains, these amino acid residues may affect the binding of MAb 3-1. To further differentiate the HK46-NT strain from the other CEF-adapted strains derived from other parent strains, MAb B69 can be used. Many CEF-adapted vaccine strains, including D78 (Intervet), IBD Blend (Sanofi Inc.), Bio-Burs (Kee-Vet), Bio-Burs I (Kee-Vet), Bursa-vac (Sterwin), Bursine (Solvary Inc.), Bur-706 (Rhone Merieux), Univax (American Sci. Lab.), and VI-Bur-G (Vineland), are B69 positive, whereas the HK46-NT strain, like its parent strain, is B69 negative (14).

TABLE 3.

Reactivities of various MAbs to different IBDV strains

Straina	Reactivity of MAb:
	B29	B69	R63	3-1	9-6	17-82	39A	44-18
D78 (CEF)	+	+	+	−	+	+	+	+
NT mutant (CEF)	+	−	+	−	+	+	+	+
HK46	+	−	+	+	+	+	+	+
G9303	+	−	+	+	+	+	+	+
F9502	+	−	+	+	+	+	+	+
G9201	+	−	+	+	+	+	+	+
CJ801	+	−	+	+	+	+	+	−
G9212	+	−	+	+	+	+	+	−

^aFor strains HK46, G9303, F9502, G9201, CJ801, and G9212, bursal homogenates from infected SPF birds were used as the source of antigen. 

Infectivity of the recombinant virus HK46-NT on CEF cells.

Cytopathogenicity of the recombinant virus HK46-NT was compared with that of D78 in CEF cells by using the MTS assay. Two individual experiments were carried out, and the results were reproducible. Figure 6 shows the relative numbers of viable CEF cells at different days postinfection after they were incubated with 100 TCID₅₀ of virus. The signal at OD₄₉₀ was directly proportional to the number of metabolically active cells. Without addition of virus, the number of viable cells in the control wells increased in the first few days and rapidly dropped to a very low level at day 5, indicating that the cells were overgrown and had died. In both experiments, the cells were lysed more rapidly by strain HK46-NT than by strain D78, implying that HK46-NT was more cytopathogenic than D78 to CEF cells.

FIG. 6.

Cytopathogenicity of recombinant virus on CEF cells. CEF cells (2 × 10⁴ in each well) were mixed with 100 TCID₅₀ of D78 or HK46-NT. At daily intervals (up to 5 days), 20 μl of MTS/PES solution was added to each well and the plates were incubated at 37°C for 2 h in a humidified chamber with 5% CO₂. The plates were then measured at OD₄₉₀. The background reading, measured from wells with culture medium and MTS/PES only, was subtracted from the data. Each value is the average of two independent experiments.

Growth curve of IBDV.

To compare the replication kinetics of strains HK46-NT and D78, CEF cells were infected with each virus and the virus titers were determined and expressed as TCID₅₀ per milliliter. Figure 7 shows the growth curve of each virus at different days postinfection. The results indicate that the mutant virus replicated at a higher speed and generated virus to a higher titer than did strain D78.

FIG. 7.

Growth curve of the mutant virus. CEF cells were infected with 100 TCID₅₀ of D78 or HK46-NT in 200 μl of culture medium. Virus harvested at daily intervals was then titered and expressed as TCID₅₀ per milliliter. Each value is the average of two independent experiments.

DISCUSSION

Heine et al. (5) aligned the VP2 amino acid sequences of virulent strains (52/70, STC, and variant E) with CEF-adapted strains of low pathogenicity (002-73, CU-1, and PBG98) and identified the heptapeptide S-W-S-A-S-G-S (residues 326 to 332) as the only conserved sequence in strains of high pathogenicity. It was hypothesized that the serine-rich motif could form hydrogen bonds which might be involved in intra- or intermolecular interactions important for virulence (5). It was also suggested that the substitution of one or more serine residues in the CEF-adapted, nonpathogenic strains might prevent such interaction and affect the virulence. In contrast, Yamaguchi et al. (17) have aligned the segment A sequence of a highly virulent strain, OKYM, and its CEF-adapted, attenuated strain, OKYMT, and discovered that both of them had the conserved heptapeptide S-W-S-A-S-G-S. There were five amino acid differences between the two Japanese strains, of which substitutions at positions 279 (Asp to Asn) and 284 (Ala to Thr) were commonly found in CEF-adapted strains.

Our results indicated that the substitution of amino acid residues at positions 279 (Asp to Asn) and 284 (Ala to Thr) of the VP2 coat protein were adequate to convert a CEF-nonpermissive, very virulent strain, HK46, to a CEF-adapted strain. In contrast, a substitution at position 330 (Ser to Arg) could not convert a CEF-nonpermissive strain to a permissive strain. Additionally, since the CEF-adapted strain HK46-NT shared identical VP1, VP3, VP4, and VP5 sequences with the wild-type CEF-nonpermissive parent strain, mutation of these viral proteins was not required for CEF adaptation. The acquisition of 279N and 284T may enable the virus to acquire a novel receptor binding site on CEF cells (10).

The principal method of controlling IBD in young chickens is by vaccination with a live attenuated strain of IBDV at the age of 0 to 5 weeks or by transferring high levels of maternal antibody induced by the administration of live and killed IBD vaccines to breeder hens (15, 16). The live attenuated vaccines on the market are generally derived by serial passage of classical or variant strains in embryonating SPF eggs or in primary cell culture, such as CEFs. Only strains with their virulence reduced or eliminated by these methods can be used as live attenuated vaccines. Inactivated vaccines are prepared by propagating the virus in cell culture or in SPF chickens, and the vaccines are subsequently inactivated by heat or chemical treatment. The attenuated strains derived from the classical and variant strains can usually protect birds from infection by their parent strains. However, many of these vaccine strains do not offer 100% protection against the very virulent strains discovered in recent years (12). Since very virulent strains could break through higher levels of maternal antibodies than are induced by the current vaccine strains (12), it is important to develop attenuated vaccine strains from very virulent strains.

Presently, live infectious bursal disease vaccines are derived from multiple passages of the virus in CEFs or in embryonating eggs. Hence, the method of producing live attenuated virus is a random, uncontrolled process; as a result, these vaccines suffer from a number of drawbacks. First, a field virus may not be adapted to CEFs after many attempts, and hence no live attenuated virus may be derived. Second, even if some of the field viruses can be adapted in CEFs, a population of viruses of different characteristics and different degrees of virulence are generated. Because uncontrolled mutations are introduced into the viral genome during serial passages, a population of virus particles heterogeneous in their virulence and immunizing properties is generated. Prior to vaccine development, a single virus has to be cloned and the pathogenicity of each cloned virus has to be tested, which is a lengthy and labor-intensive process.

Techniques for controlled manipulation of the IBDV genome developed in this project may allow the generation of CEF-adapted viruses and avoid the disadvantages of the conventional virus passage approach. The process is straightforward, controllable, and can be applied to any newly evolved strain. The genome of the generated virus is known and no virus cloning is required. To determine the potential for applying the NT mutant virus as a live vaccine against very virulent IBDV, the pathogenicity of the mutant virus must be tested in vivo on SPF chickens. Recently, Yao et al. (19) produced a recombinant virus (rD78NSΔIBDV) lacking VP5 expression by using infectious cRNA, which was derived from the CEF-adapted strain D78. The VP5-deficient mutant had a lower cytotoxic effect on CEF cells and grew to lower titers than the parent strain. The virus also failed to induce any pathological lesions or clinical signs of disease in SPF chickens. It will be interesting to produce a VP5-deficient mutant from HK46-NT to test which of the mutants is a better candidate for vaccination against very virulent IBDV strains.

In summary, this is the first report of a method for producing a recombinant CEF-adapted IBDV strain from a CEF-nonpermissive strain by direct transfection of engineered vector DNA into CEFs. The methods reported recently (8, 9, 19) involve more tedious procedures, since generation of cRNA from a T7 promoter and subsequent transfection of the cRNA into CEFs were required. More importantly, the cRNAs produced by these groups were derived from D78, a CEF-permissive strain by nature. The vectors FA-pALTER and FB-pALTER could be employed as templates for developing CEF-adapted strains from any newly evolved IBDV strain, including very virulent, variant, or classical strains, by first subcloning a partial VP2 cDNA generated from these strains by reverse transcription-PCR into the NdeI (nucleotides 646 to 651) and SpeI (nucleotides 1181 to 1186) sites on VP2 cDNA in FA-pALTER and subsequently introducing 279N and 284T into the VP2 cDNA by mutagenesis. Hence, the antigenicity and pathogenicity of the virus can be manipulated.