Download Tissue culture infectivity of different strains of infectious bursal

Journal
of General Virology (1999), 80, 2067–2076. Printed in Great Britain
...................................................................................................................................................................................................................................................................................
Tissue culture infectivity of different strains of infectious bursal
disease virus is determined by distinct amino acids in VP2
Egbert Mundt
Institute of Molecular Biology, Friedrich-Loeffler-Institutes, Federal Research Centre for Virus Diseases of Animals, D-17498 Insel Riems,
Germany
Two types of strains of serotype I of infectious bursal disease virus (IBDV) have been described, on
the basis of their ability (IBDV-TC) or inability (IBDV-BU) to infect chicken embryonic cells in culture.
However, both types infect B lymphocytes in the bursa of Fabricius of young chickens. To determine
the molecular basis for tissue culture infectivity, virus recombinants with chimeric segments A were
constructed from IBDV-TC and IBDV-BU by reverse genetics. The region responsible for the
different phenotypes was located in VP2. Site-directed mutagenesis identified single amino acids
that are responsible for the restriction in infectivity. However, the appropriate amino acid
exchanges are strain-specific.
Introduction
Infectious bursal disease in chickens was first described by
Cosgrove (1962). The causative agent of this highly contagious
and immunosuppressive disease is infectious bursal disease
virus (IBDV). Serotype I viruses are pathogenic for chickens
but individual strains differ markedly in their virulence.
Serotype II strains, isolated from fowl, turkeys and ducks
(McFerran et al., 1980), are apathogenic for chickens. Both
serotypes can be differentiated by cross-neutralization assays
(McFerran et al., 1980).
IBDV belongs to the genus Avibirnavirus of the family
Birnaviridae (Dobos et al., 1995). The genome consists of two
segments, A and B, of double-stranded RNA, which are
localized within a single-shelled icosahedral capsid of 60 nm
diameter. Recently, the complete genomic sequences of both
segments of three serotype I strains and one serotype II strain
of IBDV were determined (Mundt & Mu$ ller, 1995). The larger
segment, A, encodes a polyprotein of approximately 110 kDa
that is autoproteolytically cleaved (Hudson et al., 1986) to form
the virus proteins VP2, VP3 and VP4. A second open reading
frame (ORF), preceding and partially overlapping the polyprotein gene, has been identified (Bayliss et al., 1990 ; Spies et
al., 1989). A protein encoded by this ORF, designated VP5, has
been detected in IBDV-infected chicken embryo cells (CEC) as
well as in bursal cells of IBDV-infected chickens (Mundt et al.,
1995). Comparison of segment A sequences of different
serotype I strains showed an overall high homology. However,
Author for correspondence : Egbert Mundt.
Fax j49 38351 7151. e-mail Egbert.Mundt!rie.bfav.de
0001-6330 # 1999 SGM
a variable region is localized in VP2 (Bayliss et al., 1990).
Genome segment B encodes a 97 kDa protein, designated
VP1, that represents the putative viral RNA-dependent RNA
polymerase (Spies et al., 1987).
Serotype I field strains can cause serious problems in the
poultry industry. Several serotype I strains have been adapted
to tissue culture, Cu-1 (Nick et al., 1976), PGB98 (Baxendale,
1976), P2 (Schobries et al., 1977), OKYMT, TSKMT
(Yamaguchi et al., 1996), STC and IN (Hassan et al., 1996). After
adaptation, these strains showed reduced in vivo pathogenicity
(Cursiefen et al., 1979 ; Lange et al., 1987 ; Yamaguchi et al.,
1996 ; Hassan et al., 1996). Usually, pathogenic field strains
(bursa derived) are not easily adapted to cell culture, a process
which requires extensive passaging either in cell culture
(Hassan et al., 1996) or in the chorioallantoic membrane (CAM)
as well as in the yolk sac of embryonated eggs (Yamaguchi et
al., 1996). Several field isolates failed to become adapted to cell
culture (McFerran et al., 1980).
The aim of this study was to determine specific structural
requirements for the ability to infect cultured cells. Conversion
of IBDV strains unable to grow in cultured cells into cell
culture-infectious viruses by the use of the established reverse
genetics system of IBDV (Mundt & Vakharia, 1996) is
described. Furthermore, the application of the reverse genetics
system to generation of IBDV segment A chimeras and their
functional analysis is described.
Methods
Virus and cells. The serotype I strain D78 (Intervet, Boxmeer,
Netherlands) was propagated in CEC. The serotype I strain GLS,
obtained from the bursa of Fabricius of infected chickens, (GLS-BU) and
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 22:32:42
CAGH
E. Mundt
p10-22
pD78A-E/DeL
T7 promoter
pD78A
T7 promoter
pUC19/∆NH-D78A
pD78A-E/Del-GLS
T7 promoter
Fig. 1. Construction of chimeric cDNA clones of segment A of IBDV. A map of the genomic organization of IBDV segment A is
shown at the top of the figure. Coding sequences of segment A of strain D78 are depicted by an open box. E/Del sequences
are marked by shaded boxes and sequences of strain GLS are shown as a black box. Restriction enzymes and their cleavage
sites used for cloning are indicated. Full-length cDNAs were constructed under the control of the T7 RNA polymerase promoter.
Locations of mutated amino acids are indicated and named by the single letter code. Numbering of nucleotides and amino acids
are according to the published sequence of strain P2 (Mundt & Mu$ ller, 1995).
the tissue culture-adapted form (GLS-TC) were a generous gift from Dr
A. van Loon (Intervet). CEC derived from embryonated specific
pathogen-free (SPF) eggs (VALO, Lohmann, Cuxhaven, Germany) were
grown in Dulbecco’s minimal essential medium (DMEM) supplemented
with 10 % foetal calf serum (FCS) and were used for transfection,
propagation of recovered virus, passaging of transfection supernatants
and immunofluorescence assays (IFA). Transfection experiments and IFA
were performed by using quail muscle cells (QM-7 ; ATCC no. CRL
1962) grown in medium 199 supplemented with 10 % FCS.
Analysis of sequences of segment A of IBDV. To determine the
amino acids that are responsible for the ability of IBDV to infect CEC,
sequences of strains that are able to infect CEC were compared with
sequences of strains that are unable to do so. Amino acid sequences (aa
251–360), comprising the variable regions of sequences of VP2 from
tissue culture-infecting strains published by Yamaguchi et al. (1996)
(strains J1, accession no. D16677 ; K, D16678 ; OKYMT, D83985 ; and
TKSMT, D84071), Bayliss et al. (1990) (strains PGB98, D00868 ; and Cu1, D00867) and Mundt & Mu$ ller (1995) (strain P2, X84034) and from
strain D78 (V. N. Vakharia, personal communication), were compared to
obtain a consensus sequence. Sequences of bursa-derived IBDV strains
have been published by Vakharia et al. (1992) (strain E\Del, D10065),
Bayliss et al. (1990) (52\70, D00869), Thiry et al. (1992) (Edgar, A33255),
Kibenge et al. (1990) (STC, D00499), Yamaguchi et al. (1996) (OKYM,
D49706 ; and TKSM, D84072), Brown et al. (1994) (DV86, Z25482 ; and
UK661, X92760), Hudson et al. (1986) (002-73, X03993), Vakharia et al.
(1994) (GLS, M97346), Yamaguchi et al. (1997) (LukertBP, D16679), Lana
et al. (1992) (variant A strain, M64285), Pitcovski et al. (1998) (KS,
CAGI
L42284) and Heine et al. (1991) (variant E strain, D10065). Unpublished
sequences from strains U-26 (AF091099), Miss (AF091098) and 3212
(AF091097) (H. S. Sellers, P. N. Villegas, D. J. Jackwood & B. S. Seal,
unpublished results) were also included in the comparison. In this
manuscript, a consensus sequence was deduced and compared with the
consensus sequence of tissue culture-infecting strains. Sequences were
analysed with the GCG software package, version 8 (Genetics Computer
Group, Wisconsin, USA).
Construction of a full-length cDNA clone of segment B of
strain D78. For cloning of the full-length cDNA of segment B of
serotype I strain D78, virus was propagated in CEC and purified by
ultracentrifugation as described previously (Mu$ ller et al., 1986). Genomic
viral RNA of strain D78 was purified (Mundt & Mu$ ller, 1995), reversetranscribed into cDNA and amplified by PCR by standard procedures,
using oligonucleotides described previously (Mundt & Vakharia, 1996).
The amplification product was cloned blunt-ended and plasmids
containing appropriate PCR fragments were sequenced. The cloning
procedure to obtain a plasmid containing the full-length cDNA of
segment B (pD78B) under control of the T7 RNA polymerase promoter
corresponded to the procedure described by Mundt & Vakharia (1996)
for segment B of strain P2.
Construction of chimeric IBDV plasmids. A prerequisite for the
following site-directed mutagenesis was the modification of the plasmid
pUC18. To this end, pUC18 was cleaved with NdeI and BamHI,
electroeluted, blunt-ended by Klenow polymerase and religated to obtain
pUC18\∆NB. Plasmid pAD78\EK (Mundt et al., 1997) was cleaved with
EcoRI and KpnI to obtain the full-length sequence of segment A of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 22:32:42
R$$! S
None
None
1115–1170
833–866
1228–1252
StyI
–
–
Antisense
Sense
Antisense
Q
A
H#&$
T#)%
861–891
901–1000
SacI
NarI
Sense
Sense
R
S$$!
1094–1193
SpeI
Antisense
None
Q#&$ H
A#)% T
601–622
861–891
901–999
D78-MutRS
A44
A14
D78-MutHQ
D78-MutTA
E\Del-MutSR
P21F
E\Del-MutQH
E\Del-MutAT
CGTCCTAGTAGGGGAAGGGGTC
GAGAGCTCGTGTTCAAAACAAGCGTCCAtAG
GGGCGCCACCATCTACCTTATAGGCTTTGATGGGACTGCGGTAATCACCAGAGCTGTGG
CCGCAAACAATGGGCTGACGaCCGGCATCGACAATCTTAT
CGTAGGCTACTAGTGTGACGGGACGGAGGGCTCCTGGATAGTTGCCACCATGGATCGTC
ACTGCTAGGCTCCCcCgTGCCGACCATGACATCTGTTCCCC
GAGAGCTCGTGTTTCAAACAAGCGTCCAaGG
GGGCGCCACCATCTACCTCATAGGCTTTGATGGGACAACGGTAATCACCAGGGCTGTGG
CCGCAAACAATGGGCTGACGgCCGGCACCGACAACCTTATG
CGGAGGGCCCCTGGATAGTTGCCACCATGGATCGTCACTGCTAGGCTCCCaCTTGC
CAAGCCTCAGCGTTGGGGGAGAGC
GATCAGCTCGAAGTTGCTCACCCCA
–
SacI
NarI
Sense
Sense
Sense
Amino acid
substitution
Orientation
Position
Restriction
enzyme used
Sequence
Primer
The sequences and locations of the oligonucleotide primers used for site-directed mutagenesis and cloning are shown. The restriction sites used are underlined and the restriction enzymes
are given. Nucleotides changed in mutagenesis are in lower case and the coding nucleotide triplets are highlighted in boldface type. The positions where the primers bind (nucleotide
number) and the numbering of amino acids are in accordance with the published sequence of strain P2 (Mundt & Mu$ ller, 1995).
Table 1. Oligonucleotides used for the construction of full-length cDNA clones of IBDV segment A containing amino acid substitutions
Adaptation of IBDV to CEC by reverse genetics
serotype I strain D78, including the T7 RNA polymerase promoter site.
This fragment was ligated into EcoRI\KpnI-cleaved pUC18\∆NB to
obtain pD78A. Plasmid pD78A was used as the backbone for the
following cloning and site-directed mutagenesis procedures.
For substitution of IBDV-specific sequences, a plasmid containing the
complete coding region of the USA variant E strain E\Del was used (p1022, a generous gift from Dr A. van Loon, Intervet). p10-22 was cleaved
with restriction enzymes NdeI and SalI at nucleotides 647 and 1725,
respectively (numbering follows the full-length sequence of strain P2,
NCBI accession no. X84034), to obtain a 1078 bp fragment encompassing
coding sequences of the variable region of VP2 and partial sequences of
VP4 of strain E\Del. After ligation into NdeI\SalI-cleaved pD78A, a
chimeric full-length plasmid pD78A-E\Del, containing sequences of
segment A of strain D78 as well as E\Del, was established. Plasmids
pD78A and pD78A-E\Del were used for site-directed mutagenesis and
are depicted in Fig. 1.
A further pair of plasmids was constructed, containing the variable
regions of GLS-BU and GLS-TC. For cloning of the variable region, GLSTC was propagated in CEC and purified by ultracentrifugation as
described previously (Mu$ ller et al., 1986). A bursal homogenate of GLSBU was purified by low-speed centrifugation and the supernatant was
further processed. After 0n5 mg\ml proteinase K\0n5 % SDS digestion,
viral RNA was purified (Mundt & Mu$ ller, 1995), reverse-transcribed into
cDNA and amplified by PCR by standard procedures, using oligonucleotides A14 and A44 (Table 1). The amplification product was cloned
blunt-ended and plasmids containing appropriate PCR fragments (pGLSTC, pGLS-BU) were sequenced. For construction of chimeric segments A,
the full-length clone pD78A-E\Del was used. pGLS-TC and pGLS-BU
were digested with SacI and SpeI, respectively. Electroeluted fragments
were subsequently ligated into SacI\SpeI-digested pD78A-E\Del to
obtain pD78A-E\Del-GLS-TC and pD78A-E\Del-GLS-BU, respectively.
Maps of both plasmids are depicted in Fig. 1.
Site-directed mutagenesis. Site-directed mutagenesis was performed by PCR. Oligonucleotides contained restriction enzyme cleavage
sites and mutations leading to amino acid exchanges (Table 1). For the
establishment of full-length clones of segment A of plasmid pD78A
containing mutated codons, an NdeI–HindIII fragment of pD78A was
subcloned into NdeI\HindIII-cleaved pUC19 (pUC19\∆NH) to obtain
single restriction enzyme sites for the cloning procedures (pUC19\∆NHD78A, Fig. 1). Plasmids based on pUC19\∆NH-D78A, containing the
mutated codons for residues 253 (H to Q), 284 (T to A), 330 (R to S) and
all three amino acids, were cleaved with NdeI and SacII, appropriate
fragments were electroeluted and ligated into pD78. Mutations in pD78E\Del were generated by following standard procedures. Here, substitutions of residues 253 (Q to H), 284 (A to T) and 330 (S to R) were
performed in all seven combinations (Table 2). After PCR, amplification
fragments were cloned blunt-ended and sequenced. Fragments containing
the mutated codons were ligated into plasmids using restriction enzyme
cleavage sites as described in Fig. 1. Oligonucleotides, plasmids used for
PCR and the resulting plasmids are summarized in Table 2. Sequences of
the exchanged parts of final plasmids were analysed with the GCG
software package, version 8.
In vitro translation of segments A. To analyse processing of
viral proteins of the mutated segments A, in vitro transcription\
translation was performed by using the TNT-coupled reticulocyte lysate
system (Promega) in accordance with the manufacturer’s instructions.
Plasmid DNA for the in vitro transcription was obtained by using a
plasmid miniprep kit (Quantum Prep, Bio-Rad).
Products of the TNT reaction were radioimmunoprecipitated with
rabbit anti-IBDV antiserum and separated by SDS–PAGE under standard
conditions (Sambrook et al., 1989).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 22:32:42
CAGJ
E. Mundt
Table 2. Summary of oligonucleotides and plasmids used for site-directed mutagenesis
Plasmids used as templates for PCR and restriction enzyme cleavage sites used for cloning of the mutagenized
DNA fragments are indicated.
Oligonucleotides
Template plasmid
Cleavage sites
Plasmid product
E\Del-MutQH, A14
E\Del-MutAT, A14
E\Del-MutSR, P21F
E\Del-MutQH, E\Del-MutSR
E\Del-MutAT, E\Del-MutSR
E\Del-MutQH, A14
E\Del-MutQH, E\Del-MutSR
D78-MutHQ, A14
D78-MutTA, A14
D78-MutRS, P21F
D78-MutHQ, D78-MutRS
pD78A-E\Del
pD78A-E\Del
pD78A-E\Del
pD78A-E\Del
pD78A-E\Del
pD78A-E\Del-AT
pD78A-E\Del-AT
pD78A
pD78A
pD78A
pD78A-TA
SacI, SpeI
NarI, SpeI
SacI, SpeI
SacI, SpeI
NarI, SpeI
SacI, SpeI
SacI, SpeI
SacI, StyI
NarI, StyI
SacI, StyI
SacI, StyI
pD78A-E\Del-QH
pD78A-E\Del-AT
pD78A-E\Del-SR
pD78A-E\Del-QH-SR
pD78A-E\Del-AT-SR
PD78A-E\Del-QH-AT
pD78A-E\Del-QH-AT-SR
pD78A-HQ
pD78A-TA
pD78A-RS
pD78A-HQ-TA-RS
Table 3. Summary of transfection experiments using intergeneric full-length cDNA clones
of IBDV segment A and segment B of strain D78
Plasmids were based on the full-length cDNA clone of segment A of tissue culture-adapted, serotype I strain
D78. Sequences of bursa-derived serotype I strains GLS-BU, Delaware E (E\Del) and of the tissue cultureadapted, serotype I strain GLS-TC were substituted with D78 sequences. , Not done.
Residue*
Plasmid
pD78A-E\Del
pD78A-E\Del-QH
pD78A-E\Del-AT
pD78A-E\Del-SR
pD78A-E\Del-QH-AT
pD78A-E\Del-AT-SR
pD78A-E\Del-QH-SR
pD78A-E\Del-QH-AT-SR
pD78A
pD78A-HQ
pD78A-TA
pD78A-RS
pD78A-HQ-TA-RS
pD78A-E\Del-GLS-TC
pD78A-E\Del-GLS-BU
Tissue culture†
CAM‡
253 284 330 Transfection Passage§ Transfection Passage§
Q
H
Q
Q
H
Q
H
H
H
Q
H
H
Q
Q
Q
A
A
T
A
T
T
A
T
T
T
A
T
A
T
A
S
S
S
R
S
R
R
R
R
R
R
S
S
S
S
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
k
k
k
k
j
k
k
j
j
k
k
j
k
j
k
j
j
j
j
k
k
k
k
k
j
j
j
j


k
k
k
k
j
k
k
j
j
k
k
j
k


* The numbering of amino acids is according to the published sequence of strain P2 (Mundt & Mu$ ller, 1995).
Mutated amino acids are in boldface type.
† CEC and QM-7 cells were used for transfection experiments. IBDV antigen was detected by IFA with
rabbit anti-IBDV antiserum either 24 h after transfection or after passage onto CEC. Presence of antigen is
shown.
‡ CAM of 11-day-old embryonated eggs were transfected. CAM were analysed for IBDV antigen 6 days
after transfection by Western blot with rabbit anti-IBDV antiserum and by ELISA. Presence of antigen is
shown.
§ CEC were used for passaging of transfection supernatants. IBDV antigen was detected by IFA with rabbit
anti-IBDV antiserum after passage onto CEC. Presence of antigen is shown.
CAHA
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 22:32:42
Adaptation of IBDV to CEC by reverse genetics
Fig. 2. Immunofluorescence with rabbit anti-IBDV antiserum of QM-7 cells 24 h after transfection (A, C) and after passaging of
the supernatants of transfected cells (B, D). Cells were transfected with in vitro-transcribed cRNA of either pD78A (A) or
pD78A-E/Del (C) together with cRNA of pD78B. Transfection supernatants of pD78A (B) and pD78A-E/Del (D) were
passaged once in CEC and tested for IBDV antigen. Magnification, i300.
Virus recovery from cRNA in tissue culture. For in vitro
transcription of RNA, plasmids containing segment A (Table 3) and
pD78B were linearized by cleavage with either BsrGI or PstI. Further
treatment of linearized DNA and transcription were carried out as
described by Mundt & Vakharia (1996) with two exceptions : (i) the
transcription mixtures were not purified by phenol–chloroform extraction
and (ii) QM-7 cells and CEC were used for transfection experiments as
described previously (Mundt & Vakharia, 1996). Two days after
transfection, cells were freeze–thawed and centrifuged at 700 g to
eliminate cellular debris and the resulting supernatants were clarified
further by filtration through 0n45 µm filters and stored at k70 mC. For
immunofluorescence studies, cells were grown on sterile coverslips.
Virus recovery from cRNA after transfection of CAM. For
recovery of viruses that did not infect QM-7 cells or CEC, an alternative
method was applied. Transcription mixtures were transfected onto the
CAM of 11-day-old embryonated SPF eggs. Six days after transfection,
CAM were harvested and homogenized. Supernatants were clarified by
low-speed centrifugation and used for Western blot, ELISA and passaging
onto CEC.
Detection of IBDV antigen. IBDV antigen was detected by
indirect IFA and Western blot with rabbit anti-IBDV antiserum (Mundt
et al., 1995). Supernatants of transfected QM-7 cells, CEC and CAM were
passaged onto CEC. For IFA, QM-7 cells as well as CEC grown on
coverslips were incubated with supernatants resulting from the passaging
experiments. After 16 h, cells were acetone-fixed and processed for IFA.
For examination of IBDV replication after transfection, QM-7 cells and
CEC grown on coverslips were incubated for 24 and 48 h, acetone-fixed
and processed for IFA.
To detect IBDV antigen in transfected CAM, clarified supernatants of
homogenized CAM were analysed by Western blot with a rabbit antiIBDV antiserum. An ELISA (van Loon et al., 1994) was performed to
confirm the Western blot results,.
Analysis of tissue culture-infecting virus. IBDV able to infect
tissue cultures was passaged onto CEC and purified by ultracentrifugation
and the resulting pellets were digested with proteinase K\SDS as
described above. After precipitation, genomic RNA was reversetranscribed into cDNA and amplified by PCR under standard conditions
with oligonucleotides A14 and A44 (Table 1). PCR fragments were
cloned blunt-ended and three clones resulting from each virus were
sequenced. Sequences were analysed using the GCG software package,
version 8.
Results
Transfection experiments with chimeric cRNA
For transfection experiments, a full-length cDNA clone of
segment A of strain D78 (pD78A) and the chimeric segment A
pD78A-E\Del were transcribed into synthetic cRNA and
cotransfected with segment B (pD78B) full-length cRNA into
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 22:32:42
CAHB
E. Mundt
(A)
D78
TC
BU
E/Del
D78
TC
BU
E/Del
(B)
GLS-BU
GLS-TC
GLS-BU
GLS-TC
Fig. 3. Comparison of deduced amino acid sequences of the variable
region of VP2 of IBDV. (A) The consensus sequence of tissue cultureadapted (TC) strains is compared with the consensus sequence of
bursa-derived strains (BU) and the sequences of strains D78 and E/Del.
(B) Sequences of the bursa-derived strain GLS (GLS-BU) and the tissue
culture-adapted variant GLS-TC are aligned. Differences are indicated.
Identical amino acids are marked by dashes. Numbering of amino acids is
according to the published sequence of strain P2 (Mundt & Mu$ ller, 1995).
QM-7 cells, as well as into CEC in parallel. Two days after
transfection, cells were freeze–thawed and the resulting
supernatants were passaged twice on CEC. CEC were
incubated for up to 5 days after infection in each passage. After
freeze–thawing, transfection and passage, supernatants were
tested for IBDV antigen by IFA with CEC and QM-7 cells.
Transfection experiments were repeated three times. Virus was
generated after transfection of cRNA from plasmid pD78B in
combination with pD78A, leading to strain D78r (Fig. 2 B). In
contrast, after transfection experiments with cRNA from
pD78A-E\Del and pD78B, no tissue culture-infecting virus
could be isolated (Fig. 2 D). To analyse whether transfection
was followed by replication, transfection was carried out using
cells growing on coverslips. In both cases ,virus antigen was
detected by IFA 24 h after transfection (Fig. 2 A, C). Thus,
virus replication ensued in both cases, but it was only possible
to generate tissue culture-infecting IBDV in the case of D78r.
Comparison of viral sequences
As a result of this observation, and assuming that VP2
plays an important role during virus entry, amino acid
sequences comprising the variable region of VP2 of different
IBDV were compared. Sequences were divided into two
groups before alignment on the basis of their origin. One
group contained sequences of tissue culture-adapted (TC)
strains, whereas the second group consisted of sequences of
IBDV not able to infect tissue culture (bursa-derived strains,
BU). After alignment of amino acid sequences (aa 241–360) of
TC and BU groups, the consensus sequences of both groups
were compared (Fig. 3 A). Differences in amino acids were
found in the following positions (the residue found in TC
CAHC
strains is given first) : 241 (V I), 253 (H Q), 270 (T A),
279 (N D), 284 (T A) and 330 (R S). Within the TC
group, residues V#%" and H#&$ were conserved in 6 of 8 and 7
of 8 strains, respectively, whereas N#(* and T#)% were present
in all strains. Amino acids T#(! (5 of 8) and R$$! (4 of 8) showed
the lowest degree of conservation within the consensus
sequence of the TC group. Residues Q#&$ (16 of 17), A#)% (14
of 17) and S$$! (17 of 17) were the positions most conserved
in the BU group. Residues I#%" (9 of 17), A#(! (12 of 17) and
D#(* (10 of 17) showed a degree of identity lower than 73 %.
The amino acid sequence of the tissue culture-adapted strain
D78 was 100 % identical to the TC consensus sequence.
Comparison of the sequence of the E\Del (BU) strain with
those of the TC group sequence revealed five amino acid
exchanges (TC sequence given first), at positions 249 (Q K),
254 (G S), 286 (T I), 318 (G D) and 323 (D E), in
addition to the amino acid substitutions between the TC and
BU group. In contrast, amino acids 241 and 279 were not
different. Taken together, residues 253 and 284 were highly
conserved in both groups, whereas S$$! was the only amino
acid fully conserved within the BU group.
Transfection experiments with mutated cRNA
On the basis of the results of the sequence comparison,
several different mutated full-length cDNA clones were
established by site-directed mutagenesis. Mutated plasmids of
pD78A-E\Del were generated containing substitutions at
amino acid positions 253, 284 and 330 in all possible
combinations (Table 3). Transfection experiments and
passaging were performed three times in parallel on CEC and
QM-7 cells. The supernatants obtained were analysed for
infectivity by IFA. Infectious virus could not be isolated after
transfection of cRNA of plasmids pD78A-E\Del, pD78AE\Del-QH, pD78A-E\Del-AT, pD78A-E\Del-SR, pD78AE\Del-AT-SR and pD78A-E\Del-QH-SR in combination with
cRNA of pD78B into QM-7 cells or CEC. Transfection of
cRNA obtained from pD78A-E\Del-QH-AT or pD78AE\Del-QH-AT-SR led to generation of infectious virus (D78AE\Del-QH-AT and D78A-E\Del-QH-AT-SR). Specificity was
confirmed by IFA on CEC as well as QM-7 cells (data not
shown). This indicated that VP2 of IBDV plays a critical role in
tissue culture infection. Amino acid substitutions Q#&$ H
and A#)% T were necessary and sufficient to generate IBDV
infectious for the tissue cultures used. To confirm these results,
a second set of plasmids was constructed by using pD78A for
site-directed mutagenesis to obtain plasmids containing substitutions of either a single residue (pD78A-HQ, pD78A-TA,
pD78A-RS) or of all three amino acids (pD78A-HQ-TA-RS).
These four plasmids were used for transfection experiments in
combination with pD78B as described above. Infectious IBDV
could be generated after transfection of cRNA from pD78ARS (D78-RS) as detected by IFA. Again, residue 330 did not
have any influence on the ability of the virus generated to
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 22:32:42
Adaptation of IBDV to CEC by reverse genetics
processed into VP2, VP3 and VP4. The apparent molecular
masses of VP3 (32 kDa) and VP4 (28 kDa) showed no
differences between the two sets of full-length clones, pD78A
and derivatives (lanes 1–5) and pD78A-E\Del and derivatives
(lanes 6–13). The apparent molecular mass of VP2 differed
between the two sets of full-length clones as well as within the
sets. For example, exchange of R$$! in VP2 encoded by pD78A
(lane 1) to serine in pD78A-RS (lane 4) seemingly led to an
increase of the apparent molecular mass of approximately
2 kDa.
Transfection of chimeric cRNA containing strain GLS
sequences
Fig. 4. Radioimmunoprecipitation of in vitro-translated products of different
segments A of IBDV. Plasmids used are : pD78A (lane 1), pD78A-HQ (2),
pD78A-TA (3), pD78A-RS (4), pD78A-HQ-TA-RS (5), pD78A-E/Del (6),
pD78A-E/Del-QH (7), pD78A-E/Del-AT (8), pD78A-E/Del-SR (9),
pD78A-E/Del-QH-AT (10), pD78A-E/Del-AT-SR (11), pD78A-E/Del-QHSR (12) and pD78A-E/Del-QH-AT-SR (13). Translated proteins were
precipitated by using rabbit anti-IBDV serum. The locations of viral
proteins and molecular mass markers are indicated.
infect tissue culture, but single substitutions at positions 253 (H
to Q) or 284 (T to A) were sufficient to prevent generation of
tissue culture-infecting IBDV. All constructs were tested by
IFA for replication after transfection. IBDV antigen could be
detected 24 and 48 h after transfection, showing typical large,
intensely stained aggregates within the cytoplasm (data not
shown).
RT–PCR was performed to examine whether changes in
the mutated part of the viral genome occurred during
replication of D78r, D78-RS, D78A-E\Del-QH-AT or D78AE\Del-QH-AT-SR. Sequence analysis of the cloned PCR
fragments revealed no differences from the sequences of the
plasmids used for cRNA synthesis (data not shown).
To confirm the results of the transfection experiments with
chimeric as well as mutated plasmids, we took advantage of a
naturally occurring pair of IBDV strains. The variable regions
of VP2 of the bursa-derived strain GLS (GLS-BU) and the tissue
culture-adapted variant GLS-TC were amplified, cloned and
analysed. Comparison of the amino acid sequences of the two
strains obtained from pGLS-BU and pGLS-TC revealed one
amino acid exchange, at position 284, from A in GLS-BU to T
in GLS-TC (Fig. 3 B). Residues 253 (Q) and 330 (S) were
identical to those of the BU group, as described above. To
analyse whether the exchange at position 284 (A T) was
sufficient for the generation of tissue culture-infectious virus,
two plasmids (pD78A-E\Del-GLS-TC and pD78A-E\DelGLS-BU) were constructed containing the variable regions of
VP2 from the two variants. cRNA from pD78A-E\Del-GLSTC and pD78A-E\Del-GLS-BU was transfected with cRNA of
pD78B into QM-7 cells and CEC. After passaging of the
supernatants in tissue culture, infectious virus could be detected
by IFA and CPE occurred after transfection of cRNA of
pD78A-E\Del-GLS-TC. In several attempts, transfection of
cRNA from pD78A-E\Del-GLS-BU failed to produce tissue
culture-infectious IBDV. In vitro transcription\translation of
both plasmids showed complete processing of the polyproteins
(data not shown). Viral antigen was detected by IFA after
transfection of cRNA of both plasmids together with cRNA
from pD78B. Thus, both chimeras proved to be replicationcompetent (data not shown). In summary, a single amino acid
exchange at position 284 of VP2 of strain GLS was sufficient to
generate cell culture-infectious chimeric IBDV (Table 3).
Recovery of chimeric virus from CAM
In vitro transcription and translation of cDNA
constructs of segment A
To exclude the possibility that polyprotein processing had
been altered by cloning manipulation or amino acid substitution, each full-length clone of segment A was transcribed
and translated in vitro. Products were immunoprecipitated with
rabbit anti-IBDV antiserum and analysed after SDS–PAGE and
fluorography (Fig. 4). As expected, the viral polyprotein was
CAM of 11-day-old embryonated eggs were used for
transfection of in vitro-transcribed cRNAs. Supernatants harvested from the homogenized CAM were analysed by
Western blot and ELISA. Antigen could be detected 6 days
after transfection of CAM with pD78A, pD78A-HQ, pD78ATA, pD78A-RS, pD78A-HQ-TA-RS, pD78A-E\Del, pD78AE\Del-QH, pD78A-E\Del-AT or pD78A-E\Del-SR in conjunction with pD78B. Because the concentration of IBDV
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 22:32:42
CAHD
E. Mundt
antigen was low, the results of Western blot experiments
were confirmed by ELISA. No IBDV-specific antigen could be
detected by both Western blot and ELISA after transfection of
CAM with pD78A-E\Del-QH-AT, pD78A-E\Del-QH-SR,
pD78A-E\Del-AT-SR or pD78A-E\Del-QH-AT-SR in conjunction with pD78B. Furthermore, supernatants of transfected
CAM were passaged onto CEC. Testing of CEC supernatants
by IFA resulted in detection of IBDV antigen for D78r, D78RS, D78-E\Del-QH-AT and D78-E\Del-QH-AT-SR. Data
from the CAM experiments are summarized in Table 3.
Discussion
Two types of serotype I strains have been described for
IBDV. Strains of the first type are able to infect and replicate in
B lymphocytes located in the bursa of Fabricius (bursa-derived)
but not in CEC. In contrast, serotype I strains of the second
type infect and replicate in both B lymphocytes and CEC
(adapted strains). In this report, the structural requirements for
either phenotype were identified at the molecular level. The
viral protein responsible for the difference in tissue tropism
was determined after exchange of parts of VP2 and VP4
between bursa-derived and adapted strains that led to amino
acid exchanges in VP2. Furthermore, mutational analysis
identified specific residues in VP2 that were responsible for the
alteration of tropism from bursa-derived to tissue cultureadapted strains. Two amino acid exchanges, (BU) Q#&$ H
(TC) and (BU) A#)% T (TC), in VP2 of the bursa-derived
E\Del strain were necessary for the ability to infect CEC as
well as QM-7 cells. This was confirmed by the results of single
amino acid exchanges in the tissue culture-adapted strain D78,
where substitutions at positions 253 (H to Q) or 284 (T to A)
resulted in IBDV that was not able to infect tissue culture cells.
In contrast, only one amino acid exchange, (BU) A#)% T
(TC), was sufficient for the alteration of the tissue tropism of
GLS-BU to GLS-TC. Interestingly, residue 253 was conserved
as glutamine in both GLS strains, as in most of the bursaderived IBDV strains. In the case of the GLS strain, the
exchange of residue 284 (A to T) during natural adaptation was
sufficient to obtain tissue culture infectivity. Comparison of the
amino acid sequence of strain GLS-BU with the sequence of
strain D78 revealed eight amino acid exchanges (K#%* Q,
Q#&$ H, S#&% G, S#'* T, A#(! T, A#)% T, E$#" A
and S$$! R). Why a single amino acid exchange in the GLS
strains was sufficient to generate infectious GLS-TC is not
clear. However, each amino acid exchange may lead to
alterations within the structure of VP2 that together lead to
acquisition of the property of tissue culture infectivity.
After comparison of amino acid sequences of two tissue
culture-adapted IBDV strains with sequences of their parental
strains, Yamaguchi et al. (1996) suggested that residues N#(*
and T#)% are important for the ability to infect CEC as well as
for virulence. Comparison of the amino acid sequences of
strains D78, E\Del, GLS-BU and GLS-TC with the sequences
CAHE
reported by Yamaguchi et al. (1996) revealed that the amino
acid of the proposed tissue-culture type (glutamine) was
located at position 279 in each case. Whether residue 279
indeed plays an important role in infection of CEC remains
unclear. N#(* is conserved in nearly all tissue culture-adapted
strains, with the exception of the vaccine strain Bursine 2,
which has aspartate at position 279 (Eterradossi et al., 1998).
Here, the amino acid sequence (H#&$ and T#)%) was of the tissue
culture type. This contrasts with the proposal of Yamaguchi et
al. (1996). Based on the data obtained in this study, residue 284
appears to play a critical role in the infectivity of IBDV in tissue
culture, since all tissue culture-adapted strains analysed have
threonine at this position. Obviously, residue S$$! does not
play any role in tissue culture infection. The sequence motif
SWSAS$$!GS has been identified as being conserved in
segment A of virulent strains (Heine et al., 1991 ; Lin et al.,
1993 ; Vakharia et al., 1994), as well as in very virulent strains
in Europe and Japan (Lin et al., 1993 ; Brown et al., 1994). These
strains are not able to infect tissue culture. In summary, an
amino acid exchange at position 330 is not necessary for CEC
infection but may play a role in the virulence of IBDV in
chickens. On the basis of these data, it can be suggested that
the ability to infect CEC is determined by the variable region
of VP2. It seems that infection of CEC is not associated with
conserved amino acid alterations, since the amino acids
responsible are specific for the strain used. Therefore, any
prediction of effects caused by a particular amino acid alteration
is impossible without detailed knowledge of the threedimensional structure of VP2 and identification of the domains
involved in tissue culture tropism. Data regarding the threedimensional structure of VP2 are not available. Furthermore,
the use of theoretical models based on the amino acid sequence
is not possible, since the carboxy terminal amino acid of VP2
is unknown. However, the amino acids responsible for infection
of CEC can be determined by amino acid substitution,
accomplished by using the IBDV reverse genetics system.
During the early part of this study, it was not possible to
generate a chimeric virus based on the exchanged part of the
bursa-derived E\Del strain. IBDV antigen could be detected by
using IFA after transfection of viral cRNA, proving that
transcription and translation occurred. To overcome this
problem, it was necessary to develop a system that allowed
generation of progeny of bursa-derived IBDV in the laboratory. The successful transfection of CAM of embryonated
eggs provided a new way of generating infectious IBDV, even
from isolates that were not able to infect tissue culture. IBDV
able to infect tissue culture was not detected by Western blot
and ELISA, due to the low virus yield, although the virus was
present, as shown by CEC passage. In contrast, in only two
cases of IBDV that was unable to infect tissue culture could
antigen not be detected by both ELISA and Western blot
(D78A-E\Del-AT-SR, D78A-E\Del-QH-SR). These two may
not be viable or the amount of antigen may have been too low.
However, both are able to produce virus after transfection in
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 22:32:42
Adaptation of IBDV to CEC by reverse genetics
tissue culture, as detected by IFA, which indicates that the
system for generation of IBDV in CAM needs to be optimized.
Problems associated with this system included reproducibility
and the difficulty of detection of virus by Western blot due to
low yields. Here, the use of a highly sensitive ELISA system
(van Loon et al., 1994) was helpful. With all these constraints,
it was important to show that viable chimeric as well as
mutated IBDV could be generated.
The results obtained show new methods for the characterization of structural elements of IBDV that are responsible for
replication and infectivity. The generation of chimeric IBDV
between TC and BU strains makes ‘ tailor-made ’ vaccines a
possibility for the future.
Hudson, P. J., McKern, N. M., Power, B. E. & Azad, A. A. (1986).
Genomic structure of the large RNA segment of infectious bursal disease
virus. Nucleic Acids Research 14, 5001–5012.
Kibenge, F. S. B., Jackwood, D. J. & Mercado, C. C. (1990). Nucleotide
sequence analysis of genome segment A of infectious bursal disease virus.
Journal of General Virology 71, 569–577.
Lana, D. P., Beisel, C. E. & Silva, R. F. (1992). Genetic mechanisms of
antigenic variation in infectious bursal disease virus : analysis of a
naturally occurring variant virus. Virus Genes 6, 247–259.
Lange, H., Mu$ ller, H., Ka$ ufer, I. & Becht, H. (1987). Pathogenic and
structural properties of wild type infectious bursal disease virus (IBDV)
and virus grown in vitro. Archives of Virology 92, 187–196.
Lin, Z., Kato, A., Otaki, Y., Nakamura, T., Sasmaz, E. & Ueda, S.
(1993). Sequence comparisons of a highly virulent infectious bursal
disease virus prevalent in Japan. Avian Diseases 37, 315–323.
I thank Thomas Mettenleiter for helpful discussions and critical
reading of the manuscript, Dietlind Kretzschmar for excellent technical
assistance and A. van Loon for performing the ELISA. This study was
supported in part by DFG grant MU 1244\1-1.
References
Baxendale, W. (1976). The development of an apathogenic infectious
bursal agent vaccine : field trial results. In Proceedings of the 25th Western
Poultry Disease Conference and 10th Poultry Health Symposium, pp. 42–45.
University of California, Davis, CA, USA.
Bayliss, C. D., Spies, U., Shaw, K., Peters, R. W., Papageorgiou, A.,
Mu$ ller, H. & Boursnell, M. E. G. (1990). A comparison of the sequences
of segment A of four infectious bursal disease virus strains and
identification of a variable region in VP2. Journal of General Virology 71,
1303–1312.
Brown, M. D., Green, P. & Skinner, M. A. (1994). VP2 sequences of
recent European ‘ very virulent ’ isolates of infectious bursal disease virus
are closely related to each other but are distinct from those of ‘ classical ’
strains. Journal of General Virology 75, 675–680.
Cosgrove, A. S. (1962). An apparently new disease of chickens – avian
nephrosis. Avian Diseases 6, 385–389.
Cursiefen, D., Ka$ ufer, I. & Becht, H. (1979). Loss of virulence in a small
plaque mutant of the infectious bursal disease virus. Archives of Virology
59, 39–46.
Dobos, P., Berthiaume, L., Leong, J. A., Kibenge, F. S. B., Mu$ ller, H. &
Nicholson, B. L. (1995). Family Birnaviridae. In Virus Taxonomy. Sixth
Report of the International Committee on Taxonomy of Viruses, pp. 240–244.
Edited by F. A. Murphy, C. M. Fauquet, D. H. L. Bishop, S. A. Ghabrial,
A. W. Jarvis, G. P. Martelli, M. A. Mayo & M. D. Summers. Vienna &
New York : Springer-Verlag.
Eterradossi, N., Arnauld, C., Toquin, D. & Rivallan, G. (1998). Critical
amino acid changes in VP2 variable domain are associated with typical
and atypical antigenicity in very virulent infectious bursal disease viruses.
Archives of Virology 143, 1627–1636.
Hassan, M. K., Nielsen, C. K., Ward, L. A., Jackwood, D. J. & Saif, Y. M.
(1996). Antigenicity, pathogenicity, and immunogenicity of small and
large plaque infectious bursal disease virus clones. Avian Diseases 40,
832–836.
Heine, H.-G., Haritou, M., Failla, P., Fahey, K. & Azad, A. (1991).
Sequence analysis and expression of the host-protective immunogen VP2
of a variant strain of infectious bursal disease virus which can circumvent
vaccination with standard type I strains. Journal of General Virology 72,
1835–1843.
McFerran, J. B., McNulty, M. S., Killop, E. R., Connor, T. J., McCracken,
R. M., Collins, P. S. & Allan, G. M. (1980). Isolation and serological
studies with infectious bursal disease virus from fowl, turkeys and ducks :
demonstration of a second serotype. Avian Pathology 9, 395–404.
Mu$ ller, H., Lange, H. & Becht, H. (1986). Formation, characterization
and interfering capacity of a small plaque mutant and of incomplete virus
particles of infectious bursal disease virus. Virus Research 4, 297–309.
Mundt, E. & Mu$ ller, H. (1995). Complete nucleotide sequences of 5hand 3h-noncoding regions of both genome segments of different strains
of infectious bursal disease virus. Virology 209, 10–18.
Mundt, E. & Vakharia, V. N. (1996). Synthetic transcripts of doublestranded Birnavirus genome are infectious. Proceedings of the National
Academy of Sciences, USA 93, 11131–11136.
Mundt, E., Beyer, J. & Mu$ ller, H. (1995). Identification of a novel viral
protein in infectious bursal disease virus-infected cells. Journal of General
Virology 76, 437–443.
Mundt, E., Ko$ llner, B. & Kretzschmar, D. (1997). VP5 of infectious
bursal disease virus is not essential for viral replication in cell culture.
Journal of Virology 71, 5647–5651.
Nick, H., Cursiefen, D. & Becht, H. (1976). Structural and growth
characteristics of infectious bursal disease virus. Journal of Virology 18,
227–234.
Pitcovski, J., Goldberg, D., Levi, B. Z., Di-Castro, D., Azriel, A., Krispel,
S., Maray, T. & Shaaltiel, Y. (1998). Coding region of segment A
sequence of a very virulent isolate of IBDV – comparison with isolates
from different countries and virulence. Avian Diseases 42, 497–506.
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning : A
Laboratory Manual, 2nd edn. Cold Spring Harbor, NY : Cold Spring
Harbor Laboratory.
Schobries, H. D., Wilke, I. & Schmidt, U. (1977). Infektio$ se Bursitis
(Gumboro disease) in einem Broilerbestand. Monatshefte fuW r VeterinaW rmedizin 32, 700–704.
Spies, U., Mu$ ller, H. & Becht, H. (1987). Properties of RNA polymerase
activity associated with infectious bursal disease virus and characterization of its reaction products. Virus Research 8, 127–140.
Spies, U., Mu$ ller, H. & Becht, H. (1989). Nucleotide sequence of
infectious bursal disease virus genome segment A delineates two major
open reading frames. Nucleic Acids Research 17, 7982.
Thiry, G., Colau, D., De Wannemaeker, C. & Malarme, D. (1992). Patent
EP 0517292-A 29.
Vakharia, V. N., Ahamed, B. & He, J. (1992). Use of polymerase chain
reaction for efficient cloning of dsRNA segments of infectious bursal
disease virus. Avian Diseases 36, 736–742.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 22:32:42
CAHF
E. Mundt
Vakharia, V. N., He, J., Ahamed, B. & Snyder, D. B. (1994). Molecular
Hirai, K. (1996). Identification of sequence changes responsible for the
basis of antigenic variation in infectious bursal disease virus. Virus
Research 31, 265–273.
van Loon, A. A. W. M., Lu$ tticken, D. & Snyder, D. B. (1994). Rapid
quantification of infectious bursal disease (IBD) challenge, field or vaccine
virus strains. In Proceedings of an International Symposium on Infectious
Bursal Disease and Chicken Infectious Anemia, pp. 179–187. Rauschholzhausen, Germany.
attenuation of highly virulent infectious bursal disease virus. Virology
223, 219–223.
Yamaguchi, T., Ogawa, M., Miyoshi, M., Inoshima, Y., Fukushi, H. &
Hirai, K. (1997). Sequence and phylogenetic analyses of highly virulent
Yamaguchi, T., Ogawa, M., Inoshima, Y., Miyoshi, M., Fukushi, H. &
Received 22 March 1999 ; Accepted 10 May 1999
CAHG
infectious bursal disease virus. Archives of Virology 142, 1441–1458.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 22:32:42