_Archives
Vfrology
Arch Virol (1992) 127:89-101
© Springer-Verlag 1992 Printed in Austria
Naturally occurring-neutralizing monoclonal antibody escape variants define the epidemiology of infectious bursal disease viruses in the United States D. B. Snyder 1' 2, V. N. Vakharia 2' 1, and P. K. Savage a
t Virginia-Maryland Regional College of Veterinary Medicine and 2Center for Agricultural Biotechnology, Maryland Biotechnology Institute, Universtity of Maryland, College Park, Maryland, U.S.A.
Accepted April 20, 1992
Summary. A panel of two non-neutralizing and six neutralizing monoclonal antibodies (Mabs) were used in antigen-capture enzyme immunoassays (ACELISA) to examine the antigenicity of 1301 wild type infectious bursal disease viruses 0BDV) isolated from different poultry flocks througout the United States over a three year period. Analysis of these isolates with protective, neutralizing Mabs directed against the VP 2 structural protein of IBDV showed that four antigenically distinct groups of serotype 1 IBDV could be separated on the basis of the presence or absence of one or more Mab defined, conformation-dependent, multivalent neutralization site. AC-ELISA reactivity patterns of the Mabs with isolates demonstrated that IBDV field populations were relatively antigenically homogeneous per premise isolation. Geographically, various antigenic species were more or less prevalent, or nearly absent. Competition analysis with neutralizing Mabs coupled with AC-ELISA results suggested that neutralization epitopes for IBDV are distinct, spatially arranged, yet closely linked. Of 5 Mab defined neutralization epitopes, shown to be related to protection from virulent challenge by Classic IBDV strains isolated prior to 1985, only two of the epitopes remain unaltered on the most recent emergent variant field strain of IBDV isolated.
Introduction Infectious bursal disease virus (IBDV) is a bi-segmented, double-stranded RNA virus belonging to the family Birnaviridae [3]. The smaller genomic segment encodes VP 1 (98 kDa), the viral polymerase. The larger segment of the genome encodes three proteins VP 2 a/VP 2 b, VP 3, and VP 4, of which VP 2 and VP 3 are structural proteins, while VP 4 is the viral protease [3, 4]. VP 2 a (or VPX)
90
D.B. Snyder et al.
has an approximate molecular weight of 47 kDa, and is a precursor which is subsequently cleaved by VP 4 (28 kDa) to make VP 2 b (42 kDa). VP 2 a and VP 2 b of IBDV have been shown to be bound by neutralizing monoclonal antibodies (Mabs) [2, 4, 9], whereas VP 3 (32 kDa) has not been shown to carry neutralization epitopes [2, 4, 11]. Infection of susceptible chickens with virulent strains of IBDV leads to the rapid accumulation of high levels of viral antigen in the bursa of Fabricius. IBDV infection causes the depletion of lymphoidal follicles of the bursa of Fabricius and spleen, which initiates a highly contagious immunosuppressive disease in chickens [1, 3]. Infections caused by IBDV exacerbate infections produced by other etiologic agents, and thereby reduce the chicken's ability to respond to vaccination. The principle means of controlling infectious bursal disease in chickens are by administration of live attenuated and killed vaccines, and via the passage of high levels of vaccine induced maternal antibodies to young chickens. Two serotypes of IBDV have been defined [6-8]. Serotype 1 has been shown to cause immunologic disease in chickens, while serotpye 2 IBDV only causes sub-acute infections in turkeys. Although all reports are not in taxonomic agreement, laboratory and field strains within serotype 1 have been shown to be antigenically heterogeneous as determined by virus neutralization (VN) tests or by in vivo cross-challenge test [5-8]. In addition, other newly emergent variant strains of serotype 1 also differ pathotypically [14]. Since vaccination is the only method of control, the epidemiology of IBDV remains an important issue. However, virulent IBDV strains often are very difficult to adapt to tissue culture, and when they are adapted, they usually become less virulent. Hence, ultimately the virus adapted and assessed in VN tests may not be representative of the wild type isolate. Compounding the • problems of adaptation to in vitro substrates are the large numbers of isolates that are made and need to be typed. The virus is prevalent and environmentally stable, and when the chickens reach market age, one strain or another has infected virtually all broiler chickens we have ever examined (Snyder unpubl. data). Consequently, large scale typing of IBDV isolates has not previously been undertaken. We have previously characterized Mabs against strains of IBDV [11]. Two of the Mabs, R 63 and B 69, were shown to neutralize standard or Classic strains of IBDV to high titers in VN tests. These Mabs were employed in an antigen capture enzyme-linked immunosorbent assay (AC-ELISA) to examine the antigenicity of wild type IBDV isolates directly from clinically diseased bursal tissues [12]. We reported a pathotypic variant of IBDV could be identified by the apparent deletion or alteration of the neutralization site defined by the B 69 Mab. In this paper, we present the epidemiology of four serotype 1 IBDV field populations found in the United States which exist as naturally-occurring neutralization Mab escape mutants. Apparent progressive mutations occurring in Mab defined neutralization epitopes of field viruses are described. We also show the relative arrangement of six IBDV neutralization epitopes on VP 2 as defined
Variants define the epidemiology of IBDV
91
b y A C - E L I S A a n d c o m p e t i t i o n analysis o f Classic a n d n a t u r a l l y occurring variant strains o f I B D V .
Materials and methods Laboratory viruses, vaccines and fieM isolates Laboratory reference IBDV strains A/Del, D/Del, E/Del, G/Del, Edgar, MD, STC, and IM were obtained from previously acknowledged sources [11]. The prototype GLS IBDV variant was isolated by this laboratory during an earlier field survey in 1987, while the prototype DS326 strain was isolated during the initial stages of this study in 1988. The following commercial IBDV live vaccines were also used: S-706 (Select, Gainesville, GA); BBW and BBI (TRI BIO Gainesville, GA); BUR-1 and BUR-2 (Salisbury, Charles City, IA); UNI (American Scientific, Madison, WI); BBLEN (CEVA, Overland Park, KS); BV, BVM, and BV 4 (Sterwin Millsboro, DE); D 78 (Intervet, Millsboro, DE). In many cases, a commonly referred to laboratory strain may also be a vaccine strain. While differences could exist between the laboratory virus and the vaccine derivatives, we choose the vaccine strain as representative for this epidemiological study because they are licensed biologics used to vaccinate chickens in the field. Field isolates of IBDV were obtained from individual commercial broiler chicken flocks. The isolates were made by obtaining 10 to 15 random bursa of Fabricius samples per flock and processing them as a pool for Mab assessment. Bursal submissions were solicited from commercial broiler companies in major broiler rearing areas throughout the United States, and sampling was conducted over a period of 36 months, extending from 1988 through 1990. Field IBDV isolates were most often made from bursa obtained from commercial broiler type chickens 16 to 28 days of age.
Monoclonal antibodies Murine hybridoma cell lines secreting Mabs against the E/Del and GLS viruses were prepared and characterized in AC-ELISA, VN tests and agar gel precipitin tests as described for the B 29, R 63, and B 69 Mabs which were prepared previously [11].
Antigen-capture ELISA Bursal homogenates of field specimens were prepared as described [12]. Samples were analyzed using a modified AC-ELISA for IBDV that used IBDV specific Mabs [12]. The tests were performed after coating 96-well Immulon 1 plates (Dynatech, Chantilty, VA) with 0.1 ml of 2 ~tg/ml of protein A (PA) from Staphylococcus aureus (Pharmacia, Piscataway, N J). After 18 h at 4 °C, plates were emptied. Next, 1 : 50 dilutions of ascitic fluids containing IBDV-specific Mabs in phosphate-buffered saline (PBS) containing 0.2% of Tween 20 (PBS-T) and 2% nonfat dry milk (NFDM) were added to the wells of different PA coated plates. After 24 h at 4 °C, plates were emptied, and each well was filled with a blocking solution comprised of PBS-NFDM that additionally contained 5 ~g/ml of human IgG (Organon Teknika, Durham, NC). Following blocking, 0.1 ml of 1 : 40 dilutions of bursal homogenates, or 1 : 4 dilutions of reference viruses or vaccine viruses, in PBS-T were added to duplicate wells of each of the test plates. Four wells on each plate were. loaded with 1:40 dilutions of a positive bursal control homogenate consisting of a mixture of STC (Classic strain), E/Del, and GLS IBDVs. Four wells on each plate were loaded with 1 : 40 dilutions of normal bursa homogenates. After 1 h of incubation, plates were washed three times with PBS-T, and then each well received 0.1 ml of polyclonal chicken antiIBDV sera diluted 1 : 150 in PBS-T-NFDM. Plates were emptied and washed after a 1 h incubation. A goat anti-chicken IgG horseradish peroxidase conjugate (Kirkegaard and Perry Laboratories, Gaithersburg, MD) was added to each test well at a concentration of
92
D.B. Snyder et al.
3 lag/ml in PBS-T-NFDM. Following, 1 h of incubation, plates were emptied and washed as before, then 0.1 ml of ABTS substrate (Kirkegaard and Perry) was added. After 15 rain of incubation, tests were read at 405 nm with the aid of an automated 96-well spectrophotometer. Relative group titer levels of 0 to 9 for IBDV antigen for each Mab were calculated after subtracting the average normal bursal control homogenate absorbance value from each duplicate test sample value as described [10]. Values above 2 were considered positive.
Competitive ELISA Competitive ELISAs (C-ELISA) were performed using a modification of the AC-ELISA procedure above. For C-ELISAs, Mab B 29 directed at the 32kDa protein of IBDV [11] was bound to all PA coated plates used. After blocking, plates were reacted with either normal bursal homogenates, or homogenates containing the, STC or GLS viruses. Following this, various combinations of biotin-labeled and non-labeled competing Mabs were simultaneously added to test plates. All biotin-labeled Mabs were used at a concentration of 5 gg/mt and they were competed against constant 1 : 100 dilutions of non-labeled Mabs contained in ascitic fluids. After 1 h of incubation, wells were emptied, washed 3 times, then each well was loaded with a Streptavidin-horseradish peroxidase conjugate in PBST-NFDM (3 gg/ml; Kirkegaard and Perry). Controls consisted of biotinylated Mabs being competed against the AVS Mab which is specific for Newcastle disease virus [13].
Radioimmunoprecipitation assays ( RIPA ) Radioimmunoprecipitation assays were carried out essentially as described [2]. Briefly, D 78, E/Del, and GLS strains of IBDV were used to infect secondary chicken embryo fibroblast cell cultures at an input multiplicity of 0.5 for 30 h. After 1 h of methionine deprivation, cells were labeled with 100 ~tCi of [35S]methionine (specific activity of 1000 Ci/ mmol) per ml of medium for 24 h. The cells were washed in PBS and lysed in a buffer containing 1% sodium deoxycholate, 1% Triton X-100, 150 mM NaC1, 10 mM Tris (pH 7.8), 0.5 mM MgCI2, 0.1% sodium dodecyl sulfate (SDS). The lysates were clarified by centrifugation at 10,000 x g for 10rain and portions were reacted with Mabs or polyclonal antibodies specific for IBDV. Antibody bound proteins were precipitated with Staphylococcus aureus and resolved by electrophoresis on a 12.5% SDS-polyacrylamide gel. Precipitated proteins were detected by autoradiography of dried gets.
Passive immunization of chickens Groups of five 1-day-old specific pathogen free (SPF) chickens (SPAFAS, Inc, Norwich, CT) were inoculated intraperitoneally with 0.5 ml of sterile Mab containing ascitic fluid. The following day, the chicks were challenged with 100 mean-chicken infective doses of either the STC, E/Del, GLS, or DS 326 challenge viruses. Three days later the chickens were euthanized, and the bursa of Fabricius was removed and processed for IBDV antigen detection using the AC-ELISA procedure described above.
Results Production and characterization o f M a b s Sptenocytes o f mice i m m u n i z e d with b o t h the E/Del a n d GLS-5 viruses were used to p r o d u c e h y b r i d o m a cultures secreting Mabs. Initially, s u p e r n a t a n t s were screened for selectivity o f reactivity in A C - E L I S A against the STC (Classic strain), E/Del, GLS, a n d D S 326 I B D V strains, a n d n o r m a l bursal h o m o g e n a t e .
Variants define the epidemiology of I B D V
93
Supernatants of AC-ELISA positive clones were then screened in VN tests for neutralizing activity. Hybridomas 8, 179, 10, 57, and BK9 yielded Mabs that showed either neutralizing activity, strain preference, or both; and they were selected for further analysis and cloned. Also used in this study were three Mabs, R 63, B 69, and B 29 which were previously prepared [11]. After ascites were prepared, each Mab was tested for its ability to precipitate an IBDV antigen in agar gel precipitin tests, and to immunoprecipitate IBDV polypeptides in RIPAs. Table 1 summarizes the immunoreactivities of all anti-IBDV Mabs utilized. A panel of Mabs was prepared that could differentiate among the four selected test viruses by either positive and/or negative reactivity patterns in ACELISA. Mabs B 69 and BK9 were singularly specific for the Classic (STC strain) and E/Del strains respectively, and could identify them by a positive reaction. A positive reaction by Mab 57 could be used to separate the GLS and DS 326 viruses from the Classic and E/Del type viruses. Negative reactions by the 179 Mab could be used in conjunction with a positive reaction by Mab 57 to further differentiate the GLS and DS 326 IBDV isolates from one another. Except for Mabs BK 9 and B 29, Mab analysis in VN tests mirrored AC-
Table 1. Immunoreactivities of monoclonal antibodies (Mab) prepared against infectious bursal disease viruses (IBDV) Mab
Antigen-capture ELISA a IBDV Classic
E/Del
GLS
DS 326
B69 R63 179 8 10 57 BK 9
+ + + + + -
+ + + +
+ + + + .
+ + +
B 29
+
+
+
.
. +
VN tests b
Agargel precipitin tests °
RIPA a
VP 2/VPX site designation e
+ + + + + +
+ + + + + +
VP2/VPX VP2/VPX VP2/VPX VP 2/VPX VP2/VPX VP 2 VPX
I I I I/II II II N.D
-
-
VP 3
N.D.
.
V N Virus neutralization, R I P A radioimmunopredpitation asssay M a b s were evaluated in capture assays against 1:40 dilutions of I B D V containing bursal homogenates. The Classic I B D V was STC b, o A positive result indicates that the M a b precipitated an antigen in or neutralized all of those IBDVs it reacted with in capture assays. A negative result indicates that the M a b lacked any neutralizing or precipitating activity d All M a b s except B K 9 and 57 were reacted in R I P A s with a preparation of the Classic I B D V strain designated as D 78. M a b s B K 9 and 57 were reacted with preparations of the E/De1 and G L S I B D V strains respectively. VP 3 has a molecular weight o f 32 kDa, VPX is 47 kDa, and VP 2 is 42 k D a VP 2/VPX binding site designation was based on competitive inhibition binding assays (Table 2)
94
D.B. Snyder et al.
ELISA results (Table 1). Supernatant and ascites preparations of Mabs neutralized all viruses they reacted with in AC-ELISA. Similarly, neutralizing Mabs all precipitated an IBDV antigen in agar gel precipitin analyses, but nonneutralizing Mabs BK9 and B 29 did not. In RIPAs, Mabs B 69, R 63, 8, 179, and 10 immunoprecipitated the VP2/VPX structural protein of labeled D 78 IBDV, as shown in Fig. 1. Similarly, RIPAs with Mabs R63, BK9, and 57 showed that R 63 precipitated VP 2/VPX protein complexes, while BK 9 reacted only with the VPX protein of labeled E/Del IBDV (Fig. 2 a). Mab 57 precipitated VP 2 of labeled GLS IBDV (Fig. 2 b). The summary of RIPAs with Mabs in Table 1, and Figs. 1 and 2 show that all neutralizing Mabs bound to epitopes on VP 2. One non-neutralizing epitope was identified on VP 2 by Mab BK 9. Mab B29 was previously shown to bind VP3 [11].
Competitive inhibition assays Five of the 6 neutralizing Mabs were successfully labeled with biotin, whereas all attempts to label the B 69 Mab resulted in loss of its binding activity. Table 2 shows the results of the reciprocal C-ELISAs, and evidence that the neutralization epitopes defined with these Mabs are spatially arranged. Epitopes defined by Mabs R63, 179 and B69 are in close proximity to one another (antigenic site I, Table 1). They are linked through a central bridging epitope defined by Mab 8 (antigenic site I/II, Table 1) to two other closely linked epitopes, defined by Mabs 57 and 10 (antigenic site II, Table 1). The epitope defined by Mab B 69 appears to be located near the central epitope overlapped by Mab 8 and Mab 179.
Passive immunization of chickens Chickens passively immunized with neutralizing Mabs one day prior to challenge with virulent challenge viruses were afforded complete protection from viruses they bound in AC-ELISA or VN tests, and no IBDV antigen could be found in the bursa (Table 3). Conversely, chickens passively immunized with nonneutralizing Mabs BK 9 and B 29, or the AVS Mab were not protected upon challenge. The protection afforded by passive immunization was type specific for the challenge IBDV.
Evaluation of laboratory and vaccine strains of IBD V Laboratory and vaccine strains of serotype 1 IBDV, often referred to in literature, or commonly used in the field in the United States, were acquired for analysis by AC-ELISA with the panel of Mabs. Eleven commercially available live IBDV vaccine strains and 10 laboratory reference viruses were assessed without further passage of the original material obtained. Table 4 summarizes the AC-ELISA reactivity of these viruses with the Mabs. Nine of 11 of the
Variants define the epidemiology of IBDV
95
Fig. 1. Radioimunoprecipitation analysis of viral proteins recognized by Mabs using radiolabeled antigens from D 78 IBDV infected cells. Immunoprecipitated products were separated by 12.5% SDS-PAGE. C Mock-infected cells reacted with polyclonal antisera against IBDV; P polyclonal antisera against IBDV; 69 Mab B 69; 63 Mab R63; 8 Mab 8; 179 Mab 179; 10 Mab 10; and N C AVS Mab prepared against Newcastle disease virus Fig. 2. Radioimunoprecipitation analysis of viral proteins recognized by Mabs using radiolabeled antigens from E/Del IBDV (A) or GLS IBDV (B) infected cells. Immunoprecipitated products were separated by 12.5% SDS-PAGE. C Mock-infected cells reacted with polyclonal antisera against IBDV; P polyclonal antisera against IBDV; 69 Mab B 69; 63 Mab R63; 9 Mab BK9; and 57 Mab 57
vaccine viuses reacted similarly with the Mabs and were relegated to one group. Two other vaccine v~iruses reacted differently than the group of 9, and are presented separately for discussion purposes. Laboratory IBDV strains yielded four different sets of reactivity patterns; Classic, Delaware, GLS and DS 326.
96
D.B. Snyder etal. 2. Results of competitive inhibition assays with unlabeled and biotin-labeled neutralizing monoclonal antibodies (Mab)
Table
Biotinlabeled Mab
Percent competition by Mab R 63
179
8
10
57
B 69
R63 179 8
100 94 72 -
57
-
86 76 100 96 95
83 100 97
90 96 100
93 78
10
62 t00 84 52 52
-
AVS a
m
-
a AVS is a Mab prepared against Newcastle disease virus [13] - Value of less than 20% inhibition was observed
Table
3. Protection of 1-day-old chickens from infectious bursal disease virus (IBDV) challenge after passive administration of neutralizing monoclonal antibodies (Mab)
Mab injected
No. protected from challenge IBDV Classica
E/Del
GLS
DS 326
8 179 R 63 B 69 10 57 B 29 BK9 AVS c
5/5 5/5 5/5 4/5 4/5 0/5 0/5 ND 0/5
5/5 5/5 5/5 0/5 0/5 0/5 0/5 0/5 0/5
5/5 5/5 0/5 0/5 4/5 4/5 ND ND 0/5
5/5 0/5 0/5 0/5 5/5 5/5 ND ND 0/5
ND Not a Classic b Classic c AVS is
IBDV immunogen GLS GLS Classicb Classicb GLS GLS Classicb E/Del
done challenge virus was the STC strain immunogen virus was the D 78 strain a Mab prepared against Newcastle disease virus [13]
Geographic distribution of IBD V in the United States I B D V isolates were m a d e t h r o u g h o u t the U n i t e d States f r o m 1988 to 1990. Isolates assessed b y A C - E L I S A were classified into 4 g r o u p s b a s e d on their reactivity with M a b s . T h e typing results o f the 1301 isolates m a d e in 21 p o u l t r y p r o d u c i n g areas are presented in Table 5. Discussion
In o r d e r to address the e p i d e m i o l o g y o f I B D V a n d to o b t a i n an u n d e r s t a n d i n g o f the elements involved in I B D V antigenicity, in this study, we p r e p a r e d and
Variants define the epidemiology of IBDV
97
Table 4. Reactivity of infectious bursal disease virus (IBDV) monoclonal antibodies (Mab) against vaccine and laboratory reference strains of IBDV in antigen capture assays
Mab
IBDV vaccines
IBDV laboratory strains
1-9 a
BVM
1048 E
1-4 b
DeF
GLS-5
DS 326
B29 8 179
+ + +
+ + +
+ + +
+ + +
+ + +
+ + +
+ + -
R63 B69 BK9 57 10
+ + . . +
+ -
+ . . +
+ +
+ +
+
-
+ +
+ +
. .
. . +
.
a Nine vaccine viruses yielded the same reactivity patterns and were grouped. They
were: S-706, BBW, BBI, BUR-l, BUR-2, UNI, BBLEN, BV, and D 78 b Four laboratory reference strains of IBDV yielded the same reactivity patterns and were grouped. They were: IM, STC, MD, and Edgar strains ° Four reference strains of IBDV yielded the same reactivity patterns and were grouped. They were: A/Del, D/Del, E/Del, and G/Del
characterized additional Mabs against the GLS and Delaware viruses. The new Mabs were either type specific for, or neutralized the viruses (Table 1). As we had previously shown with the R 63 and B 69 neutralizing Mabs [11], the new neutralizing Mabs 8, 179, 57, and 10 also precipitated a multivalent antigen in double immunodiffusion analysis with IBDV containing bursal homogenates, whereas non-neutralizing Mabs did not (Table 1). Results of RIPAs (Table 1; Figs. 1 and 2) showed that neutralizing Mabs bound to VP 2, but not VP 3, and this is in agreement with earlier observations [2, 4, 9]. In a previous study, we were not able to map the polypeptide specificities of the conformation dependent B 69 and R 63 Mabs under denaturing conditions using Western immunoblotting [11]. Similar observations have been reported with IBDV Mabs prepared by others [2, 4, 9]. There appears to be a minimum of at least five neutralization epitopes on Classic IBDV strains as defined by Mabs 8, 179, R63, B69, and 10 (Table 1). The "all or none binding" by neutralizing Mabs in AC-ELISA against the various strains of IBDV suggests that each of the epitopes is distinct, and that alterations in one site do not appreciably affect another epitope's ability to bind with another Mab, or afford protection in vivo (Table 3). This is of interest, as reciprocal-cross competitive-inhibition assays suggested that six IBDV neutralization epitopes, defined in this study, are clustered in two or three sets on VP 2, but are tightly linked through a central epitope comprised of the epitope defined by Mab 8 (Table 2). We had previously shown that polyclonal antisera to IBDV competed strongly for binding to IBDV with a biotin-labeled prep-
98
D.B. Snyder et al.
Table 5. Geographic distribution of infectious bursal disease virus (IBDV) antigen types (1988-1990) in the U.S. as defined by analysis with a panel of eight monoclonal antibodies Area~
Percent IBDVtype Classic
Delaware
GLS
DS 326
No. of isolates
Delmarva NC VA GA AL MS AR PA TN MO TX IN FL OH CA OK OR MN WA NE ME
5 t5 16 16 17 19 42 44 50 50 55 57 73 80 84 87 100 100 100 100 100
43 48 67 54 57 61 40 56 50 50 27 43 20 20 8 0 0 0 0 0 0
49 35 16 29 20 19 18 0 0 0 18 0 7 0 8 13 0 0 0 0 0
3 1 0 1 ! 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
610 175 43 83 90 31 115 18 2 2 51 7 15 5 25 8 7 7 3 3 1
Total
254
571
453
23
1301
a All areas are given as State abbreviations except for Delmarva which includes the common areas on the peninsula shared by Delaware, Maryland, and Virginia
aration o f the R 6 3 M a b [11]. In this study, we also c o m p e t e d individual polyclonal antisera prepared against the Classic, Delaware, GLS, and DS 326 strains versus neutralizing Mabs R 63, B 69, 179, and 57, a n d all antisera were able to compete with all Mabs for binding (data n o t shown). These data are considered a reflection of the M a b - M a b competition data (Table 2). They are taken as an indication that the polyclonal antisera to heterologous strains, which inhibited strain specific M a b binding, did so because the strain specific and strain c o m m o n epitopes were so closely spatially arranged. As in this study, other studies have shown that passively administered nonneutralizing Mabs directed at either VP 2 or VP 3 do n o t confer protection against virulent single strain I B D V challenge of chickens, but that neutralizing Mabs directed at VP 2 are protective [2, 4]. However, this study differs from others in that it demonstrates that both strain c o m m o n and strain specific Mabs are singularly protective against challenge. It also shows that passive protection
Variants define the epidemiologyof IBDV
99
is strain specific, mirrors the binding activities of Mabs in VN tests and ACELISA, and provides further evidence that at least five distinct, but closely linked neutralization epitopes play a role in protection of chickens from challenge with Classic strains of IBDV (Tables 1-3). Our interests in this study were not only concerned with defining essential elements of critical areas on IBDV involved in neutralization and in vivo protection, but also, in relating what we had learned and developed from laboratory studies to furthering a better overall understanding of the epidemiology of IBDV in the United States. After preparing additional IBDV Mabs (Table 1), we obtained and evaluated most reference, and live IBDV vaccine serotype 1 IBDV strains that are commonly referred to in literature, employed in the United States as live vaccines, or were isolated by this laboratory and identified as naturally occurring neutralizing Mab escape mutants (Table 411. Prior to beginning a field study, the reference and vaccine IBDV strains were assessed for Mab binding in AC-ELISA (Table 4). Thirteen of the virus strains examined, nine vaccine and four reference strains, had similar reactivity patterns and were so grouped. Most notably, they all reacted with the B 69 Mab and possessed 5 Mab defined neutralization sites. All of those strains were isolated prior to 1985 and were designated as Classic viruses. Although these 13 strains were grouped as Classic viruses, it should be noted that strains, especially the reference strains, are known to vary widely in their virulence and pathogenicity. We have also demonstrated with some of these IBDV strains, that although Mab B 69 is reactive with them in AC-ELISA, they are neutralized differently [12]. The reference variant viruses provided three additional reactivity patterns denoted by the successive loss of Mab defined neutralization sites, Delaware viruses had lost the B 69 site, GLS viruses lacked the B 69 and R63 sites, and the DS326 virus lacked Mab sites B69, R63, and 179. Two other vaccine viruses BVM and BV-4 yielded similar reactivity patterns and neither of the viruses possessed any of the strain specific epitopes defined by Mabs B 69, BK 9, or 57. The reactivity of the BV-4 virus was of interest, because this vaccine virus was derived from the wild type E/Del variant which possesses the BK 9 site, but does not possess the Mab 10 site. The difference in antigenic composition of the wild type and attenuated vaccine type Delaware virus likely occurred during the selection process. The BVM virus originally originated in Europe before 1985, and no comparisons between the vaccine virus and wild type virus could be made. The field survey of IBDV isolates was random and relied heavily on industry cooperation. A total of 21 broiler producing areas were sampled yielding 1301 Mab typeable isolates from different premises. Although sampling was uneven, a distinct geographic distribution pattern was observed which may reflect the viruses antigenic evolution. In isolated and less dense broiler growing areas, such as ME, NE, WA, OR, OK, CA, OH, and FL Classic strains of IBDV were predominant. Those areas contrasted sharply with more dense Eastern areas such as MS, AL, GA, VA, NC, and especially the heavily sampled Del-
100
D.B. Snyder et al.
marva area, where isolation of a Classic IBDV was a more rare event. Other areas, such as AR, PA, TN, MO, TX, and IN, appeared to be more in a transitional stage with respect to their IBDV populations. It was interesting to note that in areas where more than one virus type was circulating, that in only about 5% of all the isolations, were we able to detect more than one IBDV type present in a given poultry house or test sample. It appears that once a given IBDV type is introduced, it establishes itself in a house and becomes prevalent. Several factors may interact to affect the distribution of these viruses. Geographic isolation, even within an area appeared to impact. On Delmarva, the few Classic isolates that were made all came from flocks outlying the main growing areas on the Peninsula. Flock and farm density also appeared to play some role. Delmarva is one of the oldest and most dense broiler chicken growing areas, and other areas of high density in the east and southeast also yielded high numbers of variant IBDV isolates. Selective vaccination pressure may also play a role. Throughout the United States, live IBDV vaccines containing only Classic IBDV strains are used in broiler flocks, and it can be assumed that there is considerable selection pressure (Table 4). However, with only limited numbers of Classic IBDV isolates being made in the eastern United States, it could be concluded that while the vaccines composed of Classic IBDV strains may be i n d u d n g selective pressures, they are safe, and not directly affecting the composition of the field virus populations in those areas. In this paper, we have described the circulation of three naturally occurring neutralizing Mab escape mutants and the apparent sequential antigenic evolution of field strains of IBDV. At present, we know that a minimum of 5 distinct but closely linked neutralization epitopes exist on Classic IBDV strains and that only two of these remain on the most recently isolated DS 326 variant strain. It will be of interest in future studies to see if this virus, or others like it become prevalent in the field.
Acknowledgements This work was supported in part by grants from InterVet International and the Southeastern Egg & Poultry Association. The authors thank Frances S. Yancey, Stephanie Mengel, and Hamp Edwards for technical assistance. Scientific Article No. A 6214, Contribution No. 8383, from the Maryland Agricultural Experiment Station.
References 1. Cosgrove AS (1962) An apparently new disease of chickens-avian nephrosis. Avian Dis 6:385-389 2. Becht H, M/iller H, Miiller H (1988) Comparative studies and antigenic properties of two serotypes of infectious bursal disease virus. J Gen Virol 69:631-640 3. Dobos PB, Hill J, Hallett R, Becht H (1979) Biophysical and biochemical characterization of five animal viruses with bisegrnented double-stranded RNA genomes. J Virol 32:593-605 4. Fahey KJ, McWaters P, Brown MA, Erny K, Murphy VJ, Hewish DR (1991) Virus-
Variants define the epidemiology of IBDV
5. 6. 7. 8.
9. 10.
11. 12.
13. 14.
101
neutralizing and passively protective monoclonal antibodies to infectious bursal disease of chickens. Avian Dis 35:365-373 Ismail NM, Saif YM, Wigle WL, Havenstein GB, Jackson C (1990) Infectious bursal disease virus variant from commercial leghorn pullets. Avian Dis 34:141-145 Jackwood DJ, Saif YM, Hughes JH (1982) Characteristics and serologic studies of two serotypes of infectious bursal disease virus in turkeys. Avian Dis 2,6:871-882 Jackwood DH, Saif YM (1987) Antigenic diversity of infectious bursal disease viruses. Avian Dis 31: 766-770 McFerran JB, McNulty MS, McKillop ER, Connor TJ, McKracken RM, Collins DS, Allan GM (1980) Isolation and serologic studies with infectious bursal disease virus from fowl, turkeys, and ducks: demonstration of a second serotype. Avian Pathol 9: 395-405 Reddy SK, Silim A (1991) Comparison of neutralizing antigens of recent isolates of infectious bursal disease virus. Arch Virol 117:287-296 Snyder DB, Marquardt WW, Mallinson ET, Allen DA, Savage PK (1985) An enzymelinked immunosorbent assay method for the simultaneous measurement of antibody titer to multiple viral, bacterial or protein antigens. Vet Immunol Immunopathol 9: 303-317 Snyder DB, Lana DP, Cho BR, Marquardt WW (1988) Group and strain specific neutralization sites of infectious bursal disease virus defined with monoclonal antibodies. Avian Dis 32:527-534 Snyder DB, Lana DP, Savage PK, Yancey FS, Mengel SA, Marquardt WW (1988) Differentiation of infectious bursal disease viruses directly from infected tissues with neutralizing monoclonal antibodies: evidence of a major antigenic shift in recent field isolates. Avian Dis 32:535-539 Srinivasappa GB, Snyder DB, Marquardt WW, King DJ (1986) Isolation of a monoclonal antibody with specificity for commonly employed vaccine strains of Newcastle disease virus. Avian Dis 30:562-567 Van den Berg TP, Gonze M, Meulmans G (1991) Acute infectious bursal disease virus in poultry: isolation and characterization of a highly virulent strain. Avian Pathol 20: 133-143
Authors' address: D. B. Snyder, Virginia-Maryland Regional College of Veterinary Medicine, University of Maryland, College Park, MD 20742, U.S.A. Received April 16, 1992
Vfrology
Arch Virol (1992) 127:89-101
© Springer-Verlag 1992 Printed in Austria
Naturally occurring-neutralizing monoclonal antibody escape variants define the epidemiology of infectious bursal disease viruses in the United States D. B. Snyder 1' 2, V. N. Vakharia 2' 1, and P. K. Savage a
t Virginia-Maryland Regional College of Veterinary Medicine and 2Center for Agricultural Biotechnology, Maryland Biotechnology Institute, Universtity of Maryland, College Park, Maryland, U.S.A.
Accepted April 20, 1992
Summary. A panel of two non-neutralizing and six neutralizing monoclonal antibodies (Mabs) were used in antigen-capture enzyme immunoassays (ACELISA) to examine the antigenicity of 1301 wild type infectious bursal disease viruses 0BDV) isolated from different poultry flocks througout the United States over a three year period. Analysis of these isolates with protective, neutralizing Mabs directed against the VP 2 structural protein of IBDV showed that four antigenically distinct groups of serotype 1 IBDV could be separated on the basis of the presence or absence of one or more Mab defined, conformation-dependent, multivalent neutralization site. AC-ELISA reactivity patterns of the Mabs with isolates demonstrated that IBDV field populations were relatively antigenically homogeneous per premise isolation. Geographically, various antigenic species were more or less prevalent, or nearly absent. Competition analysis with neutralizing Mabs coupled with AC-ELISA results suggested that neutralization epitopes for IBDV are distinct, spatially arranged, yet closely linked. Of 5 Mab defined neutralization epitopes, shown to be related to protection from virulent challenge by Classic IBDV strains isolated prior to 1985, only two of the epitopes remain unaltered on the most recent emergent variant field strain of IBDV isolated.
Introduction Infectious bursal disease virus (IBDV) is a bi-segmented, double-stranded RNA virus belonging to the family Birnaviridae [3]. The smaller genomic segment encodes VP 1 (98 kDa), the viral polymerase. The larger segment of the genome encodes three proteins VP 2 a/VP 2 b, VP 3, and VP 4, of which VP 2 and VP 3 are structural proteins, while VP 4 is the viral protease [3, 4]. VP 2 a (or VPX)
90
D.B. Snyder et al.
has an approximate molecular weight of 47 kDa, and is a precursor which is subsequently cleaved by VP 4 (28 kDa) to make VP 2 b (42 kDa). VP 2 a and VP 2 b of IBDV have been shown to be bound by neutralizing monoclonal antibodies (Mabs) [2, 4, 9], whereas VP 3 (32 kDa) has not been shown to carry neutralization epitopes [2, 4, 11]. Infection of susceptible chickens with virulent strains of IBDV leads to the rapid accumulation of high levels of viral antigen in the bursa of Fabricius. IBDV infection causes the depletion of lymphoidal follicles of the bursa of Fabricius and spleen, which initiates a highly contagious immunosuppressive disease in chickens [1, 3]. Infections caused by IBDV exacerbate infections produced by other etiologic agents, and thereby reduce the chicken's ability to respond to vaccination. The principle means of controlling infectious bursal disease in chickens are by administration of live attenuated and killed vaccines, and via the passage of high levels of vaccine induced maternal antibodies to young chickens. Two serotypes of IBDV have been defined [6-8]. Serotype 1 has been shown to cause immunologic disease in chickens, while serotpye 2 IBDV only causes sub-acute infections in turkeys. Although all reports are not in taxonomic agreement, laboratory and field strains within serotype 1 have been shown to be antigenically heterogeneous as determined by virus neutralization (VN) tests or by in vivo cross-challenge test [5-8]. In addition, other newly emergent variant strains of serotype 1 also differ pathotypically [14]. Since vaccination is the only method of control, the epidemiology of IBDV remains an important issue. However, virulent IBDV strains often are very difficult to adapt to tissue culture, and when they are adapted, they usually become less virulent. Hence, ultimately the virus adapted and assessed in VN tests may not be representative of the wild type isolate. Compounding the • problems of adaptation to in vitro substrates are the large numbers of isolates that are made and need to be typed. The virus is prevalent and environmentally stable, and when the chickens reach market age, one strain or another has infected virtually all broiler chickens we have ever examined (Snyder unpubl. data). Consequently, large scale typing of IBDV isolates has not previously been undertaken. We have previously characterized Mabs against strains of IBDV [11]. Two of the Mabs, R 63 and B 69, were shown to neutralize standard or Classic strains of IBDV to high titers in VN tests. These Mabs were employed in an antigen capture enzyme-linked immunosorbent assay (AC-ELISA) to examine the antigenicity of wild type IBDV isolates directly from clinically diseased bursal tissues [12]. We reported a pathotypic variant of IBDV could be identified by the apparent deletion or alteration of the neutralization site defined by the B 69 Mab. In this paper, we present the epidemiology of four serotype 1 IBDV field populations found in the United States which exist as naturally-occurring neutralization Mab escape mutants. Apparent progressive mutations occurring in Mab defined neutralization epitopes of field viruses are described. We also show the relative arrangement of six IBDV neutralization epitopes on VP 2 as defined
Variants define the epidemiology of IBDV
91
b y A C - E L I S A a n d c o m p e t i t i o n analysis o f Classic a n d n a t u r a l l y occurring variant strains o f I B D V .
Materials and methods Laboratory viruses, vaccines and fieM isolates Laboratory reference IBDV strains A/Del, D/Del, E/Del, G/Del, Edgar, MD, STC, and IM were obtained from previously acknowledged sources [11]. The prototype GLS IBDV variant was isolated by this laboratory during an earlier field survey in 1987, while the prototype DS326 strain was isolated during the initial stages of this study in 1988. The following commercial IBDV live vaccines were also used: S-706 (Select, Gainesville, GA); BBW and BBI (TRI BIO Gainesville, GA); BUR-1 and BUR-2 (Salisbury, Charles City, IA); UNI (American Scientific, Madison, WI); BBLEN (CEVA, Overland Park, KS); BV, BVM, and BV 4 (Sterwin Millsboro, DE); D 78 (Intervet, Millsboro, DE). In many cases, a commonly referred to laboratory strain may also be a vaccine strain. While differences could exist between the laboratory virus and the vaccine derivatives, we choose the vaccine strain as representative for this epidemiological study because they are licensed biologics used to vaccinate chickens in the field. Field isolates of IBDV were obtained from individual commercial broiler chicken flocks. The isolates were made by obtaining 10 to 15 random bursa of Fabricius samples per flock and processing them as a pool for Mab assessment. Bursal submissions were solicited from commercial broiler companies in major broiler rearing areas throughout the United States, and sampling was conducted over a period of 36 months, extending from 1988 through 1990. Field IBDV isolates were most often made from bursa obtained from commercial broiler type chickens 16 to 28 days of age.
Monoclonal antibodies Murine hybridoma cell lines secreting Mabs against the E/Del and GLS viruses were prepared and characterized in AC-ELISA, VN tests and agar gel precipitin tests as described for the B 29, R 63, and B 69 Mabs which were prepared previously [11].
Antigen-capture ELISA Bursal homogenates of field specimens were prepared as described [12]. Samples were analyzed using a modified AC-ELISA for IBDV that used IBDV specific Mabs [12]. The tests were performed after coating 96-well Immulon 1 plates (Dynatech, Chantilty, VA) with 0.1 ml of 2 ~tg/ml of protein A (PA) from Staphylococcus aureus (Pharmacia, Piscataway, N J). After 18 h at 4 °C, plates were emptied. Next, 1 : 50 dilutions of ascitic fluids containing IBDV-specific Mabs in phosphate-buffered saline (PBS) containing 0.2% of Tween 20 (PBS-T) and 2% nonfat dry milk (NFDM) were added to the wells of different PA coated plates. After 24 h at 4 °C, plates were emptied, and each well was filled with a blocking solution comprised of PBS-NFDM that additionally contained 5 ~g/ml of human IgG (Organon Teknika, Durham, NC). Following blocking, 0.1 ml of 1 : 40 dilutions of bursal homogenates, or 1 : 4 dilutions of reference viruses or vaccine viruses, in PBS-T were added to duplicate wells of each of the test plates. Four wells on each plate were. loaded with 1:40 dilutions of a positive bursal control homogenate consisting of a mixture of STC (Classic strain), E/Del, and GLS IBDVs. Four wells on each plate were loaded with 1 : 40 dilutions of normal bursa homogenates. After 1 h of incubation, plates were washed three times with PBS-T, and then each well received 0.1 ml of polyclonal chicken antiIBDV sera diluted 1 : 150 in PBS-T-NFDM. Plates were emptied and washed after a 1 h incubation. A goat anti-chicken IgG horseradish peroxidase conjugate (Kirkegaard and Perry Laboratories, Gaithersburg, MD) was added to each test well at a concentration of
92
D.B. Snyder et al.
3 lag/ml in PBS-T-NFDM. Following, 1 h of incubation, plates were emptied and washed as before, then 0.1 ml of ABTS substrate (Kirkegaard and Perry) was added. After 15 rain of incubation, tests were read at 405 nm with the aid of an automated 96-well spectrophotometer. Relative group titer levels of 0 to 9 for IBDV antigen for each Mab were calculated after subtracting the average normal bursal control homogenate absorbance value from each duplicate test sample value as described [10]. Values above 2 were considered positive.
Competitive ELISA Competitive ELISAs (C-ELISA) were performed using a modification of the AC-ELISA procedure above. For C-ELISAs, Mab B 29 directed at the 32kDa protein of IBDV [11] was bound to all PA coated plates used. After blocking, plates were reacted with either normal bursal homogenates, or homogenates containing the, STC or GLS viruses. Following this, various combinations of biotin-labeled and non-labeled competing Mabs were simultaneously added to test plates. All biotin-labeled Mabs were used at a concentration of 5 gg/mt and they were competed against constant 1 : 100 dilutions of non-labeled Mabs contained in ascitic fluids. After 1 h of incubation, wells were emptied, washed 3 times, then each well was loaded with a Streptavidin-horseradish peroxidase conjugate in PBST-NFDM (3 gg/ml; Kirkegaard and Perry). Controls consisted of biotinylated Mabs being competed against the AVS Mab which is specific for Newcastle disease virus [13].
Radioimmunoprecipitation assays ( RIPA ) Radioimmunoprecipitation assays were carried out essentially as described [2]. Briefly, D 78, E/Del, and GLS strains of IBDV were used to infect secondary chicken embryo fibroblast cell cultures at an input multiplicity of 0.5 for 30 h. After 1 h of methionine deprivation, cells were labeled with 100 ~tCi of [35S]methionine (specific activity of 1000 Ci/ mmol) per ml of medium for 24 h. The cells were washed in PBS and lysed in a buffer containing 1% sodium deoxycholate, 1% Triton X-100, 150 mM NaC1, 10 mM Tris (pH 7.8), 0.5 mM MgCI2, 0.1% sodium dodecyl sulfate (SDS). The lysates were clarified by centrifugation at 10,000 x g for 10rain and portions were reacted with Mabs or polyclonal antibodies specific for IBDV. Antibody bound proteins were precipitated with Staphylococcus aureus and resolved by electrophoresis on a 12.5% SDS-polyacrylamide gel. Precipitated proteins were detected by autoradiography of dried gets.
Passive immunization of chickens Groups of five 1-day-old specific pathogen free (SPF) chickens (SPAFAS, Inc, Norwich, CT) were inoculated intraperitoneally with 0.5 ml of sterile Mab containing ascitic fluid. The following day, the chicks were challenged with 100 mean-chicken infective doses of either the STC, E/Del, GLS, or DS 326 challenge viruses. Three days later the chickens were euthanized, and the bursa of Fabricius was removed and processed for IBDV antigen detection using the AC-ELISA procedure described above.
Results Production and characterization o f M a b s Sptenocytes o f mice i m m u n i z e d with b o t h the E/Del a n d GLS-5 viruses were used to p r o d u c e h y b r i d o m a cultures secreting Mabs. Initially, s u p e r n a t a n t s were screened for selectivity o f reactivity in A C - E L I S A against the STC (Classic strain), E/Del, GLS, a n d D S 326 I B D V strains, a n d n o r m a l bursal h o m o g e n a t e .
Variants define the epidemiology of I B D V
93
Supernatants of AC-ELISA positive clones were then screened in VN tests for neutralizing activity. Hybridomas 8, 179, 10, 57, and BK9 yielded Mabs that showed either neutralizing activity, strain preference, or both; and they were selected for further analysis and cloned. Also used in this study were three Mabs, R 63, B 69, and B 29 which were previously prepared [11]. After ascites were prepared, each Mab was tested for its ability to precipitate an IBDV antigen in agar gel precipitin tests, and to immunoprecipitate IBDV polypeptides in RIPAs. Table 1 summarizes the immunoreactivities of all anti-IBDV Mabs utilized. A panel of Mabs was prepared that could differentiate among the four selected test viruses by either positive and/or negative reactivity patterns in ACELISA. Mabs B 69 and BK9 were singularly specific for the Classic (STC strain) and E/Del strains respectively, and could identify them by a positive reaction. A positive reaction by Mab 57 could be used to separate the GLS and DS 326 viruses from the Classic and E/Del type viruses. Negative reactions by the 179 Mab could be used in conjunction with a positive reaction by Mab 57 to further differentiate the GLS and DS 326 IBDV isolates from one another. Except for Mabs BK 9 and B 29, Mab analysis in VN tests mirrored AC-
Table 1. Immunoreactivities of monoclonal antibodies (Mab) prepared against infectious bursal disease viruses (IBDV) Mab
Antigen-capture ELISA a IBDV Classic
E/Del
GLS
DS 326
B69 R63 179 8 10 57 BK 9
+ + + + + -
+ + + +
+ + + + .
+ + +
B 29
+
+
+
.
. +
VN tests b
Agargel precipitin tests °
RIPA a
VP 2/VPX site designation e
+ + + + + +
+ + + + + +
VP2/VPX VP2/VPX VP2/VPX VP 2/VPX VP2/VPX VP 2 VPX
I I I I/II II II N.D
-
-
VP 3
N.D.
.
V N Virus neutralization, R I P A radioimmunopredpitation asssay M a b s were evaluated in capture assays against 1:40 dilutions of I B D V containing bursal homogenates. The Classic I B D V was STC b, o A positive result indicates that the M a b precipitated an antigen in or neutralized all of those IBDVs it reacted with in capture assays. A negative result indicates that the M a b lacked any neutralizing or precipitating activity d All M a b s except B K 9 and 57 were reacted in R I P A s with a preparation of the Classic I B D V strain designated as D 78. M a b s B K 9 and 57 were reacted with preparations of the E/De1 and G L S I B D V strains respectively. VP 3 has a molecular weight o f 32 kDa, VPX is 47 kDa, and VP 2 is 42 k D a VP 2/VPX binding site designation was based on competitive inhibition binding assays (Table 2)
94
D.B. Snyder et al.
ELISA results (Table 1). Supernatant and ascites preparations of Mabs neutralized all viruses they reacted with in AC-ELISA. Similarly, neutralizing Mabs all precipitated an IBDV antigen in agar gel precipitin analyses, but nonneutralizing Mabs BK9 and B 29 did not. In RIPAs, Mabs B 69, R 63, 8, 179, and 10 immunoprecipitated the VP2/VPX structural protein of labeled D 78 IBDV, as shown in Fig. 1. Similarly, RIPAs with Mabs R63, BK9, and 57 showed that R 63 precipitated VP 2/VPX protein complexes, while BK 9 reacted only with the VPX protein of labeled E/Del IBDV (Fig. 2 a). Mab 57 precipitated VP 2 of labeled GLS IBDV (Fig. 2 b). The summary of RIPAs with Mabs in Table 1, and Figs. 1 and 2 show that all neutralizing Mabs bound to epitopes on VP 2. One non-neutralizing epitope was identified on VP 2 by Mab BK 9. Mab B29 was previously shown to bind VP3 [11].
Competitive inhibition assays Five of the 6 neutralizing Mabs were successfully labeled with biotin, whereas all attempts to label the B 69 Mab resulted in loss of its binding activity. Table 2 shows the results of the reciprocal C-ELISAs, and evidence that the neutralization epitopes defined with these Mabs are spatially arranged. Epitopes defined by Mabs R63, 179 and B69 are in close proximity to one another (antigenic site I, Table 1). They are linked through a central bridging epitope defined by Mab 8 (antigenic site I/II, Table 1) to two other closely linked epitopes, defined by Mabs 57 and 10 (antigenic site II, Table 1). The epitope defined by Mab B 69 appears to be located near the central epitope overlapped by Mab 8 and Mab 179.
Passive immunization of chickens Chickens passively immunized with neutralizing Mabs one day prior to challenge with virulent challenge viruses were afforded complete protection from viruses they bound in AC-ELISA or VN tests, and no IBDV antigen could be found in the bursa (Table 3). Conversely, chickens passively immunized with nonneutralizing Mabs BK 9 and B 29, or the AVS Mab were not protected upon challenge. The protection afforded by passive immunization was type specific for the challenge IBDV.
Evaluation of laboratory and vaccine strains of IBD V Laboratory and vaccine strains of serotype 1 IBDV, often referred to in literature, or commonly used in the field in the United States, were acquired for analysis by AC-ELISA with the panel of Mabs. Eleven commercially available live IBDV vaccine strains and 10 laboratory reference viruses were assessed without further passage of the original material obtained. Table 4 summarizes the AC-ELISA reactivity of these viruses with the Mabs. Nine of 11 of the
Variants define the epidemiology of IBDV
95
Fig. 1. Radioimunoprecipitation analysis of viral proteins recognized by Mabs using radiolabeled antigens from D 78 IBDV infected cells. Immunoprecipitated products were separated by 12.5% SDS-PAGE. C Mock-infected cells reacted with polyclonal antisera against IBDV; P polyclonal antisera against IBDV; 69 Mab B 69; 63 Mab R63; 8 Mab 8; 179 Mab 179; 10 Mab 10; and N C AVS Mab prepared against Newcastle disease virus Fig. 2. Radioimunoprecipitation analysis of viral proteins recognized by Mabs using radiolabeled antigens from E/Del IBDV (A) or GLS IBDV (B) infected cells. Immunoprecipitated products were separated by 12.5% SDS-PAGE. C Mock-infected cells reacted with polyclonal antisera against IBDV; P polyclonal antisera against IBDV; 69 Mab B 69; 63 Mab R63; 9 Mab BK9; and 57 Mab 57
vaccine viuses reacted similarly with the Mabs and were relegated to one group. Two other vaccine v~iruses reacted differently than the group of 9, and are presented separately for discussion purposes. Laboratory IBDV strains yielded four different sets of reactivity patterns; Classic, Delaware, GLS and DS 326.
96
D.B. Snyder etal. 2. Results of competitive inhibition assays with unlabeled and biotin-labeled neutralizing monoclonal antibodies (Mab)
Table
Biotinlabeled Mab
Percent competition by Mab R 63
179
8
10
57
B 69
R63 179 8
100 94 72 -
57
-
86 76 100 96 95
83 100 97
90 96 100
93 78
10
62 t00 84 52 52
-
AVS a
m
-
a AVS is a Mab prepared against Newcastle disease virus [13] - Value of less than 20% inhibition was observed
Table
3. Protection of 1-day-old chickens from infectious bursal disease virus (IBDV) challenge after passive administration of neutralizing monoclonal antibodies (Mab)
Mab injected
No. protected from challenge IBDV Classica
E/Del
GLS
DS 326
8 179 R 63 B 69 10 57 B 29 BK9 AVS c
5/5 5/5 5/5 4/5 4/5 0/5 0/5 ND 0/5
5/5 5/5 5/5 0/5 0/5 0/5 0/5 0/5 0/5
5/5 5/5 0/5 0/5 4/5 4/5 ND ND 0/5
5/5 0/5 0/5 0/5 5/5 5/5 ND ND 0/5
ND Not a Classic b Classic c AVS is
IBDV immunogen GLS GLS Classicb Classicb GLS GLS Classicb E/Del
done challenge virus was the STC strain immunogen virus was the D 78 strain a Mab prepared against Newcastle disease virus [13]
Geographic distribution of IBD V in the United States I B D V isolates were m a d e t h r o u g h o u t the U n i t e d States f r o m 1988 to 1990. Isolates assessed b y A C - E L I S A were classified into 4 g r o u p s b a s e d on their reactivity with M a b s . T h e typing results o f the 1301 isolates m a d e in 21 p o u l t r y p r o d u c i n g areas are presented in Table 5. Discussion
In o r d e r to address the e p i d e m i o l o g y o f I B D V a n d to o b t a i n an u n d e r s t a n d i n g o f the elements involved in I B D V antigenicity, in this study, we p r e p a r e d and
Variants define the epidemiology of IBDV
97
Table 4. Reactivity of infectious bursal disease virus (IBDV) monoclonal antibodies (Mab) against vaccine and laboratory reference strains of IBDV in antigen capture assays
Mab
IBDV vaccines
IBDV laboratory strains
1-9 a
BVM
1048 E
1-4 b
DeF
GLS-5
DS 326
B29 8 179
+ + +
+ + +
+ + +
+ + +
+ + +
+ + +
+ + -
R63 B69 BK9 57 10
+ + . . +
+ -
+ . . +
+ +
+ +
+
-
+ +
+ +
. .
. . +
.
a Nine vaccine viruses yielded the same reactivity patterns and were grouped. They
were: S-706, BBW, BBI, BUR-l, BUR-2, UNI, BBLEN, BV, and D 78 b Four laboratory reference strains of IBDV yielded the same reactivity patterns and were grouped. They were: IM, STC, MD, and Edgar strains ° Four reference strains of IBDV yielded the same reactivity patterns and were grouped. They were: A/Del, D/Del, E/Del, and G/Del
characterized additional Mabs against the GLS and Delaware viruses. The new Mabs were either type specific for, or neutralized the viruses (Table 1). As we had previously shown with the R 63 and B 69 neutralizing Mabs [11], the new neutralizing Mabs 8, 179, 57, and 10 also precipitated a multivalent antigen in double immunodiffusion analysis with IBDV containing bursal homogenates, whereas non-neutralizing Mabs did not (Table 1). Results of RIPAs (Table 1; Figs. 1 and 2) showed that neutralizing Mabs bound to VP 2, but not VP 3, and this is in agreement with earlier observations [2, 4, 9]. In a previous study, we were not able to map the polypeptide specificities of the conformation dependent B 69 and R 63 Mabs under denaturing conditions using Western immunoblotting [11]. Similar observations have been reported with IBDV Mabs prepared by others [2, 4, 9]. There appears to be a minimum of at least five neutralization epitopes on Classic IBDV strains as defined by Mabs 8, 179, R63, B69, and 10 (Table 1). The "all or none binding" by neutralizing Mabs in AC-ELISA against the various strains of IBDV suggests that each of the epitopes is distinct, and that alterations in one site do not appreciably affect another epitope's ability to bind with another Mab, or afford protection in vivo (Table 3). This is of interest, as reciprocal-cross competitive-inhibition assays suggested that six IBDV neutralization epitopes, defined in this study, are clustered in two or three sets on VP 2, but are tightly linked through a central epitope comprised of the epitope defined by Mab 8 (Table 2). We had previously shown that polyclonal antisera to IBDV competed strongly for binding to IBDV with a biotin-labeled prep-
98
D.B. Snyder et al.
Table 5. Geographic distribution of infectious bursal disease virus (IBDV) antigen types (1988-1990) in the U.S. as defined by analysis with a panel of eight monoclonal antibodies Area~
Percent IBDVtype Classic
Delaware
GLS
DS 326
No. of isolates
Delmarva NC VA GA AL MS AR PA TN MO TX IN FL OH CA OK OR MN WA NE ME
5 t5 16 16 17 19 42 44 50 50 55 57 73 80 84 87 100 100 100 100 100
43 48 67 54 57 61 40 56 50 50 27 43 20 20 8 0 0 0 0 0 0
49 35 16 29 20 19 18 0 0 0 18 0 7 0 8 13 0 0 0 0 0
3 1 0 1 ! 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
610 175 43 83 90 31 115 18 2 2 51 7 15 5 25 8 7 7 3 3 1
Total
254
571
453
23
1301
a All areas are given as State abbreviations except for Delmarva which includes the common areas on the peninsula shared by Delaware, Maryland, and Virginia
aration o f the R 6 3 M a b [11]. In this study, we also c o m p e t e d individual polyclonal antisera prepared against the Classic, Delaware, GLS, and DS 326 strains versus neutralizing Mabs R 63, B 69, 179, and 57, a n d all antisera were able to compete with all Mabs for binding (data n o t shown). These data are considered a reflection of the M a b - M a b competition data (Table 2). They are taken as an indication that the polyclonal antisera to heterologous strains, which inhibited strain specific M a b binding, did so because the strain specific and strain c o m m o n epitopes were so closely spatially arranged. As in this study, other studies have shown that passively administered nonneutralizing Mabs directed at either VP 2 or VP 3 do n o t confer protection against virulent single strain I B D V challenge of chickens, but that neutralizing Mabs directed at VP 2 are protective [2, 4]. However, this study differs from others in that it demonstrates that both strain c o m m o n and strain specific Mabs are singularly protective against challenge. It also shows that passive protection
Variants define the epidemiologyof IBDV
99
is strain specific, mirrors the binding activities of Mabs in VN tests and ACELISA, and provides further evidence that at least five distinct, but closely linked neutralization epitopes play a role in protection of chickens from challenge with Classic strains of IBDV (Tables 1-3). Our interests in this study were not only concerned with defining essential elements of critical areas on IBDV involved in neutralization and in vivo protection, but also, in relating what we had learned and developed from laboratory studies to furthering a better overall understanding of the epidemiology of IBDV in the United States. After preparing additional IBDV Mabs (Table 1), we obtained and evaluated most reference, and live IBDV vaccine serotype 1 IBDV strains that are commonly referred to in literature, employed in the United States as live vaccines, or were isolated by this laboratory and identified as naturally occurring neutralizing Mab escape mutants (Table 411. Prior to beginning a field study, the reference and vaccine IBDV strains were assessed for Mab binding in AC-ELISA (Table 4). Thirteen of the virus strains examined, nine vaccine and four reference strains, had similar reactivity patterns and were so grouped. Most notably, they all reacted with the B 69 Mab and possessed 5 Mab defined neutralization sites. All of those strains were isolated prior to 1985 and were designated as Classic viruses. Although these 13 strains were grouped as Classic viruses, it should be noted that strains, especially the reference strains, are known to vary widely in their virulence and pathogenicity. We have also demonstrated with some of these IBDV strains, that although Mab B 69 is reactive with them in AC-ELISA, they are neutralized differently [12]. The reference variant viruses provided three additional reactivity patterns denoted by the successive loss of Mab defined neutralization sites, Delaware viruses had lost the B 69 site, GLS viruses lacked the B 69 and R63 sites, and the DS326 virus lacked Mab sites B69, R63, and 179. Two other vaccine viruses BVM and BV-4 yielded similar reactivity patterns and neither of the viruses possessed any of the strain specific epitopes defined by Mabs B 69, BK 9, or 57. The reactivity of the BV-4 virus was of interest, because this vaccine virus was derived from the wild type E/Del variant which possesses the BK 9 site, but does not possess the Mab 10 site. The difference in antigenic composition of the wild type and attenuated vaccine type Delaware virus likely occurred during the selection process. The BVM virus originally originated in Europe before 1985, and no comparisons between the vaccine virus and wild type virus could be made. The field survey of IBDV isolates was random and relied heavily on industry cooperation. A total of 21 broiler producing areas were sampled yielding 1301 Mab typeable isolates from different premises. Although sampling was uneven, a distinct geographic distribution pattern was observed which may reflect the viruses antigenic evolution. In isolated and less dense broiler growing areas, such as ME, NE, WA, OR, OK, CA, OH, and FL Classic strains of IBDV were predominant. Those areas contrasted sharply with more dense Eastern areas such as MS, AL, GA, VA, NC, and especially the heavily sampled Del-
100
D.B. Snyder et al.
marva area, where isolation of a Classic IBDV was a more rare event. Other areas, such as AR, PA, TN, MO, TX, and IN, appeared to be more in a transitional stage with respect to their IBDV populations. It was interesting to note that in areas where more than one virus type was circulating, that in only about 5% of all the isolations, were we able to detect more than one IBDV type present in a given poultry house or test sample. It appears that once a given IBDV type is introduced, it establishes itself in a house and becomes prevalent. Several factors may interact to affect the distribution of these viruses. Geographic isolation, even within an area appeared to impact. On Delmarva, the few Classic isolates that were made all came from flocks outlying the main growing areas on the Peninsula. Flock and farm density also appeared to play some role. Delmarva is one of the oldest and most dense broiler chicken growing areas, and other areas of high density in the east and southeast also yielded high numbers of variant IBDV isolates. Selective vaccination pressure may also play a role. Throughout the United States, live IBDV vaccines containing only Classic IBDV strains are used in broiler flocks, and it can be assumed that there is considerable selection pressure (Table 4). However, with only limited numbers of Classic IBDV isolates being made in the eastern United States, it could be concluded that while the vaccines composed of Classic IBDV strains may be i n d u d n g selective pressures, they are safe, and not directly affecting the composition of the field virus populations in those areas. In this paper, we have described the circulation of three naturally occurring neutralizing Mab escape mutants and the apparent sequential antigenic evolution of field strains of IBDV. At present, we know that a minimum of 5 distinct but closely linked neutralization epitopes exist on Classic IBDV strains and that only two of these remain on the most recently isolated DS 326 variant strain. It will be of interest in future studies to see if this virus, or others like it become prevalent in the field.
Acknowledgements This work was supported in part by grants from InterVet International and the Southeastern Egg & Poultry Association. The authors thank Frances S. Yancey, Stephanie Mengel, and Hamp Edwards for technical assistance. Scientific Article No. A 6214, Contribution No. 8383, from the Maryland Agricultural Experiment Station.
References 1. Cosgrove AS (1962) An apparently new disease of chickens-avian nephrosis. Avian Dis 6:385-389 2. Becht H, M/iller H, Miiller H (1988) Comparative studies and antigenic properties of two serotypes of infectious bursal disease virus. J Gen Virol 69:631-640 3. Dobos PB, Hill J, Hallett R, Becht H (1979) Biophysical and biochemical characterization of five animal viruses with bisegrnented double-stranded RNA genomes. J Virol 32:593-605 4. Fahey KJ, McWaters P, Brown MA, Erny K, Murphy VJ, Hewish DR (1991) Virus-
Variants define the epidemiology of IBDV
5. 6. 7. 8.
9. 10.
11. 12.
13. 14.
101
neutralizing and passively protective monoclonal antibodies to infectious bursal disease of chickens. Avian Dis 35:365-373 Ismail NM, Saif YM, Wigle WL, Havenstein GB, Jackson C (1990) Infectious bursal disease virus variant from commercial leghorn pullets. Avian Dis 34:141-145 Jackwood DJ, Saif YM, Hughes JH (1982) Characteristics and serologic studies of two serotypes of infectious bursal disease virus in turkeys. Avian Dis 2,6:871-882 Jackwood DH, Saif YM (1987) Antigenic diversity of infectious bursal disease viruses. Avian Dis 31: 766-770 McFerran JB, McNulty MS, McKillop ER, Connor TJ, McKracken RM, Collins DS, Allan GM (1980) Isolation and serologic studies with infectious bursal disease virus from fowl, turkeys, and ducks: demonstration of a second serotype. Avian Pathol 9: 395-405 Reddy SK, Silim A (1991) Comparison of neutralizing antigens of recent isolates of infectious bursal disease virus. Arch Virol 117:287-296 Snyder DB, Marquardt WW, Mallinson ET, Allen DA, Savage PK (1985) An enzymelinked immunosorbent assay method for the simultaneous measurement of antibody titer to multiple viral, bacterial or protein antigens. Vet Immunol Immunopathol 9: 303-317 Snyder DB, Lana DP, Cho BR, Marquardt WW (1988) Group and strain specific neutralization sites of infectious bursal disease virus defined with monoclonal antibodies. Avian Dis 32:527-534 Snyder DB, Lana DP, Savage PK, Yancey FS, Mengel SA, Marquardt WW (1988) Differentiation of infectious bursal disease viruses directly from infected tissues with neutralizing monoclonal antibodies: evidence of a major antigenic shift in recent field isolates. Avian Dis 32:535-539 Srinivasappa GB, Snyder DB, Marquardt WW, King DJ (1986) Isolation of a monoclonal antibody with specificity for commonly employed vaccine strains of Newcastle disease virus. Avian Dis 30:562-567 Van den Berg TP, Gonze M, Meulmans G (1991) Acute infectious bursal disease virus in poultry: isolation and characterization of a highly virulent strain. Avian Pathol 20: 133-143
Authors' address: D. B. Snyder, Virginia-Maryland Regional College of Veterinary Medicine, University of Maryland, College Park, MD 20742, U.S.A. Received April 16, 1992
