Antigenic diversity of Akabane virus detected by monoclonal antibodies

Virus Research 47 (1997) 187 – 196

Antigenic diversity of Akabane virus detected by monoclonal antibodies H. Akashia,*, Y. Inabab b

a National Institute of Animal Health, 3 -1 -1, Kannondai, Tsukuba, Ibaraki 305, Japan Department of Epizootiology, College of Bioresource Sciences, Nihon Uni6ersity, Kanagawa 252, Japan

Received 14 August 1996; accepted 6 November 1996

Abstract The antigenic properties of 21 Japanese field isolates and two Australian strains of Akabane (AKA) virus (Simbu serogroup, bunyavirus) isolated from 1959 to 1990 were compared by enzyme-linked immunosorbent assay (ELISA), plaque-reduction neutralization (PRNT) and hemagglutination inhibition (HI) tests using monoclonal antibodies (Mabs) to the OBE-1 strain of AKA virus. Sixteen Mabs were established by fusing P3X63Ag8U1 mouse myeloma cells and spleen cells from BALB/c mice immunized with the OBE-1 strain. Of the 16 clones, 13 produced immunoglobulin (Ig) which precipitated glycoprotein G1 and three produced Ig which precipitated nucleoprotein (N). Twelve out of 13 Mabs had both NT and HI activities to not only the homologous OBE-1 strain but also the other isolates. By the competitive binding assay, at least five antigenic regions for G1, and two for N were defined. Some of the anti-G1 Mabs which reacted to the same antigenic region had unique reactivity while anti-N Mabs recognizing the same epitope reacted with almost the same degree to all of the isolates. Finally, nine epitopes of the G1 protein in five different antigenic regions have been identified. There was no striking correlation between isolation date and place of the isolates and their reactivity to Mabs. A most interesting result is that three isolates collected in the same place over a three week period had different reactivity patterns detected by ELISA, showing great antigenic variation of the virus. AKA virus may be a single gene pool consisting of different genotypes in the field. © 1997 Elsevier Science Ireland Ltd. Keywords: Akabane virus; Antigenicity; Monoclonal antibody; Virus mutation

1. Introduction

* Corresponding author. Tel.: 81 298 387761; fax: 81 298 387880; e-mail: [email protected]

Bunyaviruses have a genome consisting of three unique species of single stranded (ss) negative sense RNA, designated L (large), M (medium)

0168-1702/97/$17.00 © 1997 Elsevier Science Ireland Ltd. All rights reserved. PII S 0 1 6 8 - 1 7 0 2 ( 9 6 ) 0 1 4 1 5 - 3

H. Akashi, Y. Inaba / Virus Research 47 (1997) 187–196

188

Table 1 Akabane virus isolates used in this experiment Isolates

Year of isolation

Place of isolation (area)

Isolation from

Passage history

JaGAr39 B8935 R7949 OBE-1 NBE-9 SAB’74 KT3377 M171 KM-29X82 KC-12X84 KC-15X84 KC-04Y84 Iriki NS-88-1 NS-88-2 YG-88-2 ON-89-2 MZ-90-1 MZ-90-2 FO-90-3 FO-90-4 KS-90-2 KS-90-3

1959 1968 1968 1974 1974 1974 1977 1979 1982 1984 1984 1984 1984 1988 1988 1988 1989 1990 1990 1990 1990 1990 1990

Gunma (C)a Australia Australia Okayama (W) Niigata (C) Okayama (W) Kagoshima (S) Ibaraki (C) Kagoshima (S) Kagoshima (S) Kagoshima (S) Kagoshima (S) Kagoshima (S) Nagasaki (S) Nagasaki (S) Yamaguchi (W) Okinawa (S) Miyazaki (S) Miyazaki (S) Fukuoka (W) Fukuoka (W) Kagoshima (S) Kagoshima (S)

Aedes 6exans Culicoides bre6itarsis Culicoides bre6itarsis Aborted fetus brain Aborted fetus brain Aborted fetus brain Blood from infected cattle Culex tritaeniorhynchus Culex tritaeniorhynchus Culicoides oxystoma Culicoides oxystoma Culicoides oxystoma Affected calf cerebellum Blood from infected cattle Blood from infected cattle Blood from infected cattle Blood from infected cattle Culicoides species Culicoides species Culicoides species Culicoides species Blood from infected cattle Blood from infected cattle

SM18 SM3 SM4 SM1 SM2 SM3 SM2 SM1 SM3 SM2 SM2 SM2 HL9 HL4 HL4 HL5 HL5 HL1, Vero 3 HL1, Vero 3 HL1 HL3, Vero 1 HL7 HL7

a

C, central; W, western; S, southern.

and S (small). Genetic studies have shown that the S RNA codes for a nucleoprotein (N), and a non-structural protein (NSs). The function of NSs is still unknown. The M RNA codes for the two envelope glycoproteins, G1 and G2, and a nonstructural protein (NSm). The L RNA encodes the large virion protein (L) which has transcriptase activity (Bishop, 1990; Elliott, 1990; Elliott et al., 1991). The epizootics of abortion and congenital arthrogryposis-hydranencephaly (AH) syndrome observed among cattle in Japan in 1972 – 74 have been shown to be caused by Akabane (AKA) virus, a member of the Simbu serogroup, genus Bunya6irus, family Bunyaviridae (Inaba et al., 1975; Kurogi et al., 1976). Since the JaGAr39 strain of AKA virus was first isolated from Aedes 6exans and Culex tritaeniorhynchus mosquitoes in Gunma Prefecture in 1959 (Oya et al., 1961), many strains of the virus have been isolated from biting midges and cattle blood specimens mostly in the central and southern parts of Japan

(Kurogi et al., 1987, 1978, 1976). Although these isolates cannot be differentiated from the original JaGAr39 strain by neutralization (NT) test, the data obtained by the technique of RNAse T1 oligonucleotide fingerprint have revealed that no two isolates exhibited identical fingerprints (unpublished data). Thus, to know to what extent variation of the viral genome is reflected in the antigenic properties, monoclonal antibodies (Mabs) to AKA virus were prepared and tested serologically with 21 Japanese field isolates and two strains of Australian AKA virus.

2. Materials and methods

2.1. Viruses and cells The viruses used in this study are listed in Table 1. The viruses were collected between 1959 and 1990, from central to southern Japan. However, the viruses were isolated mainly in Kyushu Island

H. Akashi, Y. Inaba / Virus Research 47 (1997) 187–196

189

(southern Japan). Each virus was grown three times and plaque-cloned once in monolayers of HmLu-1 cells (Kurogi et al., 1977) after the passage levels shown in Table 1. Virus inoculation and plaque formation were carried out by essentially the same methods as described in previous papers (Kurogi et al., 1976, 1977) with the addition of DEAE-dextran (50 mg/ml) in the overlay medium for plaque assay. To obtain large quantities of virus, HmLu-1 cells were grown in roller bottles with growth medium consisting of Eagle’s minimal essential medium (EMEM). They were supplemented with 10% tryptose phosphate broth (TPB) and 5% calf serum, and then infected with virus at a multiplicity of about 0.01. After virus adsorption, maintenance medium consisting of EMEM containing 10% TPB, 0.1% yeast extract, 0.5% sodium glutamate, and 0.5% glucose was added. The infected cells were incubated at 37°C for 2–3 days, then the supernatant fluid was harvested.

with a minor modification (Sato et al., 1991). Briefly, 6-week-old BALB/c mice were immunized intraperitoneally (i.p.) with the two step purified OBE-1 strain of AKA virus, followed by another injection i.p. 4 weeks later. The mice were sacrificed 3 days after the last injection. The spleen cells were fused to P3X63Ag8U1 mouse myeloma cells, at a ratio of 5:1. The supernatant of hybrid cells was screened by ELISA for antibody to the OBE-1 strain. Positive cells were cloned once in 0.6% soft agar. To obtain a large amount of antibodies, hybridomas were propagated in tissue culture without serum, and the immunoglobulin (Ig) from the culture fluid, was concentrated ten fold by precipitation in half-saturated ammonium sulfate solution, or propagated as ascitic fluid in pristane-primed BALB/c mice. The isotype of Mabs was determined by the MONOAB-ID kit (Zymed, USA).

2.2. Virus purification

2.4. ELISA

The infectious fluid was clarified by low speed centrifugation, mixed with 0.025 M zinc acetate solution (Yoshinaka and Hotta, 1971), and then stirred at 4°C for 20 min after adjustment of pH to 7.2 with 1 N NaOH. The mixture was centrifuged at 1000 g for 15 min and the resulting pellet was resuspended with saturated disodium EDTA. The virus suspension was then centrifuged at 28 000 rpm for 2 h. The virus pellet was resuspended with 0.01 M Tris – HCl buffer (pH 7.4), containing 0.15 M NaCl and 0.001 M EDTA (TEN). It was purified by banding in a 30% glycerol–50% potassium tartrate density gradient with centrifugation for 2 h at 38 000 rpm in a SW41 rotor. The virus band was diluted with TEN buffer and loaded onto a 10-70% sucrose gradient and centrifuged at 38 000 rpm for 2 h. The virus band was harvested and dialyzed with TEN buffer.

Plastic 96-well flat plates were seeded with 100 ml/well of purified viral antigen diluted in 0.05 M carbonate buffer (pH 9.6), and held at 4°C overnight. The plates were washed four times with washing saline (0.85% NaCl containing 0.02% Tween 20). Test antibodies were diluted in ELISA buffer (0.1 M KH2PO4, 0.5 M NaCl, 0.5% Tween 20 and 0.5% ovalbumin adjusted to pH 7.2 with NaOH) and 50 ml was added to the appropriate wells. After incubation at 37°C for 1 h, the plates were washed as above. Fifty microliter of peroxidase-conjugated goat anti-mouse Ig (light and heavy chain-specific, Cappel, USA), diluted 500 fold with ELISA buffer, was added to each well. The plates were incubated at room temperature for 30 min. After washing, 100 ml of substrate (100 mM Na2HPO4, 50 mM citric acid, pH 4.8 and 0.4 mg/ml of o-phenylenediamine containing 0.2 ml/ml of 30% H2O2) was added. A brown color developed in 20 min at room temperature in the dark. The enzyme reaction was stopped by adding 100 ml of 3 N H2SO4. The plates were read at 490 nm in SJeia auto reader (Sanko Junyaku, Japan).

2.3. Monoclonal antibodies Monoclonal antibody production followed the procedure reported by Ko¨hler and Milstein (1975)

190

H. Akashi, Y. Inaba / Virus Research 47 (1997) 187–196

2.5. Determination of ELISA antigen and Mab titers Antigen was titrated by the ‘checkerboard’ technique with polyclonal mouse antiserum to the OBE-1 strain. Four units of antigen were used in determining the ELISA antibody titer of the Mabs. The antigen and antibody titer was expressed as the highest dilution showing optical density (OD) over 0.5.

2.8. Plaque reduction neutralization (PRNT) and hemagglutination inhibition (HI) tests

2.6. Competiti6e binding assay The competitive binding assay was performed by an avidin-biotin labelled ELISA system. One milligram of each Mab purified from mouse ascitic fluid was biotinylated with N-hydroxysuccinimidobiotin (Pierce Chemicals, USA) by the method of Guesdon et al. (1979). Briefly 100 ml of 4 U of the OBE-1 strain antigen was coated to wells at 4°C overnight. After washing with washing saline, serial five-fold dilutions for each competing non-labelled Mabs were added to the antigen coated wells and the plates were incubated at 37°C for 1 h. Biotin-labelled Ig diluted to show an OD of about 1.0 in the absence of competing antibody was added after washing the wells. After incubation at 37°C for 1 h, peroxidase-conjugated avidin (Vector, USA) and substrate were added sequentially as described in the standard ELISA method. The percentage of competitive binding was calculated as follows, in which A is OD490 in the absence of antibody, B is OD490 in the presence of homologous antibody, and N is OD490 in the presence of competitor. Percent of competition = 100 ×

lyzed fetal bovine serum and 1.85 MBq [35S]methionine, and incubated for 4 h. The concentrated Mabs were mixed with the labelled virus infected cell lysate and then protein A-sepharose CL4B by the method described elsewhere (Ito et al., 1990; Sato et al., 1991). After centrifugation, precipitated polypeptide preparations were resolved on 10% polyacrylamide gels electrophoresis containing sodium dodecyl sulfate (SDS-PAGE).

A−N A−B

2.7. Immunoprecipitation To label viral proteins, HmLu-1 cells were infected with the OBE-1 strain at a multiplicity of five, and incubated for 13 h at 34°C. Then the infected cells were washed three times with methionine-free EMEM, the medium was replaced with methionine-free EMEM containing 2% dia-

The PRNT test was performed as previously described (Kurogi et al., 1976) with a minor modification by addition of DEAE-dextran in overlay medium. HI test was carried out by essentially the same method as described elsewhere (Goto et al., 1978) using purified virion antigen obtained from the infected cell culture fluid.

3. Results

3.1. Characterization of the Mabs Sixteen hybridomas were established from 76 ELISA positive cell clones from a fusion of mouse myeloma cells and the spleen cells from mice immunized with the OBE-1 strain of AKA virus. Specificity of the Mabs was demonstrated by ELISA using purified virions from the virus infected cell culture supernatant and the virus inoculated mouse brains (data not shown). To determine the recognition protein of Mabs, immunoprecipitation analyses have been performed. Thirteen of 16 Mabs precipitated the 120 kDa major polypeptide, G1. The remaining three Mabs precipitated a 26 kDa polypeptide, N. Anti-G1 Mabs were identified as IgG2a (11) and IgG3 (2). Only Mab 2F1 lacked NT activity, while all the Mabs had HI activity (Table 2). When antibodies from mouse ascitic fluid were used, 9G12 showed HI activity but 2F1 did not neutralize the OBE-1 strain. Three anti-N Mabs were all IgG2a. Anti-N Mabs showed neither NT nor HI activity.

H. Akashi, Y. Inaba / Virus Research 47 (1997) 187–196

191

Table 2 Characterization of monoclonal antibodies (Mabs) Clone no.

1B3 1D3 2B12 2F1 3A5 4A10 4D4 4F7 4F9 5E5 5G4 9G10 9G12 2A7 5E8 5F11

Ig isotype

G2a G2a G2a G2a G2a G3 G3 G2a G2a G2a G2a G2a G2a G2a G2a G2a

Protein specifity

G1 G1 G1 G1 G1 G1 G1 G1 G1 G1 G1 G1 G1 N N N

Homologous titera ELISA

50% PRNT

HI

12 800 51 200 25 600 6400 6400 25 600 12 800 12 800 51 200 6400 25 600 6400 320 102 400 12 800 25 600

4096 1024 32 768 B4 4096 4096 128 512 32 768 1024 4 4096 4 B4 B4 B4

128 1024 512 4 512 512 64 256 256 128 256 256 B2 B2 B2 B2

a

Titers of Mabs to OBE-1 strain in ELISA, 50% plaque reduction neutralization (PRNT) and hemagglutination inhibition (HI) tests.

3.2. Competiti6e binding assay The competitive binding assay was performed to map the epitopes of AKA OBE-1 strain. The results are summarized in Table 3. Fifty percent of competition was observed when 4.5 (2B12) to 160 ng (4D4) of the homologous antibody was used as a competitor (data not shown). The results of the competitive binding assay clearly show that anti-G1 Mabs were assigned to five (A–E) topologically distinct antigenic regions and three anti-N Mabs were divided into two (N-A and N-B) groups (Table 4). Mabs belonging to Group A (1B3 and 4F9), Group C (2B12 and 3A5) and Group E (9G10 and 9G12) showed 85 – 100% competition with each other within the same group, but were not inhibited with a Mab in the other group. On the contrary, although Mabs in Group B (1D3 and 4F7) competed 100% with each other, 1B3 showed weak competition (about 40%) with those of Group B. The degree of competition among Mabs in Group D varied. Mabs 4A10, 4D4 and 5G4 showed complete reciprocal competition, while 5E5 and 2F1 showed

one-way competition with one of the antibodies in the group.

3.3. Reacti6ity pattern of the isolates with Mabs Results for 23 AKA isolates tested by ELISA with Mabs are presented in Table 4. Among the 23 isolates, none except FO-90-3 and FO-90-4 exhibited the same pattern of reactivity to antiG1 Mabs. However, some isolates have very similar reactivity patterns to Mabs, i.e. JaGAr39, M171, FO-90-3 and FO-90-4; KC12X84, KC-04Y84 and KS-90-3; KM-29X82, MZ-90-1 and Iriki. There was no clear association between the reactivity of isolates to Mabs and the date and place of virus isolation. Whereas Mabs belonging to Group A and C, and three Mabs in Group D (4A10, 4D4 and 5G4) showed reciprocal competition with each other, each AKA isolate reacted to these antibodies in a different manner. This suggests that an antigenic region may contain overlapping epitopes. Antigenic regions A and D contained two and four epitopes, respectively. Finally we recognized nine epitopes in five antigenic regions

400 B10 B10 51 200 B10 B10 800 400 400 B10 B10 B10 B10 1600 1600 20 3200 20 20 400 800 10 160

B10 B10 B10 12 800 B10 B10 B10 B10 B10 B10 B10 B10 B10 B10 B10 B10 B10 B10 B10 B10 B10 B10 B10

JaGAr39 B8935 R7949 OBE-1 NBE-9 SAB’74 KT3377 M171 KM-29X82 KC-12X84 KC-15X84 KC-04Y84 Iriki NS-88-1 NS-88-2 YG-88-2 ON-89-2 MZ-90-1 MZ-90-2 FO-90-3 FO-90-4 KS-90-2 KS-90-3

a

lD3

A2 4F9

A1 lB3

Isolates 12 800 12 800 25 600 12 800 6400 6400 12 800 12 800 6400 25 600 25 600 25 600 12 800 6400 6400 6400 51 200 12 800 B10 51 200 25 600 B10 51 200

4F7 12 800 1600 12 800 25 600 6400 6400 400 12 800 6400 12 800 25 600 25 600 25 600 12 800 12 800 10 102 400 25 600 12 800 102 400 102 400 102 400 102 400

2B12

C

6400 1600 6400 6400 1600 1600 10 6400 3200 6400 25 600 12 800 51 200 6400 3200 10 25 600 12 800 3200 25 600 25 600 6400 25 600

3A5

Antigenic regions are determined by the results of competitive binding assay.

25 600 12 800 51 200 51 200 12 800 12 800 25 600 25 600 25 600 51 200 51 200 25 600 51 200 51 200 12 800 6400 51 200 12 800 10 51 200 25 600 B10 51 200

B

A

Antigenic regiona

Table 4 ELISA tests with Akabane virus monoclonal antibodies

3200 800 6400 25 600 1600 12 800 40 12 800 10 3200 B10 3200 B10 80 40 40 1600 160 400 800 1600 200 800

D1 4A10

D

3200 B10 10 12 800 1600 B10 1600 6400 800 6400 6400 3200 1600 1600 800 1600 6400 800 160 3200 3200 400 1600

D2 4D4 12 800 B10 10 25 600 6400 B10 6400 25 600 3200 12 800 400 12 800 400 6400 1600 3200 12 800 6400 6400 12 800 25 600 800 12 800

D2 5G4 1600 1600 6400 6400 1600 1600 B10 3200 B10 B10 6400 B10 B10 160 40 20 25 600 20 40 1600 1600 200 20

D3 5E5 1600 1600 12 800 6400 400 400 6400 6400 1600 6400 64 00 6400 25 600 25 600 20 800 25 600 12 800 160 25 600 25 600 6400 12 800

D4 2F1 3200 B10 12 800 6400 1600 1600 3200 12 800 1600 12 800 12 800 6400 12 800 3200 400 200 6400 1600 1600 3200 6400 800 3200

9G10

E

160 B10 320 320 80 160 160 320 40 320 320 320 160 80 20 80 320 160 160 160 640 160 160

9G12

51 200 51 200 51 200 102 400 25 600 51 200 102 400 102 400 51 200 102 400 102 400 102 400 102 400 51 200 25 600 51 200 51 200 102 400 51 200 51 200 102 400 102 400 102 400

2A7

N–A

12 800 6400 12 800 25 600 12 800 12 800 12 800 25 600 12 800 25 600 25 600 25 600 12 800 51 200 12 800 25 600 25 600 25 600 12 800 12 800 51 200 12 800 25 600

5F11

6400 12 800 12 800 12 800 6400 12 800 6400 12 800 3200 12 800 12 800 12 800 6400 3200 3200 6400 B10 6400 6400 6400 12 800 6400 6400

5E8

N–B

192 H. Akashi, Y. Inaba / Virus Research 47 (1997) 187–196

H. Akashi, Y. Inaba / Virus Research 47 (1997) 187–196

193

Table 3 Results of competitive binding assay using OBE-1 strain of Akabane virus

on the G1 molecule of AKA virus. In contrast, all anti-N Mabs reacted with almost the same ELISA titer as that of the homologous OBE-1 strain whereas 5E8 did not react with ON-89-2 (Table 4).

3.4. Comparison of reacti6ity patterns in ELISA, PRNT and HI tests Nine AKA isolates including the homologous OBE-1 strain have been tested with 11 neutralizing Mabs in PRNT and HI tests, then compared with the results of ELISA. The reactivity of the isolates to the Mabs detected by each serological test was expressed as a ratio calculated as the homologous titer/the highest titer which the Mab showed with the isolates used. There was a good correlation among the results of ELISA, PRNT and HI tests. However, some isolates were not neutralized by one or two Mabs although those Mabs reacted with the isolate very strongly in ELISA and HI tests (Fig. 1).

4. Conclusion Bunyaviruses have four structural proteins (L, G1, G2 and N) and two non-structural proteins. In our experiments we recognized three structural viral proteins of AKA virus which have 120, 35 and 26 kDa molecular weights. Three proteins were presumed to be G1, G2 and N, respectively, by the data of McPhee and Della-Porta (1988). To analyze the function of major virion proteins Mabs have been established to several bunyaviruses (Blackburn et al., 1987; Franko et al., 1983; Gonzalez-Scarano et al., 1982; Grady et al., 1983; Kingsford et al., 1983; Najjar et al., 1985; Pifat et al., 1988), but none to Simbu serogroup virus has been reported. Few papers have compared the antigenicity of bunyaviruses using Mabs (Artsob et al., 1992; Kingsford et al., 1983). Sixteen clones of hybridoma cells which secrete antibody to AKA virus were established. Of the 16 clones, 13 produced antibody to the G1 protein and the remaining three produced antibody to the N protein. The competitive binding

194

H. Akashi, Y. Inaba / Virus Research 47 (1997) 187–196

Fig. 1. A comparison of anti-G1 monoclonal antibody titers obtained in ELISA (E), plaque neutralization (N) and hemagglutination inhibition (H) tests using nine Akabane isolates. Mabs 2F1 and 9G12 were not included in this comparison because of non-neutralizing activity and low avidity to the virus, respectively.

assay suggested that at least five independent antigenic sites exist on G1 and at least two on N. For the G1 protein of La Crosse virus (LAC; California serogroup), several epitopes have been found on the molecule. Kingsford et al. (1983); Kingsford and Boucquey (1990) reported that 23 Mabs to the G1 of LAC bound to 16 epitopes located within eight antigenic regions identified by a competitive binding assay. On the other hand, Najjar et al. (1985) defined 11 epitopes to five antigenic regions which were interrelated and might be part of a single immunodominant antigenic site. In our experiments we defined nine epitopes to five topologically different antigenic regions, and these results are very similar to those of Najjar et al. We only found one non-neutralizing Mab (2F1) to the G1 protein of AKA virus, while this Mab showed HI activity. Other groups have reported several non-neutralizing Mabs to the LAC G1 (GonzalezScarano et al., 1982; Grady et al., 1983; Kingsford et al., 1983). These disparity in the number of

non-neutralizing Mabs might explain the differences on the number of epitopes between our data and those of Kingsford et al. (1983). The reactivity of AKA isolates to Mabs established by immunization with the OBE-1 strain varied (Table 4). Only the OBE-1 strain responded with Mab 1B3, thus suggesting that its epitope is absent from the other viruses studied. On the other hand, Mab 4F9, which recognizes the same antigenic site (A) as 1B3, reacted weakly to some isolates. This region consists of overlapping epitopes (A1 and A2). Antibodies to other regions (D) also showed differing reactivity patterns, thus suggesting the existence of four overlapping epitopes (D1–4). There was no obvious correlation between the date and place of isolation and their reactivity to Mabs. This might suggest that antigenic, probably genetic mutation, occurs randomly for AKA virus in the field. Reviews of several workers (Beaty and Bishop, 1988; Calisher, 1988; Kingsford, 1991) suggest that the

H. Akashi, Y. Inaba / Virus Research 47 (1997) 187–196

evolution of bunyaviruses due to genetic mutation (genetic drift) and reassortment (shift) is caused by the virus replication cycle alternating between vertebrate and arthropod hosts. A most interesting finding obtained from the ELISA tests with isolates is that three isolates (KC-12X84, KC15X84 and KC-04Y84), collected from Culicoides species in the same place over a 3 week period, had different reactivity patterns, showing great antigenic variation. AKA virus may be a single gene pool consisting of different genotypes in the field. Our findings add to the evolution theory for bunyaviruses. Of 13 anti-G1 Mabs, 12 had both NT and HI activities to the homologous OBE-1 strain. There was a good correlation between the results of all three serological tests (Fig. 1). These facts suggest that the G1 protein acts as the viral hemagglutinin, and the domain which binds to erythrocytes overlaps with the domain which binds to cells, confirming previous work (Gonzalez-Scarano et al., 1982; Grady et al., 1983; Kingsford et al., 1983). In contrast, Mab 2F1 had HI but no NT activity. A few antibodies having these characteristics were also reported by others (GonzalezScarano et al., 1982; Grady et al., 1983), indicating that the binding sites to erythrocytes and cells may differ somewhat in structure or function. This hypothesis is supported by the facts obtained from the comparison of reactivity patterns of the Mabs in three serological tests. We found that some Mabs responded to the isolates in ELISA and HI test, but these failed to neutralize the viruses. The results suggest that a part of the binding site to erythrocytes plays a role in binding to virus susceptible cells.

Acknowledgements We are very grateful to K. Yatsuhashi and K. Terui for their technical help. We would like to thank Drs Ian M. Personson and Peter D. Kirkland for their critical reading and helpful comments on the manuscript. This work was supported by grants from the Ministry of Agriculture, Forestry and Fisheries, Japan.

195

References Artsob, H., Spence, L., Brodeur, B.R. and Th’ng, C. (1992) Monoclonal antibody characterization of Jamestown Canyon (California serogroup) virus topotypes isolated in Canada. Viral Immunol. 5, 233 – 242. Beaty, B.J. and Bishop, D.H.L. (1988) Bunyavirus-vector interactions. Virus Res. 10, 289 – 302. Bishop, D.H.L. (1990) Bunyaviridae and their replication. Part I: Bunyaviridae. In: B.N. Fields, D.M. Knipe (Eds.), Virology, 2nd edn. Raven Press, New York, pp. 1155 – 1173. Blackburn, N.K., Besselaar, R.G., Shepherd, A.J. and Swanepoel, R. (1987) Preparation and use of monoclonal antibodies for identifying Crimean-Congo hemorrhagic fever virus. Am. J. Trop. Med. Hyg. 37, 392 – 397. Calisher, C.H. (1988) Evolutionary significance of the taxonomic data regarding bunyaviruses of the family Bunyaviridae. Intervirology 29, 268 – 276. Elliott, R.M. (1990) Molecular biology of the Bunyaviridae. J. Gen. Virol. 71, 501 – 522. Elliott, R.M., Schmaljohn, C.S. and Collett, M.S. (1991) Bunyaviridae genome structure and gene expression. Curr. Top. Microbiol. Immunol. 169, 91 – 141. Franko, M.C., Gibbs, C.J., Lee, P.W. and Gajkusek, D.C. (1983) Monoclonal antibodies specific for Hantaan virus. Proc. Natl. Acad. Sci. 80, 4149 – 4153. Gonzalez-Scarano, F., Shope, R.E., Calisher, C.E. and Nathanson, N. (1982) Characterization of monoclonal antibodies against the G1 and N proteins of La Crosse and Tahyna, two California serogroup Bunyaviruses. Virology 120, 42 – 53. Goto, Y., Inaba, Y., Miura, Y., Kurogi, H., Takahashi, E., Sato, K., Omori, T., Hanaki, T., Sazawa, H. and Matumoto, M. (1978) Hemagglutination-inhibition test applied to the study of Akabane virus infection in domestic animals. Vet. Microbiol. 3, 89 – 99. Grady, L.J., Srihongse, S., Grayson, M.A. and Deibel, R. (1983) Monoclonal antibodies against La Crosse virus. J. Gen. Virol. 64, 1699 – 1704. Guesdon, J-L., Ternynck, T. and Avrameas, S. (1979) The use of avidin-biotin interaction in immunoenzymatic techniques. J. Histochem. Cytochem. 27, 1131 – 1139. Inaba, Y., Kurogi, H. and Omori, T. (1975) Akabane disease: Epizootic abortion, premature birth, stillbirth and congenital arthrogryposis-hydranencephaly in cattle, sheep and goat caused by Akabane virus. Aust. Vet. J. 51, 584 – 585. Ito, T., Akashi, H., Yamakawa, M., Kurogi, H. and Inaba, Y. (1990) Structural polypeptides of a Palyam serogroup orbivirus isolated in Japan. Aust. Vet. J. 67, 153 – 154. Kingsford, L. (1991) Antigenic variance. Curr. Top. Microbiol. Immunol. 169, 181 – 216. Kingsford , L. and Boucquey, K.H. (1990) Monoclonal antibodies specific for the G1 glycoprotein of La Crosse virus that react with other California serogroup viruses. J. Gen. Virol. 71, 523 – 530. Kingsford, L., Ishizawa, L.D. and Hill, D.W. (1983) Biological activities of monoclonal antibodies reactive with antigenic

H. Akashi, Y. Inaba / Virus Research 47 (1997) 187–196

196

sites mapped on the G1 glycoprotein of La Crosse virus. Virology 129, 443–455. Ko¨hler, G. and Milstein, C. (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256, 495 – 497. Kurogi, H., Akiba, K., Inaba, Y. and Matumoto, M. (1987) Isolation of Akabane virus from the biting midge Culicoides oxystoma in Japan. Vet. Microbiol. 15, 243–248. Kurogi, H., Inaba, Y., Takahashi, E., Sato, K., Akashi, H., Satoda, K. and Omori, T. (1978) Pathogenicity of different strains of Akabane virus for mice. Nat. Inst. Anim. Health Q. 18, 1 – 7. Kurogi, H., Inaba, Y., Takahashi, E., Sato, K., Omori, T., Miura, Y., Goto, Y., Fujiwara, Y., Hatano, Y., Kodama, K., Fukuyama, S., Sasaki, N. and Matumoto, M. (1976) Epizootic congenital arthrogryposis-hydranencephaly syndrome in cattle: Isolation of Akabane virus from affected fetuses. Arch. Virol. 51, 67–74. Kurogi, H., Inaba, Y., Takahashi, E., Sato, K., Satoda, K. and Omori, T. (1977) Development of Akabane virus and its immunogen in HmLu-1 cell culture. Nat. Inst. Anim. Health Q. 17, 27–28.

.

McPhee, D.A. and Della-Porta, A.J. (1988) Biochemical and serological comparisons of Australian bunyaviruses belonging to the Simbu serogroup. J. Gen. Virol. 69, 1007 – 1017. Najjar, J.A., Gentsch, J.R., Nathanson, N. and GonzalezScarano, F. (1985) Epitopes of the G1 glycoprotein of La Crosse virus form overlapping clusters within a single antigenic site. Virology 144, 426 – 432. Oya, A., Okuno, T., Ogata, T., Kobayashi, I. and Matsuyama, T. (1961) Akabane, a new arborvirus isolated in Japan. Jpn. J. Med. Sci. Biol. 14, 101 – 108. Pifat, D.Y., Osterling, M.C. and Smith, J.F. (1988) Antigenic analysis of Punta Toro virus and identification of protective determinants with monoclonal antibodies. Virology 167, 442 – 450. Sato, M., Akashi, H., Hirai, S., Kimura, Y. and Takahashi, M. (1991) Production of monoclonal antibodies against the Kakegawa strain of bovine coronavirus and their characterization. J. Vet. Med. Sci. 53, 147 – 148. Yoshinaka, Y. and Hotta, S. (1971) Purification of arboviruses grown in tissue culture. PSEBM 137, 1047 – 1053.

.