Journal of Medical Virology 32:31-38 (1990)
Defined Epitope Blocking With Murray Valley Encephalitis Virus and Monoclonal Antibodies: Laboratory and Field Studies R.A. Hawkes, J.T. Roehrig, C.R. Boughton, H.M. Naim, R. Orwell, and P. Anderson-Stuart Schools of Microbiology and Medicine, University o f New South Wales, Kensington (R.A.H., C.R.B., H.M.N.), Faculty of Life Sciences, University of Technology, St. Leonards (R.O.),and Bioclone Laboratories, Marrickville (P.A.S.), Australia; Division of Vector-Borne Viral Diseases, Centers for Disease Control, Fort Collins, Colorado (J.T.R.) In an attempt to develop a specific serological test for Murray Valley encephalitis (MVE) virus antibodies, a panel of MVE monoclonal antibodies was utilised in defined-epitope blocking ELlSA tests. In sera of mice immunised singly and in combinations of MVE, Alfuy (ALF), and Kunjin (KUN) viruses, blocking patterns usually distinguished MVE infections from those of the other flaviviruses. When blocking tests with selected MAbs were applied to 468 flavivirus antibody positive sera collected from human subjects throughout New South Wales, sera with blocking patterns consistent with previous MVE infection were found in 18 subjects. All were long-term residents of areas previously frequented by MVE, and all were of an age to have been exposed to the virus in past epidemics. No such sera were found in subjects living in coastal areas of NSW where MVE has never been reported.
KEY WORDS: flavivirus, serology, epidemiolW Y
INTRODUCTION Murray Valley encephalitis (MVE) virus is a n arthropod-borne f lavivirus which periodically causes serious epidemics of encephalitis in south-eastern Australia, and which also causes sporadic cases in northern Australia and Papua New Guinea. Laboratory diagnosis and knowledge of the natural history of the virus depend chiefly upon serological avenues, and on virus isolation from arthropods. Unfortunately serological approaches to MVE diagnosis and epidemiology are complicated by the fact that extensive antigenic cross reactivity occurs within Flaviviridae. At least 6 flaviviruses (MVE, Kunjin (KUN), Alfuy (ALF), Stratford (STR), Kokobera (KOK), and Edge Hill (EH) viruses) have been isolated on the Australian mainland [Karabatsos, 19851. With the conventional 0 1990 WILEY-LISS, INC.
haemagglutination-inhibition (HI) and neutralisation (NT) tests, especially the former, infection with one or more of these viruses may induce antibodies detectable with MVE virus. Hence a need exists to provide a serological test which is specific for MVE. The recent development of a panel of monoclonal antibodies (MAbs) some of which are uniquely specific for MVE [Hawkes et al., 19881 provided a n opportunity to design and assess a method for detection of specific MVE antibody, uncomplicated by cross-reacting antibody induced by other flaviviruses. The approach adopted was modelled on a defined-epitope blocking enzyme-linked immunosorbent assay (DEB-ELISA) used to distinguish dengue virus infections from those caused by Japanese encephalitis (JE) virus [Burke et al., 19871. The present study is presented in 2 phases. Firstly the evaluation of candidate MAbs was undertaken, using a panel of antisera produced by immunisation and hyperimmunisation of mice with MVE and 2 other Australian flaviviruses. The second phase consisted of applying the optimised method with selected MAbs to a large panel of human sera from flavivirus antibody positive subjects from New South Wales (NSW) collected during 1981 and 1982.
MATERIALS AND METHODS Viruses All viruses were from mouse-adapted strains of: MVE strain 96961/53; Kunjin (KUN) strain MRM16; Alfuy (ALF) strain MRM3929; Sepik (SEP) strain MK7148; Saumarez Reef (SRE) strain CSIRO 4; Stratford (STR) strain C338; Kokobera (KOK) strain MRM32; Edge Hill (EH) strain C281 [for details of passage histories, see Hawkes et al., 19851.
Accepted for publication May 25, 1990. Address reprint requests to Dr. R.A. Hawkes, School of Microbiology, University of New South Wales, P.O. Box 1,Kensington, NSW 2033 Australia.
Hawkes et al.
Haemagglutinating (HA) and ELISA Antigens HA antigens of infective suckling mouse brain material were prepared by acetone-sucrose extraction [Shope and Sather, 19791. The antigen used in the DEB-ELISA test consisted of HA antigen, diluted as described below. Laboratory-Produced Antisera Female mice (QS strain) 4-5 weeks of age were inoculated intraperitoneally (IP) with 0.2 ml volumes of 10% suckling mouse brain suspension of either MVE, ALF, or KUN viruses. All 3 viruses are within the same antigenic complex of Flaviviridae, but ALF is more closely related to MVE than is KUN virus [Calisher et al., 19891. Groups of mice were terminally bled a t either 2 or 4 weeks following the initial injection. In addition, other groups of mice were injected I P with 0.2 ml of the same or a different flavivirus 2 weeks following the first injection, and terminally bled 2 weeks later. Sera from all the single-dose (2 week and 4 week post-inoculum) and double-dose mice (4 weeks postfirst inoculum) were separated from the clot and stored a t -20°C.
TABLE I. Characteristics of Flavivirus MAbs Used in This Study Elicititinp virus MVE MVE MVE MVE MVE MVE SLE
I
MAb 4B6C-2 4B5A-2 4B1D-3 4B3B-6 4B6B-10 4A1B-9 6B6C-1
Epitope” E-lc E-ld
Reactivityb MVE MVE
E-5 E-6 E-4b
MVE, J E , Usutu MVE, J E , Alfuy All flaviviruses
~
“For details of epitope distribution see Hawkes et al. [19881. bReactivity as determined in direct binding ELISA assays [Hawkeset al., 19881.
Flaviviridae, was also used [Roehrig et al., 19831. The antigenic reactivities of each of the MAbs is given in Table I. MAbs, 4B6C-2, 4B5A-2, 4BID-3, 4B3B-6 and 4B6B-10, are referred to as “specific” MAbs in future discussion. For use in the DEB-ELISA test, each MAb was purified and coupled to horseradish peroxidase (HRP) by established methods [Wilson and Nakane, 19781.
The Defined-Epitope Blocking Enzyme-Linked Immunosorbent Assay (DEB-ELISA) Briefly stated, the assay assumes that the population of antibodies elicited by infection with a flavivirus will contain sub-populations with a range of differing specificities. Some antibodies will react only with the uniquely specific epitopes of the eliciting virus, some will react with epitopes common to several closely related members of Flaviviridae, and some will react with epitopes common to all members of the family. Monoclonal antibodies, by definition, are capable of Serological Tests binding to one epitope only. In principle then, a monoThe haemagglutination-inhibition (HI) test was per- clonal antibody to a n epitope which is shared among all formed by a standard method [Shope and Sather, 19791. flaviviruses such as anti-E-4b ought to be inhibited in The neutralisation (NT) test employed serial 4-fold di- its binding to MVE virus by group-reactive antibody lutions of heat-inactivated sera (56”C, 30 minutes), which has developed in response to any flaviviral inmixed in duplicate, with equal volumes (100 p1) of a fection. At the other extreme, MAbs reactive with calculated 100 plaque-forming units (pfu) of MVE or MVE-specific epitopes, such as anti-E-lc or anti-E-ld, KUN virus. After a n incubation of 1hour at 37”C, 104.7 ought to be blocked only by antibodies elicited by MVE cells per well in 500 p1 of a continuous porcine cell line infection. Between these extremes are MAbs which re(PS clone D) in growth medium were added, followed 2 act with epitopes shared by MVE and a few other hours later by a final overlay of 1.5% carboxymethyl closely related flaviviruses, such as anti-E-5 or anticellulose in growth medium. After 4-5 days incubation E-6, which might be expected to be inhibited by antia t 37”C, plaques were stained with naphthalene black. bodies elicited only by MVE virus and the other closely Titres of neutralising antibody were established by the related flaviviruses. The assay used was a modification of that recently serum dilution effecting a 70% reduction in control plaque count. The test was carried out in 24-well Cos- employed by Burke e t al. [19871. tar plates (cat. 3524). 1. Polystyrene wells (Dynatech Immulon 2 ReMonoclonal Antibodies Used in DEB-ELISA Test movawell strips) were sensitised with either a) 100 p1 The panel of monoclonal antibodies to MVE virus of a 1 5 0 0 dilution of a globulin extract derived from used in these studies (MVE 4BID-3 excepted) has been pooled human sera with high HI antibody titres to all described elsewhere [Hawkes, 19881. In addition, the Australian flaviviruses or b) 100 p.1 of a 1:2,500 diluMAb SLE 6B6C-1, induced by St. Louis encephalitis tion of purified monoclonal antibody JE6B4A-10 [Roe(SLE) virus, and widely reactive to all members of hrig, 19861. Either antibody when diluted in alkaline Human Survey Sera A previous survey of about 17,000 subjects throughout New South Wales [Boughton et al., 1984; Hawkes e t al., 19851 provided 468 human sera possessing HI antibodies to one or more flaviviruses. Most biophysical regions of the State were well represented. Flavivirus antibody negative sera from subjects living in areas not prone to arboviruses were used as negative controls to standardise the DEB-ELISA test, when human sera were being assayed.
Defined Epitope Blocking With MVE Virus
33
carbonate buffer efficiently captures MVE and closely studies of blocking of different MAbs with such sera related viruses onto the solid phase. After 4 hours in- difficult to evaluate. As expected, all 28 day post-injection sera, regardless cubation at room temperature, the capture antibody was discarded and 200 pl of 0.5% bovine serum albu- of the immunising virus, possessed antibodies which min (BSA) in phosphate-buffered saline (PBS-BSA) were able to combine with flavivirus group-reactive was added. After 2 hours at room temperature, the epitopes, and thus block the 2 broadly reactive antiplates were washed with PBS containing 0.05% Tween E-4b MAbs in the DEB-ELISA test. (These 2 MAbs behaved in a similar fashion throughout the whole 20. 2. Fifty haemagglutinating units of MVE acetone- study.) With the MVE antisera, the DEB-ELISA blocking sucrose extracted viral antigen in 25 pl of 90% PBSBSA and 10% acetone extracted normal mouse serum titres against the more specific epitopes tended to be (AENMS) were added. The plate was incubated over- similar to those shown against the group-reactive night at 4°C to allow attachment of the virus to anti- epitope. This implied the existence in MVE antisera of body on the solid phase and then washed and drained. substantial levels of MVE (type)-specific antibodies. It was evident that antibodies elicited by ALF and 3. The blocking step. Twenty-five microlitres of untreated test serum, diluted in PBS-BSA, was added and KUN infections of some mice also possessed the ability to block the attachment of MAbs which had been reacleft for 2 hours a t room temperature. 4. Without removing the test serum, 25 p1 of HRP- tive with only the MVE type-specific epitopes in direct conjugated MAb, diluted in 90% PBS-BSA/lO% binding assays (Table I). This anomalous finding was AENMS, was added. Each conjugate used had been di- more marked with ALF sera than with KUN sera, parluted to achieve approximately the same optical den- alleling the closer antigenic relationship which ALF sity when reacted against 50 HA of un-blocked MVE has to MVE virus [Calisher et al., 19891. Even with antigen. After a n additional 1hour incubation at room ALF sera, however, the titre of blocking antibodies to temperature, the plates were again washed and thor- the specific MAbs was often significantly lower than t h a t shown to the group-reactive (E-4b) epitope. Unexoughly drained. 5. The substrate 3,3’,5,5’-tetramethylbenzidinepectedly, the MAb enabling the best discrimination be(TMB) in citrate acetate buffer pH 6.0 was added in 100 tween MVE and ALF was the anti-E-6 which reacts pl volumes to each well. The reaction was stopped after with both MVE and ALF in direct binding assays (Ta3 minutes by the addition of 50 pl of 2M H2S0,, and the ble I). Anti-E-ld, anti-E-lc, and anti-E-5 MAbs were less optical density (OD) was measured at 450 nm. Sera were usually screened at a dilution of 15or 1:lO. Serial able to discriminate between MVE antisera and those 2-fold dilutions of positive sera were then tested. Sig- induced by a single injection of the other flaviviruses. Antisera from mice injected with 2 doses of vinificant blocking of epitopes of MVE was deemed to have occurred when the OD of the test sample was less rus. Surveys done in New South Wales [Hawkes et al., than half that shown by a panel of normal antibody 19851 have indicated that a minor proportion of the negative sera of the same species as that under test. human population experiences multiple sequential flaPreliminary experiments indicated that no improve- vivirus infections. In a n attempt to mimic this situament in the sensitivity or specificity of the assay could tion experimentally, mice were injected, a t a n interval be gained by the use of untreated 10% brain suspension of 2 weeks, with a second dose of the same or a different of MVE as antigen, or by raising the temperature of the flavivirus. Five of the possible 9 sequential combinations were used, namely MVE/MVE, ALF/ALF, KUN/ blocking step to 37°C. KUN, ALF/KUN; KUNIALF. (It was of some interest to know whether sequential infections with 2 non-MVE viruses could simulate the blocking pattern generated RESULTS by MVE virus itself.) DEB-ELISA and HI results on Evaluation of the DEB-ELISA With the 28 day sera are shown in Table 111. Mouse Antisera As in the singly injected mice (Table 11), all sera As a prerequisite to field testing, sera from immu- blocked group E-4b epitopes. Sera from mice injected nised and hyperimmunised mice were tested for their twice with MVE possessed type-specific antibodies caability to block the attachment of HRP-conjugated pable of blocking the more specific MAbs, usually to a MVE MAbs of varying specificities, to MVE. The similar titre as the E-4b group-reactive MAbs. The sera broadly reacting MAb SLE 6B6C-1 (antiE-4b) was also from non-MVE immunisation schedules also possessed antibodies to the more MVE-specific epitopes but, alused for comparison. Antisera from mice irqjected with one dose of vi- most without exception the blocking titres were signifrus. Mice were sacrificed at 14 and 28 days after a icantly lower than those to the anti-E-4b MAbs. Although not inferior to the other MAbs in this resingle injection with either MVE, ALF, or KUN viruses. DEB-ELISA and HI results are detailed on Table spect, anti-E-6 MAb was not markedly superior in be2. In some of the 14 day sera, DEB-ELISA antibody ing able to distinguish MVE infection sera from those titres were low or undetectable, making comparative elicited with other flaviviruses.
Hawkes et al.
34
TABLE 11. DEB-ELESA and HI Results on Antisera From Mice Given One I.P. Injection of 0.2 ml. Living 10%smb Suspension of Virus and Sampled 14 or 28 Days Later* DEB-ELISA titre blocking anti-epitope MAb Virus injected MVE
Days post injection 14
MVE
28
ALF
14
ALF
28
KUN
14
KUN
28
Mouse No. 1 2 3 1 2 3 4 1 2 3 1 2 3 4 5 1 2 3 1 2 3 4 5
MVE 10 10 40 640 21,280 640 21,280 40 20 20 320 160 320 160 160 40 20 40 640 320 640 320 320
HI titre vs ALF KUN 10 10 10 10 10 10 21,280 320 21,280 640 21,280 320 21,280 1,280 80 80 20 10 40 10 21,280 40 640 40 21,280 80 640 80 640 40 40 160 40 80 80 21,280 21,280 21,280 21,280 640 21,280 21,280 640 640 640 640
E-ld (4B5A-2)
Defined Epitope Blocking With Murray Valley Encephalitis Virus and Monoclonal Antibodies: Laboratory and Field Studies R.A. Hawkes, J.T. Roehrig, C.R. Boughton, H.M. Naim, R. Orwell, and P. Anderson-Stuart Schools of Microbiology and Medicine, University o f New South Wales, Kensington (R.A.H., C.R.B., H.M.N.), Faculty of Life Sciences, University of Technology, St. Leonards (R.O.),and Bioclone Laboratories, Marrickville (P.A.S.), Australia; Division of Vector-Borne Viral Diseases, Centers for Disease Control, Fort Collins, Colorado (J.T.R.) In an attempt to develop a specific serological test for Murray Valley encephalitis (MVE) virus antibodies, a panel of MVE monoclonal antibodies was utilised in defined-epitope blocking ELlSA tests. In sera of mice immunised singly and in combinations of MVE, Alfuy (ALF), and Kunjin (KUN) viruses, blocking patterns usually distinguished MVE infections from those of the other flaviviruses. When blocking tests with selected MAbs were applied to 468 flavivirus antibody positive sera collected from human subjects throughout New South Wales, sera with blocking patterns consistent with previous MVE infection were found in 18 subjects. All were long-term residents of areas previously frequented by MVE, and all were of an age to have been exposed to the virus in past epidemics. No such sera were found in subjects living in coastal areas of NSW where MVE has never been reported.
KEY WORDS: flavivirus, serology, epidemiolW Y
INTRODUCTION Murray Valley encephalitis (MVE) virus is a n arthropod-borne f lavivirus which periodically causes serious epidemics of encephalitis in south-eastern Australia, and which also causes sporadic cases in northern Australia and Papua New Guinea. Laboratory diagnosis and knowledge of the natural history of the virus depend chiefly upon serological avenues, and on virus isolation from arthropods. Unfortunately serological approaches to MVE diagnosis and epidemiology are complicated by the fact that extensive antigenic cross reactivity occurs within Flaviviridae. At least 6 flaviviruses (MVE, Kunjin (KUN), Alfuy (ALF), Stratford (STR), Kokobera (KOK), and Edge Hill (EH) viruses) have been isolated on the Australian mainland [Karabatsos, 19851. With the conventional 0 1990 WILEY-LISS, INC.
haemagglutination-inhibition (HI) and neutralisation (NT) tests, especially the former, infection with one or more of these viruses may induce antibodies detectable with MVE virus. Hence a need exists to provide a serological test which is specific for MVE. The recent development of a panel of monoclonal antibodies (MAbs) some of which are uniquely specific for MVE [Hawkes et al., 19881 provided a n opportunity to design and assess a method for detection of specific MVE antibody, uncomplicated by cross-reacting antibody induced by other flaviviruses. The approach adopted was modelled on a defined-epitope blocking enzyme-linked immunosorbent assay (DEB-ELISA) used to distinguish dengue virus infections from those caused by Japanese encephalitis (JE) virus [Burke et al., 19871. The present study is presented in 2 phases. Firstly the evaluation of candidate MAbs was undertaken, using a panel of antisera produced by immunisation and hyperimmunisation of mice with MVE and 2 other Australian flaviviruses. The second phase consisted of applying the optimised method with selected MAbs to a large panel of human sera from flavivirus antibody positive subjects from New South Wales (NSW) collected during 1981 and 1982.
MATERIALS AND METHODS Viruses All viruses were from mouse-adapted strains of: MVE strain 96961/53; Kunjin (KUN) strain MRM16; Alfuy (ALF) strain MRM3929; Sepik (SEP) strain MK7148; Saumarez Reef (SRE) strain CSIRO 4; Stratford (STR) strain C338; Kokobera (KOK) strain MRM32; Edge Hill (EH) strain C281 [for details of passage histories, see Hawkes et al., 19851.
Accepted for publication May 25, 1990. Address reprint requests to Dr. R.A. Hawkes, School of Microbiology, University of New South Wales, P.O. Box 1,Kensington, NSW 2033 Australia.
Hawkes et al.
Haemagglutinating (HA) and ELISA Antigens HA antigens of infective suckling mouse brain material were prepared by acetone-sucrose extraction [Shope and Sather, 19791. The antigen used in the DEB-ELISA test consisted of HA antigen, diluted as described below. Laboratory-Produced Antisera Female mice (QS strain) 4-5 weeks of age were inoculated intraperitoneally (IP) with 0.2 ml volumes of 10% suckling mouse brain suspension of either MVE, ALF, or KUN viruses. All 3 viruses are within the same antigenic complex of Flaviviridae, but ALF is more closely related to MVE than is KUN virus [Calisher et al., 19891. Groups of mice were terminally bled a t either 2 or 4 weeks following the initial injection. In addition, other groups of mice were injected I P with 0.2 ml of the same or a different flavivirus 2 weeks following the first injection, and terminally bled 2 weeks later. Sera from all the single-dose (2 week and 4 week post-inoculum) and double-dose mice (4 weeks postfirst inoculum) were separated from the clot and stored a t -20°C.
TABLE I. Characteristics of Flavivirus MAbs Used in This Study Elicititinp virus MVE MVE MVE MVE MVE MVE SLE
I
MAb 4B6C-2 4B5A-2 4B1D-3 4B3B-6 4B6B-10 4A1B-9 6B6C-1
Epitope” E-lc E-ld
Reactivityb MVE MVE
E-5 E-6 E-4b
MVE, J E , Usutu MVE, J E , Alfuy All flaviviruses
~
“For details of epitope distribution see Hawkes et al. [19881. bReactivity as determined in direct binding ELISA assays [Hawkeset al., 19881.
Flaviviridae, was also used [Roehrig et al., 19831. The antigenic reactivities of each of the MAbs is given in Table I. MAbs, 4B6C-2, 4B5A-2, 4BID-3, 4B3B-6 and 4B6B-10, are referred to as “specific” MAbs in future discussion. For use in the DEB-ELISA test, each MAb was purified and coupled to horseradish peroxidase (HRP) by established methods [Wilson and Nakane, 19781.
The Defined-Epitope Blocking Enzyme-Linked Immunosorbent Assay (DEB-ELISA) Briefly stated, the assay assumes that the population of antibodies elicited by infection with a flavivirus will contain sub-populations with a range of differing specificities. Some antibodies will react only with the uniquely specific epitopes of the eliciting virus, some will react with epitopes common to several closely related members of Flaviviridae, and some will react with epitopes common to all members of the family. Monoclonal antibodies, by definition, are capable of Serological Tests binding to one epitope only. In principle then, a monoThe haemagglutination-inhibition (HI) test was per- clonal antibody to a n epitope which is shared among all formed by a standard method [Shope and Sather, 19791. flaviviruses such as anti-E-4b ought to be inhibited in The neutralisation (NT) test employed serial 4-fold di- its binding to MVE virus by group-reactive antibody lutions of heat-inactivated sera (56”C, 30 minutes), which has developed in response to any flaviviral inmixed in duplicate, with equal volumes (100 p1) of a fection. At the other extreme, MAbs reactive with calculated 100 plaque-forming units (pfu) of MVE or MVE-specific epitopes, such as anti-E-lc or anti-E-ld, KUN virus. After a n incubation of 1hour at 37”C, 104.7 ought to be blocked only by antibodies elicited by MVE cells per well in 500 p1 of a continuous porcine cell line infection. Between these extremes are MAbs which re(PS clone D) in growth medium were added, followed 2 act with epitopes shared by MVE and a few other hours later by a final overlay of 1.5% carboxymethyl closely related flaviviruses, such as anti-E-5 or anticellulose in growth medium. After 4-5 days incubation E-6, which might be expected to be inhibited by antia t 37”C, plaques were stained with naphthalene black. bodies elicited only by MVE virus and the other closely Titres of neutralising antibody were established by the related flaviviruses. The assay used was a modification of that recently serum dilution effecting a 70% reduction in control plaque count. The test was carried out in 24-well Cos- employed by Burke e t al. [19871. tar plates (cat. 3524). 1. Polystyrene wells (Dynatech Immulon 2 ReMonoclonal Antibodies Used in DEB-ELISA Test movawell strips) were sensitised with either a) 100 p1 The panel of monoclonal antibodies to MVE virus of a 1 5 0 0 dilution of a globulin extract derived from used in these studies (MVE 4BID-3 excepted) has been pooled human sera with high HI antibody titres to all described elsewhere [Hawkes, 19881. In addition, the Australian flaviviruses or b) 100 p.1 of a 1:2,500 diluMAb SLE 6B6C-1, induced by St. Louis encephalitis tion of purified monoclonal antibody JE6B4A-10 [Roe(SLE) virus, and widely reactive to all members of hrig, 19861. Either antibody when diluted in alkaline Human Survey Sera A previous survey of about 17,000 subjects throughout New South Wales [Boughton et al., 1984; Hawkes e t al., 19851 provided 468 human sera possessing HI antibodies to one or more flaviviruses. Most biophysical regions of the State were well represented. Flavivirus antibody negative sera from subjects living in areas not prone to arboviruses were used as negative controls to standardise the DEB-ELISA test, when human sera were being assayed.
Defined Epitope Blocking With MVE Virus
33
carbonate buffer efficiently captures MVE and closely studies of blocking of different MAbs with such sera related viruses onto the solid phase. After 4 hours in- difficult to evaluate. As expected, all 28 day post-injection sera, regardless cubation at room temperature, the capture antibody was discarded and 200 pl of 0.5% bovine serum albu- of the immunising virus, possessed antibodies which min (BSA) in phosphate-buffered saline (PBS-BSA) were able to combine with flavivirus group-reactive was added. After 2 hours at room temperature, the epitopes, and thus block the 2 broadly reactive antiplates were washed with PBS containing 0.05% Tween E-4b MAbs in the DEB-ELISA test. (These 2 MAbs behaved in a similar fashion throughout the whole 20. 2. Fifty haemagglutinating units of MVE acetone- study.) With the MVE antisera, the DEB-ELISA blocking sucrose extracted viral antigen in 25 pl of 90% PBSBSA and 10% acetone extracted normal mouse serum titres against the more specific epitopes tended to be (AENMS) were added. The plate was incubated over- similar to those shown against the group-reactive night at 4°C to allow attachment of the virus to anti- epitope. This implied the existence in MVE antisera of body on the solid phase and then washed and drained. substantial levels of MVE (type)-specific antibodies. It was evident that antibodies elicited by ALF and 3. The blocking step. Twenty-five microlitres of untreated test serum, diluted in PBS-BSA, was added and KUN infections of some mice also possessed the ability to block the attachment of MAbs which had been reacleft for 2 hours a t room temperature. 4. Without removing the test serum, 25 p1 of HRP- tive with only the MVE type-specific epitopes in direct conjugated MAb, diluted in 90% PBS-BSA/lO% binding assays (Table I). This anomalous finding was AENMS, was added. Each conjugate used had been di- more marked with ALF sera than with KUN sera, parluted to achieve approximately the same optical den- alleling the closer antigenic relationship which ALF sity when reacted against 50 HA of un-blocked MVE has to MVE virus [Calisher et al., 19891. Even with antigen. After a n additional 1hour incubation at room ALF sera, however, the titre of blocking antibodies to temperature, the plates were again washed and thor- the specific MAbs was often significantly lower than t h a t shown to the group-reactive (E-4b) epitope. Unexoughly drained. 5. The substrate 3,3’,5,5’-tetramethylbenzidinepectedly, the MAb enabling the best discrimination be(TMB) in citrate acetate buffer pH 6.0 was added in 100 tween MVE and ALF was the anti-E-6 which reacts pl volumes to each well. The reaction was stopped after with both MVE and ALF in direct binding assays (Ta3 minutes by the addition of 50 pl of 2M H2S0,, and the ble I). Anti-E-ld, anti-E-lc, and anti-E-5 MAbs were less optical density (OD) was measured at 450 nm. Sera were usually screened at a dilution of 15or 1:lO. Serial able to discriminate between MVE antisera and those 2-fold dilutions of positive sera were then tested. Sig- induced by a single injection of the other flaviviruses. Antisera from mice injected with 2 doses of vinificant blocking of epitopes of MVE was deemed to have occurred when the OD of the test sample was less rus. Surveys done in New South Wales [Hawkes et al., than half that shown by a panel of normal antibody 19851 have indicated that a minor proportion of the negative sera of the same species as that under test. human population experiences multiple sequential flaPreliminary experiments indicated that no improve- vivirus infections. In a n attempt to mimic this situament in the sensitivity or specificity of the assay could tion experimentally, mice were injected, a t a n interval be gained by the use of untreated 10% brain suspension of 2 weeks, with a second dose of the same or a different of MVE as antigen, or by raising the temperature of the flavivirus. Five of the possible 9 sequential combinations were used, namely MVE/MVE, ALF/ALF, KUN/ blocking step to 37°C. KUN, ALF/KUN; KUNIALF. (It was of some interest to know whether sequential infections with 2 non-MVE viruses could simulate the blocking pattern generated RESULTS by MVE virus itself.) DEB-ELISA and HI results on Evaluation of the DEB-ELISA With the 28 day sera are shown in Table 111. Mouse Antisera As in the singly injected mice (Table 11), all sera As a prerequisite to field testing, sera from immu- blocked group E-4b epitopes. Sera from mice injected nised and hyperimmunised mice were tested for their twice with MVE possessed type-specific antibodies caability to block the attachment of HRP-conjugated pable of blocking the more specific MAbs, usually to a MVE MAbs of varying specificities, to MVE. The similar titre as the E-4b group-reactive MAbs. The sera broadly reacting MAb SLE 6B6C-1 (antiE-4b) was also from non-MVE immunisation schedules also possessed antibodies to the more MVE-specific epitopes but, alused for comparison. Antisera from mice irqjected with one dose of vi- most without exception the blocking titres were signifrus. Mice were sacrificed at 14 and 28 days after a icantly lower than those to the anti-E-4b MAbs. Although not inferior to the other MAbs in this resingle injection with either MVE, ALF, or KUN viruses. DEB-ELISA and HI results are detailed on Table spect, anti-E-6 MAb was not markedly superior in be2. In some of the 14 day sera, DEB-ELISA antibody ing able to distinguish MVE infection sera from those titres were low or undetectable, making comparative elicited with other flaviviruses.
Hawkes et al.
34
TABLE 11. DEB-ELESA and HI Results on Antisera From Mice Given One I.P. Injection of 0.2 ml. Living 10%smb Suspension of Virus and Sampled 14 or 28 Days Later* DEB-ELISA titre blocking anti-epitope MAb Virus injected MVE
Days post injection 14
MVE
28
ALF
14
ALF
28
KUN
14
KUN
28
Mouse No. 1 2 3 1 2 3 4 1 2 3 1 2 3 4 5 1 2 3 1 2 3 4 5
MVE 10 10 40 640 21,280 640 21,280 40 20 20 320 160 320 160 160 40 20 40 640 320 640 320 320
HI titre vs ALF KUN 10 10 10 10 10 10 21,280 320 21,280 640 21,280 320 21,280 1,280 80 80 20 10 40 10 21,280 40 640 40 21,280 80 640 80 640 40 40 160 40 80 80 21,280 21,280 21,280 21,280 640 21,280 21,280 640 640 640 640
E-ld (4B5A-2)
