HYBRIDOMA Volume 11, Number 6, 1992 Mary Ann Liebert, Inc., Publishers
Epitope Specificity of Monoclonal Antibodies Against Newcastle Disease Virus: Competitive Fluorogenic Enzyme Immunoassay JONATHAN P. WONG, ROBERTA E. and YUNUS M. SIDDIQUI
FULTON,
Biomédical Defence Section, Defence Research Establishment Suffield, Box 4000, Medicine Hat, Alberta, TÍA 8K6, Canada
A test to determine the epitope specificity of monoclonal antibodies (MCA) was developed for hybridoma clones producing antibodies against Newcastle disease virus (NDV). The virus was first immobilized on nitrocellulose membranes of Millititer1" HA plates. Dilutions of MCA were then added, singly, or simultaneously in pairs, and bound antibody was quantitated with alkaline phosphat¬ ase- labelled detector antibody and a fluorogenic substrate, Fluorescence count was 4-methylumbelliferyl phosphate (4-MUP). measured fluorometrically. Additivity indices were calculated and plotted against dilu¬ tions of paired MCA. Antibodies that recognized identical epitopes displayed non-additivity at saturating antibody dilutions, followed by partial additivity and by total additivity at low, non-saturating In contrast, MCA that recognized distinct epitopes dilutions. exhibited total additivity throughout the curve. MCA that exhibited partial additivity were interpreted as competing for overlapping shared epitopes, or, distinct epitopes in close proximity, resulting in steric hinderance.
In the selection of immunoreagents for use in the development of immunoassays for the detection of macromolecular antigens, it is often desirable to identify MCA which are reactive against the same Methods antigen but which have different epitope specificity. currently used to determine whether MCA recognize the same or different antigenic regions are cumbersome and time-consuming, involving purification and labelling of each MCA to be tested (1, 2) or preliminary titration of each MCA to ensure antigen saturation We describe here a technique which circumvents these (3, 4). problems, by combining a modification of the immunoenzymatic approach described by Friguet et al. (3) with a highly sensitive fluorogenic enzyme-linked immunosorbent assay (FELISA) described by Fulton et al. (5), to determine the epitope specificity of a panel of MCA to NDV. The test is simple to perform and does not require of reagents or preliminary titration of MCA required to saturate the antigen. Alkaline phosphatase-labelled affinity purified goat anti-mouse IgG was purchased from Bio-Rad Laboratories (Mississauga, Ont.).
labelling
829
4-methylumbelliferyl phosphate (4-MUP)
was purchased from Sigma NDV strain NJ-La Sota was Company (St. Louis, MO). purchased from American Type Culture Collection (Rockville, MD) and cultivated, by established methods (6), in the allantoic cavity of embryonated hens' eggs. Harvested allantoic fluids were assayed for The virus was virus titer by hemagglutination (HA) test (7) purified by two cycles of differential centrifugation followed by single cycles of discontinuous and continuous sucrose gradient centrifugation, as described by Fulton et al. (5). Hybridoma clones producing MCA directed against the NJ-La Sota strain of NDV were prepared, under contract, by the Department of Immunology, University of Alberta (Edmonton, Alta.). The procedure used was that originally described by Köhler and Milstein (8) and, briefly, was as follows. Myeloma cells (THT) (Hyclone Laboratories, Logan, UT) were fused to spleen cells derived from Balb/c mice (Charles River Ltd., St. Constant, Que.) which had previously been
Chemical
.
with NDV. The fused hybrid cells were cloned and RPMI-1640 medium (Gibco/ BRL, Burlington, Ont.) supplemented with 15% fetal bovine serum Gibco/BRL) and containing 10 10 adenine (1.5 M) and thymidine (3.2 M), aminopterin (4 10 M) (Sigma Chemical Co.). Balb/c peritoneal cells were used as feeder cells. Culture supernatants from hybridoma clones were screened for the presence of NDV antibodies by an indirect FELISA, described below. Clones which were strongly positive were amplified by growth in RPMI-1640 medium, supplemented as described above, and
hyperimmunized cultured
in
10 cells/mouse) into Balb/c mice. injected intraperitoneally (1 Two weeks prior to injection, mice were primed by intraperitoneal administration of 500 µ of pristane (Sigma Chemical Co.) and, 24 h prior to injection, mice were irradiated, at a dose of 500 rad, by Ascites a Gamma Cell 40 (Atomic Energy of Canada, Nepean, Ont.). fluids were harvested by paracentesis after a period of 2-4 weeks, when the peritoneal cavity of the mice had become distended. Specific antibodies to NDV in the ascites fluids were assayed by indirect FELISA. The optimal working dilution of alkaline phosphatase-labelled goat anti-mouse IgG for use in the competitive FELISA and the indirect FELISA, was determined by checkerboard titration against a representative strongly positive ascites fluid, in Millititer™ wells sensitized with 50 µ of optimal concentration of purified virus protein (20 pg/ml) (1 pg/well). The optimal concentration of virus protein, defined as the lowest concentration required to saturate the nitrocellulose membranes, was determined experimentally (results not An shown). optimal dilution of 1:1,000 for alkaline phophatase-labelled goat anti-mouse IgG yielded the highest ratio of positive to background fluorescence and hence, was adopted for routine use. Optimal assay conditions such as incubation time, components of the washing solutions, conditions of incubation, and number of washes were adopted for use from the FELISA previously described by Fulton et al. (5). Competitive FELISA procedures were carried out in 96-well Millititer™ HA filtration plates (Millipore Corp., Mississauga, Ont.) in which 0.45 µ nitrocellulose membranes formed the bottom surface. Immediately prior to use, wells were washed three times with phosphate buffered saline, pH 7.4 (PBS). Washing steps and removal of unbound reagents were achieved by vacuum filtration in a Millititer™ filtration system (Millipore Corp.). Wells were first coated with NDV by adding 50 µ of the optimal concentration of purified NDV protein (20 pg/ml), diluted in The plates were 0.05 M carbonate-bicarbonate buffer, pH 9.6. incubated overnight at 4°C and were then washed three times with 200 µ of PBS. Unoccupied binding sites on the membranes were blocked for 1 h at 37°C with 200 µ of PBS containing 2% bovine serum
830
albumin (BSA) and 0.1% Tween 20 (T) (PBS-BSA-T). The plates were washed once with PBS, after which, the blocking step was repeated twice. The plates were then washed three times with PBS. Ascites fluids to be tested for epitope specificity were serially diluted in PBS-BSA-T. For testing of ascites fluids singly, 50 µ of each dilution was added to the appropriate wells; for testing of ascites fluids in pairs, equal volumes of each dilution were mixed by vortexing and 100 µ of the mixtures were then added to the appropriate wells. The plates were incubated for 1 h at 37°C and then washed five times with PBS containing 0.05% T. Wells were then incubated at 37°C for 1 h with 50 µ of the optimal dilution (1:1,000) of alkaline phosphatase-labelled goat anti-mouse IgG, diluted in PBS-BSA-T. After six cycles of washing with PBS containing 0.05% T, 200 µ of the enzyme substrate solution, 10"* M 4-MUP in 10% diethanolamine buffer, pH 9.8, was added and the plates incubated for 15 min at room temperature in the dark. The fluorescent product, resulting from hydrolysis of the substrate by
bound enzyme-labelled antibody conjugate, was measured directly on the Millititer™ HA plates by a microFLUOR* fluorometer (Dynatech Laboratories, Alexandria, VA) fitted with 365 nm and 450 nm filters for excitation and emission, respectively. The fluorometer was blanked on wells which received enzyme substrate solution only and
appropriate controls
were
included.
Ascites fluids and hybridoma supernatants were screened for the presence of antibodies to NDV by an indirect FELISA. The procedures were carried out as in the competitive FELISA, with the exception that the varying dilutions of the ascites fluids and hybridoma supernatants were added independently and were not paired with each other. A panel of MCA was analyzed by competitive FELISA, both singly and in pairs, to determine whether members of a pair competed with each other for binding sites on NDV immobilized on nitrocellulose membranes. If members of a MCA pair recognized different epitopes, the amount of antibody bound to the virus when these MCA were paired and added simultaneously, would be expected to correspond to the sum of the amount bound when each was added separately, regardless of the dilution of the MCA added. In contrast, when members of a pair of MCA recognized the same epitope, the binding of each MCA at saturation concentrations would be expected to be non-additive and
RECIPROCAL OF MCA DILUTIONS
Figure .
1
Epitope specificities of MCA pairs. Dilutions of each MCA were added separately, or simultaneously, to the antigen-coated solid phase. The relative amount of MCA bound at each dilution (fluorescence count) was quantitated by FELISA. Data points are the mean of triplicate determinations. and C.
831
bound would be approximately equal to the each was added separately, since these MCA would be competing for the same binding site on the virus. The epitope specificities of three representative pairs of MCA {al ß6 and 3 ß2),(a2 ß6 and al ß5) and {al ß5 and aU ßl) analyzed by competitive FELISA, are presented (Figure 1). There was no competition observed between MCA al ß6 and 3 ß2 (Figure ). The fluorescence count (FC) of the MCA pair, when added simultaneously, corresponded, at all dilutions, to the sum of the FC obtained for each MCA when added separately. This finding suggested that the two MCA bound to distinct epitopes on the virus. In contrast, MCA pair al ß6 and al ß5 competed with each other for the same antigenic site (Figure IB) The FC for the pair at saturating antibody dilutions (1:20 to 1:80) was approximately equal to the mean of the FC At obtained for these MCA when added separately. high, non-saturating dilution (1:10,240), summation of the FC was obtained, while partial summation of the FC was observed at antibody dilutions of 1:160 to 1:5,120. These findings suggested that the two MCA were reactive against the same virus epitope. Partial competition was observed between MCA pair al ß5 and 4 ßl (Figure IC). Throughout the curve, the FC for the pair, when added simultaneously, was between the sum and the mean of the FC when each was added separately. In order to quantitate the degree of competition between MCA pairs, the 'additivity index' (AI), derived by Friguet et al. (3), was adopted as follows:
antibody
the amount of amount bound when
,
.
% AI
=
(
2F, F
2
.
F
1
)
100
-
_
where F1 F2 and F1+, represented, respectively, the FC obtained when the first MCA was added separately, the second MCA was added separately, and the two MCA were added simultaneously. If a MCA pair bound to the same epitope, F1+, would be expected to equal the mean of F. + F, and AI would equal 0%. If a MCA+ pair bound to distinct sites, would be to the sum of F.. equal F, and AI would F1+. equal 100%. To ascertain the level of competition between MCA pair binding to an antigenic determinant, the calculated AI were plotted against dilutions of paired MCA (Figure 2). AI of 80-100%, 30-79% and 0-29% were arbitrarily defined as exhibiting total, partial and The MCA pair al ß6 and 3 ß2 non-additivity, respectively. exhibited total additivity throughout the curve (Figure 2A), suggesting that these MCA recognized distinct epitopes. In con¬ trast, the MCA pair al ß6 and al ß5 exhibited non-additivity at low, saturating dilutions, followed by partial additivity with increasing dilutions and total additivity at high non-saturating dilutions (Figure 2B) suggesting that this MCA pair recognized the same epitope. The MCA pair al ß5 and 4 ßl exhibited partial additivity at all dilutions (Figure 2C), suggesting partial competition between the MCA pair for binding to the virus epitope. The results presented in this study indicate that the competitive FELISA is a relatively simple technique that could be used to determine if two MCA, produced by two different hybridoma clones, recognize the same or distinct epitopes on a virus. The method that we propose combines a sensitive fluorogenic enzyme immunoassay on nitrocellulose membranes, described by Fulton et al. (5) , with a modification of a competitive antibody-binding assay described by Friguet et al. (3). ,
,
832
100
RECIPROCAL OF MCA DILUTIONS
Figure
2
Comparison of additivity indices of three MCA pairs tested for epitope specificity. Additivity indices were plotted as a function of dilution of MCA pairs 06 and 3 02 (A); al 06 and al 05 ( ); and al 05 and o4 B1 (C), respectively. Legend: total additivity (80-100%); partial additivity (30-79%); non-additivity (0-29%).
The method most commonly used to determine the epitope specificity of MCA involves the use of 125iodine-labelled MCA in a competitive antibody-binding assay for antigenic determinants (1, 2). This approach, however, requires purification of each MCA to be tested and, in addition, the labelling of each MCA with radioactive isotopes, procedures which are labour-intensive and which could In addition, adversely affect the immunoreactivity of the MCA. radioiodinated MCA have short shelflives of only a few weeks and handling of radioisotopes could pose potential health hazards to laboratory personnel. To overcome these drawbacks inherent in the use of radioisotope-labelled MCA in competitive antibody-binding two enzyme-linked chromogenic non-radioisotopic assays, The double immunosorbent assay (ELISA) systems were developed. antibody system, described by Friguet et al. (3), involved the reaction of predetermined saturating dilutions of two different MCA, to added singly or simultaneously, antigen-coated polystyrene microtiter plates. Additivity of the amount of antibody bound, suggestive of participation of different epitopes, was then quantitated in a chromogenic enzymatic reaction, with an enzymelabelled anti-immunoglobulin conjugate. Similarly, Hendry et al. (4), described a competitive binding assay in a monoclonal capture ELISA, in which, the binding of antigen to the capture MCA on the solid phase was inhibited by preincubation of the antigen with a predetermined saturating amount of a second competing MCA. Although the requirement for radioisotopic labelling of purified MCA was eliminated in the ELISA systems described by Friguet et al. (3) and Hendry et al. (4), there are certain disadvantages in their methodologies, which have been overcome in the competitive FELISA described in this study. In both of the previously described ELISA systems for determination of epitope specificity, the concentration of MCA required to saturate a fixed amount of antigen was predetermined for each MCA to be tested, a procedure which is time-consuming. In the competitive FELISA described in this paper, competion of MCA for a fixed concentration of immobilized antigen was evaluated in a single assay over a range of MCA dilutions, thus obviating a requirement for predetermination of MCA saturation In concentrations and hence, providing for more rapid analysis. as solid million times more sensitive ELISA systems which utilize polystyrene solid supports (5). This enhanced sensitivity is attributable not 15 17-fold a to to greater absorptive capacity of
addition, the FELISA system, which utilizes nitrocellulose
phase support,
is than conventional
phase only
approximately
two
833
nitrocellulose for proteins, compared to polystyrene (5), but also that fluorescent end determined to the fact products, spectrofluorometrically, are detected at lower concentrations than are colored products, determined spectrophotometrically (10, 11). This enhanced sensitivity of the FELISA system would be expected to result in signal amplification and hence increased resolution of the Such a significant increase in resolution, epitope analysis. offered by the competitive FELISA, would undoubtedly facilitate discrimination of subtle differences in the antigen-combining sites of two closely-related MCA, which could be directed against epitopes that were similar, but differed only slightly in amino acid sequence. In addition, because the competitive FELISA is performed in plates of 96-well format, large numbers of samples can be conveniently analyzed, simultaneously. Millititer™ HA plates can be precoated with the virus antigen, frozen, and stored for weeks prior to the assay, without significant loss in antigenic activity of the immobilized virus protein (results not shown), thus results can be readily obtained within 3-4 h. MCA function as specific probes for small regions of complex macromolecules and, when characterized according to their epitope specificity, are valuable tools in defining subtle differences in NDV MCA, characterized with respect to the protein sequence. antigenic determinants against which they are directed, could therefore be utilized in elucidating differences, at the macro molecular level, in protein composition among different strains of NDV and among individual members of the paramyxoviruses (12, 13). In addition, MCA directed against different antigenic determinants on NDV, could be used to construct enzyme immunoassay systems of the "sandwich"-type format, in which two MCA of distinct epitope specificity, could be used to capture and detect, respectively, the antigen. Such assay formulations, which utilize a MCA on the solid phase to capture the antigen, would be expected to result in enhancement of test sensitivity (5). We have described, in this paper, a sensitive competitive fluorogenic enzyme immunoassay which utilizes nitrocellulose membranes as solid phase support, for determination of epitope Purification and specificity of MCA directed against NDV. subsequent radiolabelling of MCA are not required and tests can be performed directly on hybridoma supernatants or ascites fluids. In addition, preliminary determination of the concentration of MCA required to saturate the antigen is not required and the degree of competition between members of MCA pairs can be estimated in a single test titration. The test is simple to perform, large numbers of samples can be simultaneously analyzed, and results are readily obtainable within 3 to 4 h. We propose that the competitive FELISA will be a valuable alternative for the non-isotopie analysis of epitope specificities of MCA.
REFERENCES
1.
Nowinski, R.C., Strong, M.R., Tarn, M.R., Lostrom, M.E., Burnette, W.H. and O'Donnell, P.V. (1981) Mapping of viral
proteins with monoclonal antibodies: analysis In: proteins of murine leukemia viruses. McKearn,
T.J.
Antibodies.
analyses.
and
Bechtol,
Hybridomas: Plenum
Press,
K.B. a
New
new
dimension
envelope
Kennet,
R.H. Monoclonal
(Eds.).
York, 295-315.
834
of the
in
,
biological
2.
(1983) Nonoverlapping anti¬ sites of woodchuck hepatitis virus surface antigen and their cross-reactivity with ground squirrel hepatitis virus and surface antigens. J. Virol. 47:15-23. hepatitis Cote, P.J., Jr. and Gerin, J.L.
genic
3.
Friguet, B., Djavadi-Ohaniance, L., Pages, J., Bussard, A. and Goldberg, M. (1983) A convenient enzyme-linked immunosorbent assay for testing whether monoclonal antibodies recognize the same antigenic site. Application to hybridomas specific for the
ß-subunit of Escherichia coli Methods 60: 351-358.
Immunol. 4.
tryptophan synthase.
J.
Hendry, R.M., Fernie, B.F., Anderson, L.J., Godfrey, E. and Monoclonal capture antibody ELISA for Mclntosh, K. (1985) respiratory syncytial virus: detection of individual viral antigens and determination of monoclonal antibody specifici¬ ties.
5.
J. Immunol. Methods 77:247-258.
Fulton, R.E., Wong, J.P., Siddiqui, Y.M. andTso, M.-S. (1988) Sensitive fluorogenic enzyme immunoassay on nitrocellulose membranes for quantitation of virus. Methods J. Virol. 22:149-164.
6.
Hawkes,
(1979) General principles underlaying laboratory In: viral infections. Lennette, E.H. and (Eds.). Diagnostic Procedures for Viral,
R.A.
diagnosis
of
Schmidt, N.J. Rickettsial and Chlamydial Infections. Health Assoc, Washington, D.C. 3-48. 7.
Grist, N.R., Ross, CA. and Bell, E.J.
5th ed.
Amer.
Public
(1974) Hemagglutination
and hemagglutination inhibition tests. In: Diagnostic Methods in Clinical Virology. 2nd ed. Blackwell Scientific Publica¬ tions, London. 103-111.
8.
Köhler, G. and Milstein, C. fused
cells secreting Nature 256:495. 9.
(1975) Continuous cultures of
antibody
of
predefined specificity.
Layne, E. (1957) Spectrophotometric and turbidimetric methods for measuring proteins. In: Colowich, S.P. and Kaplan, N.O. (Eds.). Methods in Enzymology. Vol. 3. Academic Press, New York. 447-454.
10.
Fernley,
H.N.
and
Walker, P.C
(1965) Kinetic behaviour of
calf-intestinal alkaline phosphatase with phosphate. Biochem. J. 97: 95-103. 11.
Kato, K.Y., Hamaguchi, Y., Fukui, H.
and
4-methylumbelliferyl Ishikawa, E. (1976)
Enzyme-linked immunoassay. Conjugation of rabbit anti-(human immunoglobulin G) antibody with ß-B-galactosidase from Escherichia coli and its use for human immunoglobulin G assay. Eur. J.
Biochem.
62:185-192.
P.H. and Alexander, D.J. (1983) Antigenic variations of Newcastle disease virus strains detected by monoclonal antibodies. Arch. Virol. 75: 243-253.
12.
Russell,
13.
Nishikawa,
K.
antibodies
to
,
et al. (1983) Monoclonal of Newcastle disease virus:
Isomura, S., Suzuki, S. the HN
glycoprotein
835
biological characterization Virology 130:318-330. Address
and
use
for
strain
reprint request
comparison.
to:
Jonathan Wong Defence Research Establishment Suffield Box 4000 Medicine Hat, Alberta Canada TÍA 8K6
Received for publication: 4/30/92 after revision as a short ccmnunication: 6/5/92
Accepted
836
This article has been cited by: 1. John J. Weiland, James V. Anderson, Brant B. Bigger. 2007. Inexpensive chemifluorescent detection of antibody–alkaline phosphatase conjugates on Western blots using 4-methylumbelliferyl phosphate. Analytical Biochemistry 361:1, 140-142. [CrossRef] 2. W LEE, H THOMPSON, J HALL, R FULTON, J WONG. 1993. Rapid immunofiltration assay of Newcastle disease virus using a silicon sensor. Journal of Immunological Methods 166:1, 123-131. [CrossRef]
Epitope Specificity of Monoclonal Antibodies Against Newcastle Disease Virus: Competitive Fluorogenic Enzyme Immunoassay JONATHAN P. WONG, ROBERTA E. and YUNUS M. SIDDIQUI
FULTON,
Biomédical Defence Section, Defence Research Establishment Suffield, Box 4000, Medicine Hat, Alberta, TÍA 8K6, Canada
A test to determine the epitope specificity of monoclonal antibodies (MCA) was developed for hybridoma clones producing antibodies against Newcastle disease virus (NDV). The virus was first immobilized on nitrocellulose membranes of Millititer1" HA plates. Dilutions of MCA were then added, singly, or simultaneously in pairs, and bound antibody was quantitated with alkaline phosphat¬ ase- labelled detector antibody and a fluorogenic substrate, Fluorescence count was 4-methylumbelliferyl phosphate (4-MUP). measured fluorometrically. Additivity indices were calculated and plotted against dilu¬ tions of paired MCA. Antibodies that recognized identical epitopes displayed non-additivity at saturating antibody dilutions, followed by partial additivity and by total additivity at low, non-saturating In contrast, MCA that recognized distinct epitopes dilutions. exhibited total additivity throughout the curve. MCA that exhibited partial additivity were interpreted as competing for overlapping shared epitopes, or, distinct epitopes in close proximity, resulting in steric hinderance.
In the selection of immunoreagents for use in the development of immunoassays for the detection of macromolecular antigens, it is often desirable to identify MCA which are reactive against the same Methods antigen but which have different epitope specificity. currently used to determine whether MCA recognize the same or different antigenic regions are cumbersome and time-consuming, involving purification and labelling of each MCA to be tested (1, 2) or preliminary titration of each MCA to ensure antigen saturation We describe here a technique which circumvents these (3, 4). problems, by combining a modification of the immunoenzymatic approach described by Friguet et al. (3) with a highly sensitive fluorogenic enzyme-linked immunosorbent assay (FELISA) described by Fulton et al. (5), to determine the epitope specificity of a panel of MCA to NDV. The test is simple to perform and does not require of reagents or preliminary titration of MCA required to saturate the antigen. Alkaline phosphatase-labelled affinity purified goat anti-mouse IgG was purchased from Bio-Rad Laboratories (Mississauga, Ont.).
labelling
829
4-methylumbelliferyl phosphate (4-MUP)
was purchased from Sigma NDV strain NJ-La Sota was Company (St. Louis, MO). purchased from American Type Culture Collection (Rockville, MD) and cultivated, by established methods (6), in the allantoic cavity of embryonated hens' eggs. Harvested allantoic fluids were assayed for The virus was virus titer by hemagglutination (HA) test (7) purified by two cycles of differential centrifugation followed by single cycles of discontinuous and continuous sucrose gradient centrifugation, as described by Fulton et al. (5). Hybridoma clones producing MCA directed against the NJ-La Sota strain of NDV were prepared, under contract, by the Department of Immunology, University of Alberta (Edmonton, Alta.). The procedure used was that originally described by Köhler and Milstein (8) and, briefly, was as follows. Myeloma cells (THT) (Hyclone Laboratories, Logan, UT) were fused to spleen cells derived from Balb/c mice (Charles River Ltd., St. Constant, Que.) which had previously been
Chemical
.
with NDV. The fused hybrid cells were cloned and RPMI-1640 medium (Gibco/ BRL, Burlington, Ont.) supplemented with 15% fetal bovine serum Gibco/BRL) and containing 10 10 adenine (1.5 M) and thymidine (3.2 M), aminopterin (4 10 M) (Sigma Chemical Co.). Balb/c peritoneal cells were used as feeder cells. Culture supernatants from hybridoma clones were screened for the presence of NDV antibodies by an indirect FELISA, described below. Clones which were strongly positive were amplified by growth in RPMI-1640 medium, supplemented as described above, and
hyperimmunized cultured
in
10 cells/mouse) into Balb/c mice. injected intraperitoneally (1 Two weeks prior to injection, mice were primed by intraperitoneal administration of 500 µ of pristane (Sigma Chemical Co.) and, 24 h prior to injection, mice were irradiated, at a dose of 500 rad, by Ascites a Gamma Cell 40 (Atomic Energy of Canada, Nepean, Ont.). fluids were harvested by paracentesis after a period of 2-4 weeks, when the peritoneal cavity of the mice had become distended. Specific antibodies to NDV in the ascites fluids were assayed by indirect FELISA. The optimal working dilution of alkaline phosphatase-labelled goat anti-mouse IgG for use in the competitive FELISA and the indirect FELISA, was determined by checkerboard titration against a representative strongly positive ascites fluid, in Millititer™ wells sensitized with 50 µ of optimal concentration of purified virus protein (20 pg/ml) (1 pg/well). The optimal concentration of virus protein, defined as the lowest concentration required to saturate the nitrocellulose membranes, was determined experimentally (results not An shown). optimal dilution of 1:1,000 for alkaline phophatase-labelled goat anti-mouse IgG yielded the highest ratio of positive to background fluorescence and hence, was adopted for routine use. Optimal assay conditions such as incubation time, components of the washing solutions, conditions of incubation, and number of washes were adopted for use from the FELISA previously described by Fulton et al. (5). Competitive FELISA procedures were carried out in 96-well Millititer™ HA filtration plates (Millipore Corp., Mississauga, Ont.) in which 0.45 µ nitrocellulose membranes formed the bottom surface. Immediately prior to use, wells were washed three times with phosphate buffered saline, pH 7.4 (PBS). Washing steps and removal of unbound reagents were achieved by vacuum filtration in a Millititer™ filtration system (Millipore Corp.). Wells were first coated with NDV by adding 50 µ of the optimal concentration of purified NDV protein (20 pg/ml), diluted in The plates were 0.05 M carbonate-bicarbonate buffer, pH 9.6. incubated overnight at 4°C and were then washed three times with 200 µ of PBS. Unoccupied binding sites on the membranes were blocked for 1 h at 37°C with 200 µ of PBS containing 2% bovine serum
830
albumin (BSA) and 0.1% Tween 20 (T) (PBS-BSA-T). The plates were washed once with PBS, after which, the blocking step was repeated twice. The plates were then washed three times with PBS. Ascites fluids to be tested for epitope specificity were serially diluted in PBS-BSA-T. For testing of ascites fluids singly, 50 µ of each dilution was added to the appropriate wells; for testing of ascites fluids in pairs, equal volumes of each dilution were mixed by vortexing and 100 µ of the mixtures were then added to the appropriate wells. The plates were incubated for 1 h at 37°C and then washed five times with PBS containing 0.05% T. Wells were then incubated at 37°C for 1 h with 50 µ of the optimal dilution (1:1,000) of alkaline phosphatase-labelled goat anti-mouse IgG, diluted in PBS-BSA-T. After six cycles of washing with PBS containing 0.05% T, 200 µ of the enzyme substrate solution, 10"* M 4-MUP in 10% diethanolamine buffer, pH 9.8, was added and the plates incubated for 15 min at room temperature in the dark. The fluorescent product, resulting from hydrolysis of the substrate by
bound enzyme-labelled antibody conjugate, was measured directly on the Millititer™ HA plates by a microFLUOR* fluorometer (Dynatech Laboratories, Alexandria, VA) fitted with 365 nm and 450 nm filters for excitation and emission, respectively. The fluorometer was blanked on wells which received enzyme substrate solution only and
appropriate controls
were
included.
Ascites fluids and hybridoma supernatants were screened for the presence of antibodies to NDV by an indirect FELISA. The procedures were carried out as in the competitive FELISA, with the exception that the varying dilutions of the ascites fluids and hybridoma supernatants were added independently and were not paired with each other. A panel of MCA was analyzed by competitive FELISA, both singly and in pairs, to determine whether members of a pair competed with each other for binding sites on NDV immobilized on nitrocellulose membranes. If members of a MCA pair recognized different epitopes, the amount of antibody bound to the virus when these MCA were paired and added simultaneously, would be expected to correspond to the sum of the amount bound when each was added separately, regardless of the dilution of the MCA added. In contrast, when members of a pair of MCA recognized the same epitope, the binding of each MCA at saturation concentrations would be expected to be non-additive and
RECIPROCAL OF MCA DILUTIONS
Figure .
1
Epitope specificities of MCA pairs. Dilutions of each MCA were added separately, or simultaneously, to the antigen-coated solid phase. The relative amount of MCA bound at each dilution (fluorescence count) was quantitated by FELISA. Data points are the mean of triplicate determinations. and C.
831
bound would be approximately equal to the each was added separately, since these MCA would be competing for the same binding site on the virus. The epitope specificities of three representative pairs of MCA {al ß6 and 3 ß2),(a2 ß6 and al ß5) and {al ß5 and aU ßl) analyzed by competitive FELISA, are presented (Figure 1). There was no competition observed between MCA al ß6 and 3 ß2 (Figure ). The fluorescence count (FC) of the MCA pair, when added simultaneously, corresponded, at all dilutions, to the sum of the FC obtained for each MCA when added separately. This finding suggested that the two MCA bound to distinct epitopes on the virus. In contrast, MCA pair al ß6 and al ß5 competed with each other for the same antigenic site (Figure IB) The FC for the pair at saturating antibody dilutions (1:20 to 1:80) was approximately equal to the mean of the FC At obtained for these MCA when added separately. high, non-saturating dilution (1:10,240), summation of the FC was obtained, while partial summation of the FC was observed at antibody dilutions of 1:160 to 1:5,120. These findings suggested that the two MCA were reactive against the same virus epitope. Partial competition was observed between MCA pair al ß5 and 4 ßl (Figure IC). Throughout the curve, the FC for the pair, when added simultaneously, was between the sum and the mean of the FC when each was added separately. In order to quantitate the degree of competition between MCA pairs, the 'additivity index' (AI), derived by Friguet et al. (3), was adopted as follows:
antibody
the amount of amount bound when
,
.
% AI
=
(
2F, F
2
.
F
1
)
100
-
_
where F1 F2 and F1+, represented, respectively, the FC obtained when the first MCA was added separately, the second MCA was added separately, and the two MCA were added simultaneously. If a MCA pair bound to the same epitope, F1+, would be expected to equal the mean of F. + F, and AI would equal 0%. If a MCA+ pair bound to distinct sites, would be to the sum of F.. equal F, and AI would F1+. equal 100%. To ascertain the level of competition between MCA pair binding to an antigenic determinant, the calculated AI were plotted against dilutions of paired MCA (Figure 2). AI of 80-100%, 30-79% and 0-29% were arbitrarily defined as exhibiting total, partial and The MCA pair al ß6 and 3 ß2 non-additivity, respectively. exhibited total additivity throughout the curve (Figure 2A), suggesting that these MCA recognized distinct epitopes. In con¬ trast, the MCA pair al ß6 and al ß5 exhibited non-additivity at low, saturating dilutions, followed by partial additivity with increasing dilutions and total additivity at high non-saturating dilutions (Figure 2B) suggesting that this MCA pair recognized the same epitope. The MCA pair al ß5 and 4 ßl exhibited partial additivity at all dilutions (Figure 2C), suggesting partial competition between the MCA pair for binding to the virus epitope. The results presented in this study indicate that the competitive FELISA is a relatively simple technique that could be used to determine if two MCA, produced by two different hybridoma clones, recognize the same or distinct epitopes on a virus. The method that we propose combines a sensitive fluorogenic enzyme immunoassay on nitrocellulose membranes, described by Fulton et al. (5) , with a modification of a competitive antibody-binding assay described by Friguet et al. (3). ,
,
832
100
RECIPROCAL OF MCA DILUTIONS
Figure
2
Comparison of additivity indices of three MCA pairs tested for epitope specificity. Additivity indices were plotted as a function of dilution of MCA pairs 06 and 3 02 (A); al 06 and al 05 ( ); and al 05 and o4 B1 (C), respectively. Legend: total additivity (80-100%); partial additivity (30-79%); non-additivity (0-29%).
The method most commonly used to determine the epitope specificity of MCA involves the use of 125iodine-labelled MCA in a competitive antibody-binding assay for antigenic determinants (1, 2). This approach, however, requires purification of each MCA to be tested and, in addition, the labelling of each MCA with radioactive isotopes, procedures which are labour-intensive and which could In addition, adversely affect the immunoreactivity of the MCA. radioiodinated MCA have short shelflives of only a few weeks and handling of radioisotopes could pose potential health hazards to laboratory personnel. To overcome these drawbacks inherent in the use of radioisotope-labelled MCA in competitive antibody-binding two enzyme-linked chromogenic non-radioisotopic assays, The double immunosorbent assay (ELISA) systems were developed. antibody system, described by Friguet et al. (3), involved the reaction of predetermined saturating dilutions of two different MCA, to added singly or simultaneously, antigen-coated polystyrene microtiter plates. Additivity of the amount of antibody bound, suggestive of participation of different epitopes, was then quantitated in a chromogenic enzymatic reaction, with an enzymelabelled anti-immunoglobulin conjugate. Similarly, Hendry et al. (4), described a competitive binding assay in a monoclonal capture ELISA, in which, the binding of antigen to the capture MCA on the solid phase was inhibited by preincubation of the antigen with a predetermined saturating amount of a second competing MCA. Although the requirement for radioisotopic labelling of purified MCA was eliminated in the ELISA systems described by Friguet et al. (3) and Hendry et al. (4), there are certain disadvantages in their methodologies, which have been overcome in the competitive FELISA described in this study. In both of the previously described ELISA systems for determination of epitope specificity, the concentration of MCA required to saturate a fixed amount of antigen was predetermined for each MCA to be tested, a procedure which is time-consuming. In the competitive FELISA described in this paper, competion of MCA for a fixed concentration of immobilized antigen was evaluated in a single assay over a range of MCA dilutions, thus obviating a requirement for predetermination of MCA saturation In concentrations and hence, providing for more rapid analysis. as solid million times more sensitive ELISA systems which utilize polystyrene solid supports (5). This enhanced sensitivity is attributable not 15 17-fold a to to greater absorptive capacity of
addition, the FELISA system, which utilizes nitrocellulose
phase support,
is than conventional
phase only
approximately
two
833
nitrocellulose for proteins, compared to polystyrene (5), but also that fluorescent end determined to the fact products, spectrofluorometrically, are detected at lower concentrations than are colored products, determined spectrophotometrically (10, 11). This enhanced sensitivity of the FELISA system would be expected to result in signal amplification and hence increased resolution of the Such a significant increase in resolution, epitope analysis. offered by the competitive FELISA, would undoubtedly facilitate discrimination of subtle differences in the antigen-combining sites of two closely-related MCA, which could be directed against epitopes that were similar, but differed only slightly in amino acid sequence. In addition, because the competitive FELISA is performed in plates of 96-well format, large numbers of samples can be conveniently analyzed, simultaneously. Millititer™ HA plates can be precoated with the virus antigen, frozen, and stored for weeks prior to the assay, without significant loss in antigenic activity of the immobilized virus protein (results not shown), thus results can be readily obtained within 3-4 h. MCA function as specific probes for small regions of complex macromolecules and, when characterized according to their epitope specificity, are valuable tools in defining subtle differences in NDV MCA, characterized with respect to the protein sequence. antigenic determinants against which they are directed, could therefore be utilized in elucidating differences, at the macro molecular level, in protein composition among different strains of NDV and among individual members of the paramyxoviruses (12, 13). In addition, MCA directed against different antigenic determinants on NDV, could be used to construct enzyme immunoassay systems of the "sandwich"-type format, in which two MCA of distinct epitope specificity, could be used to capture and detect, respectively, the antigen. Such assay formulations, which utilize a MCA on the solid phase to capture the antigen, would be expected to result in enhancement of test sensitivity (5). We have described, in this paper, a sensitive competitive fluorogenic enzyme immunoassay which utilizes nitrocellulose membranes as solid phase support, for determination of epitope Purification and specificity of MCA directed against NDV. subsequent radiolabelling of MCA are not required and tests can be performed directly on hybridoma supernatants or ascites fluids. In addition, preliminary determination of the concentration of MCA required to saturate the antigen is not required and the degree of competition between members of MCA pairs can be estimated in a single test titration. The test is simple to perform, large numbers of samples can be simultaneously analyzed, and results are readily obtainable within 3 to 4 h. We propose that the competitive FELISA will be a valuable alternative for the non-isotopie analysis of epitope specificities of MCA.
REFERENCES
1.
Nowinski, R.C., Strong, M.R., Tarn, M.R., Lostrom, M.E., Burnette, W.H. and O'Donnell, P.V. (1981) Mapping of viral
proteins with monoclonal antibodies: analysis In: proteins of murine leukemia viruses. McKearn,
T.J.
Antibodies.
analyses.
and
Bechtol,
Hybridomas: Plenum
Press,
K.B. a
New
new
dimension
envelope
Kennet,
R.H. Monoclonal
(Eds.).
York, 295-315.
834
of the
in
,
biological
2.
(1983) Nonoverlapping anti¬ sites of woodchuck hepatitis virus surface antigen and their cross-reactivity with ground squirrel hepatitis virus and surface antigens. J. Virol. 47:15-23. hepatitis Cote, P.J., Jr. and Gerin, J.L.
genic
3.
Friguet, B., Djavadi-Ohaniance, L., Pages, J., Bussard, A. and Goldberg, M. (1983) A convenient enzyme-linked immunosorbent assay for testing whether monoclonal antibodies recognize the same antigenic site. Application to hybridomas specific for the
ß-subunit of Escherichia coli Methods 60: 351-358.
Immunol. 4.
tryptophan synthase.
J.
Hendry, R.M., Fernie, B.F., Anderson, L.J., Godfrey, E. and Monoclonal capture antibody ELISA for Mclntosh, K. (1985) respiratory syncytial virus: detection of individual viral antigens and determination of monoclonal antibody specifici¬ ties.
5.
J. Immunol. Methods 77:247-258.
Fulton, R.E., Wong, J.P., Siddiqui, Y.M. andTso, M.-S. (1988) Sensitive fluorogenic enzyme immunoassay on nitrocellulose membranes for quantitation of virus. Methods J. Virol. 22:149-164.
6.
Hawkes,
(1979) General principles underlaying laboratory In: viral infections. Lennette, E.H. and (Eds.). Diagnostic Procedures for Viral,
R.A.
diagnosis
of
Schmidt, N.J. Rickettsial and Chlamydial Infections. Health Assoc, Washington, D.C. 3-48. 7.
Grist, N.R., Ross, CA. and Bell, E.J.
5th ed.
Amer.
Public
(1974) Hemagglutination
and hemagglutination inhibition tests. In: Diagnostic Methods in Clinical Virology. 2nd ed. Blackwell Scientific Publica¬ tions, London. 103-111.
8.
Köhler, G. and Milstein, C. fused
cells secreting Nature 256:495. 9.
(1975) Continuous cultures of
antibody
of
predefined specificity.
Layne, E. (1957) Spectrophotometric and turbidimetric methods for measuring proteins. In: Colowich, S.P. and Kaplan, N.O. (Eds.). Methods in Enzymology. Vol. 3. Academic Press, New York. 447-454.
10.
Fernley,
H.N.
and
Walker, P.C
(1965) Kinetic behaviour of
calf-intestinal alkaline phosphatase with phosphate. Biochem. J. 97: 95-103. 11.
Kato, K.Y., Hamaguchi, Y., Fukui, H.
and
4-methylumbelliferyl Ishikawa, E. (1976)
Enzyme-linked immunoassay. Conjugation of rabbit anti-(human immunoglobulin G) antibody with ß-B-galactosidase from Escherichia coli and its use for human immunoglobulin G assay. Eur. J.
Biochem.
62:185-192.
P.H. and Alexander, D.J. (1983) Antigenic variations of Newcastle disease virus strains detected by monoclonal antibodies. Arch. Virol. 75: 243-253.
12.
Russell,
13.
Nishikawa,
K.
antibodies
to
,
et al. (1983) Monoclonal of Newcastle disease virus:
Isomura, S., Suzuki, S. the HN
glycoprotein
835
biological characterization Virology 130:318-330. Address
and
use
for
strain
reprint request
comparison.
to:
Jonathan Wong Defence Research Establishment Suffield Box 4000 Medicine Hat, Alberta Canada TÍA 8K6
Received for publication: 4/30/92 after revision as a short ccmnunication: 6/5/92
Accepted
836
This article has been cited by: 1. John J. Weiland, James V. Anderson, Brant B. Bigger. 2007. Inexpensive chemifluorescent detection of antibody–alkaline phosphatase conjugates on Western blots using 4-methylumbelliferyl phosphate. Analytical Biochemistry 361:1, 140-142. [CrossRef] 2. W LEE, H THOMPSON, J HALL, R FULTON, J WONG. 1993. Rapid immunofiltration assay of Newcastle disease virus using a silicon sensor. Journal of Immunological Methods 166:1, 123-131. [CrossRef]
