- Email: [email protected]
Differentiation of adenovirus hexon epitopes with monoclonal antibodies by gel diffusion assays
Molecular Immunology, Vol. 22, No. 8, pp. 967-971, 1985
0161-5890/85 $3.00 + 0.00 Q 1985 Pergamon Press Ltd
Printed in Great Britain
DIFFERENTIATION OF ADENOVIRUS HEXON EPITOPES WITH MONOCLONAL ANTIBODIES BY GEL DIFFUSION ASSAYS &A
ADAM,*
ISTVAN NAsz,*t
ANNA LENGYEL,*
J~NOS ERDEI$
and J~ZSEF FACHET$ *Institute of Microbiology, iInstitute of Pathophysiology,
Semmelweis Medical School, Budapest, and University Medical School, Debrecen, Hungary
(Received 26 December
1984; accepted 6 February 1985)
Abstract-Nine monoclonal antibodies (MAbs) directed against crystallized human adenovirus type I (Ad h 1) hexon were tested with purified homologous and heterologous hexon preparations by gel diffusion. Six MAbs formed a single line with the homologous hexon in a 2-well pattern, and 3 MAbs formed lines only in biclonal combinations with an appropriate MAb. All of the 6 precipitating MAbs formed a continuous line of complete identity when tested simultaneously against homologous and different heterologous hexons. With Ad h 1 hexon a line of double partial identity (double spur) was formed when some pair combinations of 2 MAbs were placed in 2 juxtaposed wells. Other MAbs in the adjacent wells formed a line of identity. The MAbs could be divided into 2 antibody groups (groups A and B) based on this phenomenon. Members of antibody groups A and B apparently identified 2 sterically distinct epitopes: one of them is presumably the genus-specific epitope of the hexons (group A) and the other(s) should be intertype-specific epitope(s). Thus, the gel diffusion method can be used for selecting pairs of MAbs for their specificity to sterically independent epitopes. Mixtures of 2 MAbs belonging to the different antibody groups formed double lines with Ad h 1 hexons. Members of group A showed some helper effect to the members of group B for their precipitin line formation.
INTRODUCTION
C), 12 (subgenus A), 7 and 35 (subgenus B) and 8,9, 10 and 13 (subgenus D) were propagated in HEp-2 cells and the soluble protein components were extracted and separated by ultracentrifugation and purified by repeated anion exchange chromatography and gel filtration as previously described (Lengyel and N&Z, 1970). Bovine adenovirus type 3 (Ad bos 3) was kindly supplied by Dr A. Bartha. Highly purified hexon capsomers of adenovirus type 1 were crystallized by dialysis against 0.5 M acetate buffer (pH 4.5) (Adim and N&z, 1980). Antihexon immune sera were produced in rabbits as described previously (NBsz et al., 1972).
Polyclonal antihexon immune sera form 1 precipitin line when counterdiffused to purified hexon antigen prepared from homologous adenovirus types (species). Immune sera form also one precipitin line in a 2-well double-diffusion pattern with hexons prepared from heterologous adenovirus types containing a genus-specific antigenic determinant (Wigand et al., 1982). Examining 2 different hexons in a 3-well pattern, a line of partial identity is formed, i.e. a spur formation appears between the homologous and heterologous hexons. Studying afly 2 different hexons with a third (heterotypic) polyclonal antihexon serum, no spur formation could be detected, i.e. a single continuous line of complete identity is formed (N&z et al., 1972). This report gives information on the findings that adenovirus hexon antigen may form 2 precipitation lines and a double spur when tested with 2 monoclonal antibodies (MAbs) directed against 2 different hexon epitopes, and that this method can be a tool for the easy differentiation between MAbs for their specificity to independent epitopes.
MAbs
MATERIALS AND METHODS
Antigens Human
and antisera adenovirus
types
1, 2, 5 and 6 (subgenus
tAddress correspondence to: Istvin NBsz, Institute of Microbiology, Semmelweis University Medical School, Nagyvkrad tt’r 4, H-1089 Budapest, Hungary. 967
Production of hybridomas and mouse ascitic fluids were carried out as described earlier (Adim et al., 1985). Washed and redissolved crystals of adenovirus type 1 hexon were used for immunization of the spleen-cell donor BALB/c female mice. The Sp2/0 nonsecreting parental myeloma cell line was used for cell fusion. An adenovirus type 1 hexon preparation was used for the selection of specific MAbs. The antibodies were reactive specifically with both the crystallized and noncrystallized purified homologous hexon preparations by an enzyme-linked immunosorbent assay (ELISA) and a passive hemagglutination test (HA) (Adim et al., in press, 1985). In a complement fixation test with type 1 hexon, only ascites lA3, 2C2 and H12 were positive in a 1: 100 titre and 1D6 in a 1:75 titre. In virus neutral-
ization experiments the 9 MAbs failed to neutralize the infectivity of adenovirus type 1, even in different biclonal combinations. The antibody isotype of all of the 9 MAbs was found to be IgG2a (Adam et al., 1985). Gel dzfusion One per cent agar gel (Difco, Noble) was used as described earlier (N&z et al., 1972), containing 5% PFG 4000. The well dia was 2 or 3 mm, and the well distance was 3 or 4mm. The antigens were used in 0.5 mg/ml concn. The ascitic fluids were used neat or diluted to the optimal ratio for precipitation, Some of the antibodies were mixed and used as biclonal antibodies (see Results).
Fig. 1. Gel diffusion test of MAb lA3 (central purified adenovirus hexons of the indicated
8
well) with types.
1
RESULTS
Gel difSusion analysis of MAbs For the present studies, 9 ascitic fluids were selected, based on their different reactivity patterns with the 10 heterologous hexon types (Table 1). These MAbs were tested against hexons of human adenovirus type 1 (Ad h 1) and Ad bos 3 for precipitin line formation in agar gel containing 5% PEG. Six MAbs formed a single precipitin line with Ad h 1 and 2 of them with Ad bos 3 hexons too in a 2-well pattern (Table 2). Three of the 9 MAbs formed lines only in biclonal combinations with an appropriate MAb (see later). When 1 precipitating MAb was tested by radial diffusion in a 5- or 7-well pattern against homologous Ad h 1 and heterologous hexons belonging to different subgenera, a continuous line of complete identity was formed (Fig. 1). In contrast, when a polyclonal antihexon serum was tested in a similar pattern, spur formation appeared between the homologous and the heterologous hexons (Fig. 2). Polyclonal antihexon sera against type 1, 5 and 8 hexons spurred over all precipitating MAbs tested in a 3-well pattern using either homologous or heterologous hexons as antigens (not shown). When the homologous type 1 hexon or one of the used heterologous hexons were placed in a central well and the surrounding wells were filled with different precipitating MAbs, a confluent line of complete identity was formed between some MAbs.
Table
Fig. 2. Gel diffusion test of polyclonal antihexon serum of Ad h 1 (central well) with purified hexon of the indicated types.
In contrast, some MAbs mutually spurred over the other MAb in the adjacent wells and formed lines of double partial identity (Fig. 3). By testing the precipitating MAbs this way, they could be divided into 2 groups (antibody groups A and B). To group A belong ascites lA3 and 2C2, and to group B ascites H12, 2A1, lD6 and 2Dl. The members of the same group always formed lines of complete identity when placed in adjacent wells either with homologous or heterologous hexons. In contrast, when a member of group A and a member of group B were placed in adjacent wells and diffused against homologous or heterologous hexons, double-spur formation was always found (Fig. 4, Table 3). Cooperative reaction between 2 MAbs Mixtures of 2 MAbs belonging to different antibody groups were tested against homologous type 1 hexon by double diffusion. Six doubling dilutions of ascites lD6 (group B) in ascites lA3 (group A) (diluted 1:2) were diffused against highly purified adenovirus type 1 hexon preparations containing
1. Cross-reactivity pattern of MAbs with 10 heterologous hexon adenovirus tvues 2. 5. 6. 7. 8. 9, 10, 12, 13 and 35) MAbs 2C2
H12
ELISA
10”
10
IO
HA
10
10
10
Method
lA3
types
(human
to Ad h 1
2Al
ID6
2Dl
1B2
2A4
IBI
(1;)’
(3:)
(12835)
(12835)
(P,)
(12:35)
10
10 (&
(8, I:, 12)
(8,qo 12,35i
(768 10,‘3$
“Number of heterologous hexons reacting with the given MAbs. ‘Figures in parentheses indicate hexon types not reacting with the given MAb
969
Gel diffusion of monoclonal antibodies to adenovirus hexon Table 4. Double-spur formation of MAbs and biclonal mixtures of MAbs in juxtaposed wells tested against ditkent hexons”
201 2Al
2A4
lD6
MAb
H12
IA3 2c2 HI2 2Al ID6 2DI
lD6 Fig. 3. Gel diffusion tests of the indicated MAbs with adenovirus type 1 hexon (upper central well) and with type 5 hexon (lower central well).
lD6 12
2c2 Fig.
4. Gel diffusion test of the indicated MAbs adenovirus type 5 hexon (unlabelled well).
with
Fig. 5. Gel diffusion test of biclonal mixtures with adenovirus type 1 hexon (central well). Serial dilutions of MAb lD6 in MAb IA3 (outer wells) were filled clockwise (first dilution of MAb lD6 at the upper left).
0.5 mg/ml protein. This biclonal mixture formed single lines at 1:2-l: 8 dilutions and double lines at 1: 16, 1: 32 and 1: 64 dilutions of ascites lD6 (Fig. 5). This dilution of ascites lD6 does not form precipitin line alone; the second line therefore should be formed as a result of cooperative reaction with ascites lA3.
Table 2. Precipitation
tests of MAbs hexons”
HeXI”
“+ indicates
lA3
2C2
H12
2Al
ID6
2Dl
lB2
1Bl
2A4
+ +
+ +
+ _
+ _
+ _
+ _
-
-
-
line formation;
_
“Abbreviations
+ + _ _ _
+ + _ _
as in Table 3.
A double line formed also between the members of the different antibody groups used in appropriate dilutions in a 3-well pattern when the distance between the 2 antibody wells was twice the distance betweenthe antigen well and the antibody wells. The localization of the second line is characteristic of the 2 antibody wells and indicates also that this precipitin line was formed by the cooperative reaction of 2 MAbs (Fig. 6) belonging to the different antibody groups. The 3 MAbs separately non-precipitating (2A4, 1Bl and 1B2) (Table 2) were mixed in equal vols (neat ascites) in all the 3 possible combinations and diffused against type 1 hexon. These biclonal mixtures formed 1 hardly visible precipitin line in all combinations (not shown). When these 3 MAbs were premixed separately with type 1 hexon and counterdiffused against either of the 2 other MAbs, no precipitin line formation could be detected. Mixtures of 1 of these non-precipitating MAbs and 1 member of group A formed a line of double partial identity with the latter MAb, indicating that these separately non-precipitating MAbs belong to the group B of MAbs and that the members of group A can provide some cooperative or helper effect for their precipitin line formation (Fig. 7, Table 4).
with Ad h 1 and Ad bos3
MAbs to Ad h 1
Ad h I Ad bos 3
+ + _
lB2 1Bl + IA3 or 2C2
~ indicates
Fig. 6. Gel diffusion of MAb 2C2 (diluted 1: 10) and the separately non-precipitating dilution (1: 10) of MAb H12 with adenovirus type 1 hexon (unlabelled well).
lA3 (1:3)
no line formation.
Table 3. Double-spur formation of MAbs” in juxtaposed wells tested against different hexons” MAb
IA3
2C2
HI2
2Al
ID6
2Dl
IA3 2c2 HI2 2Al ID6 2Dl
+ t + +
~ t t t t
+ t -
+ t ~ ~
+ t -
+ t -
“MAbs against Ad h I. h+ indicates double-spur formation; indicates no double-spur formation.
lA3y:lO)
lA3(1:3)
2A4
2i4 lA3(1:20) -
Fig. 7. Gel diffusion test of MAb lA3 and biclonal mixtures of the different dilutions (indicated) of MAb lA3 and the separately non-precipitating MAb 2A4 (neat) with adenovirus type 1 hexon (central well).
DISCUSSION
previous experiments MAbs were tested against crystallized Ad h 1 and it was demonstrated that there are different, partially similar or overlapping epitopes to be found on the surface of the adenovirus hexon capsomer. These Ad h 1 hexon related epitopes can be found in various combinations on the surface of the different heterologous hexon types, characteristic of the given hexon type (Ad8m et t61., in press, 1985). Considering that the complete hexon protein consists of 3 identical poiypeptide subunits, presumably identical epitopes can be found on each subunit. Consequently it can be supposed that 3 identical copies of the different epitopes are present on the surface of a complete hexon. By testing highly purified hexon protein of Ad h 1 with a panel of MAbs to crystallized Ad h 1 in gel diffusion assays, it was shown that adenovirus hexon antigen formed a line of double partial identity (double spur) when some pair combinations of 2 MAbs were placed in 2 juxtaposed wells. Other MAbs in the adjacent wells formed a confluent line of complete identity when tested either with the homologous Ad h 1 or with different heterologous hexon antigens (Figs 3 and 4). Based on the doublespur-forming ability, the MAbs could be divided into 2 antibody groups (groups A and B). Any member of group A formed a line of identity with the other member of group A in juxtaposed wells when diffused against hexon antigens and formed a double spur with any member of group B, and vice versa. The biclonal mixtures of MAbs belonging to the different antibody groups formed double lines with 1 purified either homologous or heterologous hexon preparation (Fig. 5). The finding that the 9 Ad 1 antihexon MAbs could be divided into 2 mutually exclusive groups indicates that the members of the 2 groups identified 2 sterically distinct epitopes on the surface of hexon antigens. The MAbs which do not form a double spur (i.e. belonging to the same group) presumably react with the same or overlapping epitope. The different reactivity patterns of the 9 MAbs by ELISA and HA (Table 1) indicate that this gel diffusion method cannot distinguish 2 MAbs of different affinities to the same epitope nor of different specificities of 2 MAbs to overlapping epitopes. However, it can be used as a simple procedure for testing a group of MAbs for their specificity to sterically independent epitopes, because no expensive equipment and no labelling of the reactants is required in this method. The mechanism of double-spur and double-line formation is not easily explained by the conventional understanding of immunoprecipitation. However, the hypothesis of Molinaro and Eby (1984) is quite acceptable in our test system, i.e. the double spur indicates that the MAb formed soluble immune complexes with the hexons that diffused past the major precipitin lines and reacted with the other MAb. The In
our
double line indicates a bimodal distribution of the precipitating immune complexes. The cooperative or helper effect can be explained by the hypothesis of Molinaro er al. (1984), i.e. a separately nonprecipitating MAb formed hexon dimers (dimerizing MAb) which can be linked into larger insoluble immune complexes by the appropriate second MAb (linking MAb). In the experiments of Steensgaard et al. (1980) combinations of 2 MAbs directed against spatially distinct epitopes produced complexes exhibiting marked turbidity. In our experiments, the linking MAbs (group A) formed line with the hexon separately too (Figs 6 and 7). However, some MAb molecules remained apparently free or also soluble complexes formed beside the insoluble ones, and these diffused past the major precipitin lines and reacted with the hexon dimers formed by the other dimerizing MAb. The epitopes identified by MAbs lA3 and 2C2 (group A) should be the genus-specific epitope of adenovirus hexon (Wigand et al., 1982) because they reacted with the hexons of all human adenovirus types examined by ELISA and HA and with Ad bos 3 in a gel precipitation experiment. The epitope(s) identified by the members of antibody group B should be intertypespecific (Table 1) and/or overlapping to which the different MAbs have different affinities and they should be localized at the hexon surface on sterically distinct places relative to the genus-specific epitope. This is consistent with the double-spur and doubleline forming ability of the different MAbs and with the helper effect of group A members for precipitin line formation by some members of group B.
Acknott,le~gemen~.~-We thank Miss Zsuzsanna and I. Balazs for excellent technical assistance.
Bakonyi
REFERENCES
Adam E., Erdei J., Lengyel A.. Berencsi Gy., Fachet J. and NQsz 1. Delineation of antigenic deternlinants of adenovirus hexons by means of monoclonal antibodies. iniert;iroiogy (in press). Adam E., Erdei J., Lengyel A., Katona A., Berencsi Gy., Fachet J. and Ndsz I. (1985) Reactivity of mouse ascitic fluids containing monoclonal antibodies directed against adenovirus hexon. Acta microbial. hung. 32, 113-1~4. Adam E. and N&z I. (\ 1980) Electron microsconic studies on the crystaliization of adenovirus hexon capsomers. Infervirology 13, l-6. Lengyel A. and N&z I. (1970) Soluble components of adenovirus type 8. J. Viral. 6, 406-413. Molinaro G. A. and Eby W. C. (1984) One antigen may form two precipitin lines and two spurs when tested with two monoclonal antibodies by gel diffusion assays. Molec. ~~~~~. 21, 181-184. Molinaro G. A., Ebv W. C., Molinaro C. A.. Bartholomew R. M. and David G. (1984) Two monoclonal antibodies to two different epitopes of human growth hormone form a precipitin line when counterdiffused as soluble immune complexes. Molec. Immun. 21, 771-774. Nasz I., Lengyel A. and Cserba I. (1972) Comparative studies of adenovirus hexon antigens. Arch ges. Virusjiirxch. 36, 80-92. 1
Gel diffusion
of monoclonal
Steensgaard J., Jacobson C., Lowe J., Hardie D., Ling N. R. and Jefferis R. (1980) The development of difference turbidimetric analysis for monoclonal antibodies to human IgG. Molec. Immun. 17, 1315-I 318.
antibodies
to adenovirus
hexon
971
Wigand R., Bartha A., Dreizin R. S., Esche H., Ginsberg H. S., Green M., Hierholzer J. C., Kalter S. S., McFerran J. B., Petterson U., Russel W. C. and Wade11 G. (1982) Adenoviridae: second report. Intervirology 18, 169-176.
0161-5890/85 $3.00 + 0.00 Q 1985 Pergamon Press Ltd
Printed in Great Britain
DIFFERENTIATION OF ADENOVIRUS HEXON EPITOPES WITH MONOCLONAL ANTIBODIES BY GEL DIFFUSION ASSAYS &A
ADAM,*
ISTVAN NAsz,*t
ANNA LENGYEL,*
J~NOS ERDEI$
and J~ZSEF FACHET$ *Institute of Microbiology, iInstitute of Pathophysiology,
Semmelweis Medical School, Budapest, and University Medical School, Debrecen, Hungary
(Received 26 December
1984; accepted 6 February 1985)
Abstract-Nine monoclonal antibodies (MAbs) directed against crystallized human adenovirus type I (Ad h 1) hexon were tested with purified homologous and heterologous hexon preparations by gel diffusion. Six MAbs formed a single line with the homologous hexon in a 2-well pattern, and 3 MAbs formed lines only in biclonal combinations with an appropriate MAb. All of the 6 precipitating MAbs formed a continuous line of complete identity when tested simultaneously against homologous and different heterologous hexons. With Ad h 1 hexon a line of double partial identity (double spur) was formed when some pair combinations of 2 MAbs were placed in 2 juxtaposed wells. Other MAbs in the adjacent wells formed a line of identity. The MAbs could be divided into 2 antibody groups (groups A and B) based on this phenomenon. Members of antibody groups A and B apparently identified 2 sterically distinct epitopes: one of them is presumably the genus-specific epitope of the hexons (group A) and the other(s) should be intertype-specific epitope(s). Thus, the gel diffusion method can be used for selecting pairs of MAbs for their specificity to sterically independent epitopes. Mixtures of 2 MAbs belonging to the different antibody groups formed double lines with Ad h 1 hexons. Members of group A showed some helper effect to the members of group B for their precipitin line formation.
INTRODUCTION
C), 12 (subgenus A), 7 and 35 (subgenus B) and 8,9, 10 and 13 (subgenus D) were propagated in HEp-2 cells and the soluble protein components were extracted and separated by ultracentrifugation and purified by repeated anion exchange chromatography and gel filtration as previously described (Lengyel and N&Z, 1970). Bovine adenovirus type 3 (Ad bos 3) was kindly supplied by Dr A. Bartha. Highly purified hexon capsomers of adenovirus type 1 were crystallized by dialysis against 0.5 M acetate buffer (pH 4.5) (Adim and N&z, 1980). Antihexon immune sera were produced in rabbits as described previously (NBsz et al., 1972).
Polyclonal antihexon immune sera form 1 precipitin line when counterdiffused to purified hexon antigen prepared from homologous adenovirus types (species). Immune sera form also one precipitin line in a 2-well double-diffusion pattern with hexons prepared from heterologous adenovirus types containing a genus-specific antigenic determinant (Wigand et al., 1982). Examining 2 different hexons in a 3-well pattern, a line of partial identity is formed, i.e. a spur formation appears between the homologous and heterologous hexons. Studying afly 2 different hexons with a third (heterotypic) polyclonal antihexon serum, no spur formation could be detected, i.e. a single continuous line of complete identity is formed (N&z et al., 1972). This report gives information on the findings that adenovirus hexon antigen may form 2 precipitation lines and a double spur when tested with 2 monoclonal antibodies (MAbs) directed against 2 different hexon epitopes, and that this method can be a tool for the easy differentiation between MAbs for their specificity to independent epitopes.
MAbs
MATERIALS AND METHODS
Antigens Human
and antisera adenovirus
types
1, 2, 5 and 6 (subgenus
tAddress correspondence to: Istvin NBsz, Institute of Microbiology, Semmelweis University Medical School, Nagyvkrad tt’r 4, H-1089 Budapest, Hungary. 967
Production of hybridomas and mouse ascitic fluids were carried out as described earlier (Adim et al., 1985). Washed and redissolved crystals of adenovirus type 1 hexon were used for immunization of the spleen-cell donor BALB/c female mice. The Sp2/0 nonsecreting parental myeloma cell line was used for cell fusion. An adenovirus type 1 hexon preparation was used for the selection of specific MAbs. The antibodies were reactive specifically with both the crystallized and noncrystallized purified homologous hexon preparations by an enzyme-linked immunosorbent assay (ELISA) and a passive hemagglutination test (HA) (Adim et al., in press, 1985). In a complement fixation test with type 1 hexon, only ascites lA3, 2C2 and H12 were positive in a 1: 100 titre and 1D6 in a 1:75 titre. In virus neutral-
ization experiments the 9 MAbs failed to neutralize the infectivity of adenovirus type 1, even in different biclonal combinations. The antibody isotype of all of the 9 MAbs was found to be IgG2a (Adam et al., 1985). Gel dzfusion One per cent agar gel (Difco, Noble) was used as described earlier (N&z et al., 1972), containing 5% PFG 4000. The well dia was 2 or 3 mm, and the well distance was 3 or 4mm. The antigens were used in 0.5 mg/ml concn. The ascitic fluids were used neat or diluted to the optimal ratio for precipitation, Some of the antibodies were mixed and used as biclonal antibodies (see Results).
Fig. 1. Gel diffusion test of MAb lA3 (central purified adenovirus hexons of the indicated
8
well) with types.
1
RESULTS
Gel difSusion analysis of MAbs For the present studies, 9 ascitic fluids were selected, based on their different reactivity patterns with the 10 heterologous hexon types (Table 1). These MAbs were tested against hexons of human adenovirus type 1 (Ad h 1) and Ad bos 3 for precipitin line formation in agar gel containing 5% PEG. Six MAbs formed a single precipitin line with Ad h 1 and 2 of them with Ad bos 3 hexons too in a 2-well pattern (Table 2). Three of the 9 MAbs formed lines only in biclonal combinations with an appropriate MAb (see later). When 1 precipitating MAb was tested by radial diffusion in a 5- or 7-well pattern against homologous Ad h 1 and heterologous hexons belonging to different subgenera, a continuous line of complete identity was formed (Fig. 1). In contrast, when a polyclonal antihexon serum was tested in a similar pattern, spur formation appeared between the homologous and the heterologous hexons (Fig. 2). Polyclonal antihexon sera against type 1, 5 and 8 hexons spurred over all precipitating MAbs tested in a 3-well pattern using either homologous or heterologous hexons as antigens (not shown). When the homologous type 1 hexon or one of the used heterologous hexons were placed in a central well and the surrounding wells were filled with different precipitating MAbs, a confluent line of complete identity was formed between some MAbs.
Table
Fig. 2. Gel diffusion test of polyclonal antihexon serum of Ad h 1 (central well) with purified hexon of the indicated types.
In contrast, some MAbs mutually spurred over the other MAb in the adjacent wells and formed lines of double partial identity (Fig. 3). By testing the precipitating MAbs this way, they could be divided into 2 groups (antibody groups A and B). To group A belong ascites lA3 and 2C2, and to group B ascites H12, 2A1, lD6 and 2Dl. The members of the same group always formed lines of complete identity when placed in adjacent wells either with homologous or heterologous hexons. In contrast, when a member of group A and a member of group B were placed in adjacent wells and diffused against homologous or heterologous hexons, double-spur formation was always found (Fig. 4, Table 3). Cooperative reaction between 2 MAbs Mixtures of 2 MAbs belonging to different antibody groups were tested against homologous type 1 hexon by double diffusion. Six doubling dilutions of ascites lD6 (group B) in ascites lA3 (group A) (diluted 1:2) were diffused against highly purified adenovirus type 1 hexon preparations containing
1. Cross-reactivity pattern of MAbs with 10 heterologous hexon adenovirus tvues 2. 5. 6. 7. 8. 9, 10, 12, 13 and 35) MAbs 2C2
H12
ELISA
10”
10
IO
HA
10
10
10
Method
lA3
types
(human
to Ad h 1
2Al
ID6
2Dl
1B2
2A4
IBI
(1;)’
(3:)
(12835)
(12835)
(P,)
(12:35)
10
10 (&
(8, I:, 12)
(8,qo 12,35i
(768 10,‘3$
“Number of heterologous hexons reacting with the given MAbs. ‘Figures in parentheses indicate hexon types not reacting with the given MAb
969
Gel diffusion of monoclonal antibodies to adenovirus hexon Table 4. Double-spur formation of MAbs and biclonal mixtures of MAbs in juxtaposed wells tested against ditkent hexons”
201 2Al
2A4
lD6
MAb
H12
IA3 2c2 HI2 2Al ID6 2DI
lD6 Fig. 3. Gel diffusion tests of the indicated MAbs with adenovirus type 1 hexon (upper central well) and with type 5 hexon (lower central well).
lD6 12
2c2 Fig.
4. Gel diffusion test of the indicated MAbs adenovirus type 5 hexon (unlabelled well).
with
Fig. 5. Gel diffusion test of biclonal mixtures with adenovirus type 1 hexon (central well). Serial dilutions of MAb lD6 in MAb IA3 (outer wells) were filled clockwise (first dilution of MAb lD6 at the upper left).
0.5 mg/ml protein. This biclonal mixture formed single lines at 1:2-l: 8 dilutions and double lines at 1: 16, 1: 32 and 1: 64 dilutions of ascites lD6 (Fig. 5). This dilution of ascites lD6 does not form precipitin line alone; the second line therefore should be formed as a result of cooperative reaction with ascites lA3.
Table 2. Precipitation
tests of MAbs hexons”
HeXI”
“+ indicates
lA3
2C2
H12
2Al
ID6
2Dl
lB2
1Bl
2A4
+ +
+ +
+ _
+ _
+ _
+ _
-
-
-
line formation;
_
“Abbreviations
+ + _ _ _
+ + _ _
as in Table 3.
A double line formed also between the members of the different antibody groups used in appropriate dilutions in a 3-well pattern when the distance between the 2 antibody wells was twice the distance betweenthe antigen well and the antibody wells. The localization of the second line is characteristic of the 2 antibody wells and indicates also that this precipitin line was formed by the cooperative reaction of 2 MAbs (Fig. 6) belonging to the different antibody groups. The 3 MAbs separately non-precipitating (2A4, 1Bl and 1B2) (Table 2) were mixed in equal vols (neat ascites) in all the 3 possible combinations and diffused against type 1 hexon. These biclonal mixtures formed 1 hardly visible precipitin line in all combinations (not shown). When these 3 MAbs were premixed separately with type 1 hexon and counterdiffused against either of the 2 other MAbs, no precipitin line formation could be detected. Mixtures of 1 of these non-precipitating MAbs and 1 member of group A formed a line of double partial identity with the latter MAb, indicating that these separately non-precipitating MAbs belong to the group B of MAbs and that the members of group A can provide some cooperative or helper effect for their precipitin line formation (Fig. 7, Table 4).
with Ad h 1 and Ad bos3
MAbs to Ad h 1
Ad h I Ad bos 3
+ + _
lB2 1Bl + IA3 or 2C2
~ indicates
Fig. 6. Gel diffusion of MAb 2C2 (diluted 1: 10) and the separately non-precipitating dilution (1: 10) of MAb H12 with adenovirus type 1 hexon (unlabelled well).
lA3 (1:3)
no line formation.
Table 3. Double-spur formation of MAbs” in juxtaposed wells tested against different hexons” MAb
IA3
2C2
HI2
2Al
ID6
2Dl
IA3 2c2 HI2 2Al ID6 2Dl
+ t + +
~ t t t t
+ t -
+ t ~ ~
+ t -
+ t -
“MAbs against Ad h I. h+ indicates double-spur formation; indicates no double-spur formation.
lA3y:lO)
lA3(1:3)
2A4
2i4 lA3(1:20) -
Fig. 7. Gel diffusion test of MAb lA3 and biclonal mixtures of the different dilutions (indicated) of MAb lA3 and the separately non-precipitating MAb 2A4 (neat) with adenovirus type 1 hexon (central well).
DISCUSSION
previous experiments MAbs were tested against crystallized Ad h 1 and it was demonstrated that there are different, partially similar or overlapping epitopes to be found on the surface of the adenovirus hexon capsomer. These Ad h 1 hexon related epitopes can be found in various combinations on the surface of the different heterologous hexon types, characteristic of the given hexon type (Ad8m et t61., in press, 1985). Considering that the complete hexon protein consists of 3 identical poiypeptide subunits, presumably identical epitopes can be found on each subunit. Consequently it can be supposed that 3 identical copies of the different epitopes are present on the surface of a complete hexon. By testing highly purified hexon protein of Ad h 1 with a panel of MAbs to crystallized Ad h 1 in gel diffusion assays, it was shown that adenovirus hexon antigen formed a line of double partial identity (double spur) when some pair combinations of 2 MAbs were placed in 2 juxtaposed wells. Other MAbs in the adjacent wells formed a confluent line of complete identity when tested either with the homologous Ad h 1 or with different heterologous hexon antigens (Figs 3 and 4). Based on the doublespur-forming ability, the MAbs could be divided into 2 antibody groups (groups A and B). Any member of group A formed a line of identity with the other member of group A in juxtaposed wells when diffused against hexon antigens and formed a double spur with any member of group B, and vice versa. The biclonal mixtures of MAbs belonging to the different antibody groups formed double lines with 1 purified either homologous or heterologous hexon preparation (Fig. 5). The finding that the 9 Ad 1 antihexon MAbs could be divided into 2 mutually exclusive groups indicates that the members of the 2 groups identified 2 sterically distinct epitopes on the surface of hexon antigens. The MAbs which do not form a double spur (i.e. belonging to the same group) presumably react with the same or overlapping epitope. The different reactivity patterns of the 9 MAbs by ELISA and HA (Table 1) indicate that this gel diffusion method cannot distinguish 2 MAbs of different affinities to the same epitope nor of different specificities of 2 MAbs to overlapping epitopes. However, it can be used as a simple procedure for testing a group of MAbs for their specificity to sterically independent epitopes, because no expensive equipment and no labelling of the reactants is required in this method. The mechanism of double-spur and double-line formation is not easily explained by the conventional understanding of immunoprecipitation. However, the hypothesis of Molinaro and Eby (1984) is quite acceptable in our test system, i.e. the double spur indicates that the MAb formed soluble immune complexes with the hexons that diffused past the major precipitin lines and reacted with the other MAb. The In
our
double line indicates a bimodal distribution of the precipitating immune complexes. The cooperative or helper effect can be explained by the hypothesis of Molinaro er al. (1984), i.e. a separately nonprecipitating MAb formed hexon dimers (dimerizing MAb) which can be linked into larger insoluble immune complexes by the appropriate second MAb (linking MAb). In the experiments of Steensgaard et al. (1980) combinations of 2 MAbs directed against spatially distinct epitopes produced complexes exhibiting marked turbidity. In our experiments, the linking MAbs (group A) formed line with the hexon separately too (Figs 6 and 7). However, some MAb molecules remained apparently free or also soluble complexes formed beside the insoluble ones, and these diffused past the major precipitin lines and reacted with the hexon dimers formed by the other dimerizing MAb. The epitopes identified by MAbs lA3 and 2C2 (group A) should be the genus-specific epitope of adenovirus hexon (Wigand et al., 1982) because they reacted with the hexons of all human adenovirus types examined by ELISA and HA and with Ad bos 3 in a gel precipitation experiment. The epitope(s) identified by the members of antibody group B should be intertypespecific (Table 1) and/or overlapping to which the different MAbs have different affinities and they should be localized at the hexon surface on sterically distinct places relative to the genus-specific epitope. This is consistent with the double-spur and doubleline forming ability of the different MAbs and with the helper effect of group A members for precipitin line formation by some members of group B.
Acknott,le~gemen~.~-We thank Miss Zsuzsanna and I. Balazs for excellent technical assistance.
Bakonyi
REFERENCES
Adam E., Erdei J., Lengyel A.. Berencsi Gy., Fachet J. and NQsz 1. Delineation of antigenic deternlinants of adenovirus hexons by means of monoclonal antibodies. iniert;iroiogy (in press). Adam E., Erdei J., Lengyel A., Katona A., Berencsi Gy., Fachet J. and Ndsz I. (1985) Reactivity of mouse ascitic fluids containing monoclonal antibodies directed against adenovirus hexon. Acta microbial. hung. 32, 113-1~4. Adam E. and N&z I. (\ 1980) Electron microsconic studies on the crystaliization of adenovirus hexon capsomers. Infervirology 13, l-6. Lengyel A. and N&z I. (1970) Soluble components of adenovirus type 8. J. Viral. 6, 406-413. Molinaro G. A. and Eby W. C. (1984) One antigen may form two precipitin lines and two spurs when tested with two monoclonal antibodies by gel diffusion assays. Molec. ~~~~~. 21, 181-184. Molinaro G. A., Ebv W. C., Molinaro C. A.. Bartholomew R. M. and David G. (1984) Two monoclonal antibodies to two different epitopes of human growth hormone form a precipitin line when counterdiffused as soluble immune complexes. Molec. Immun. 21, 771-774. Nasz I., Lengyel A. and Cserba I. (1972) Comparative studies of adenovirus hexon antigens. Arch ges. Virusjiirxch. 36, 80-92. 1
Gel diffusion
of monoclonal
Steensgaard J., Jacobson C., Lowe J., Hardie D., Ling N. R. and Jefferis R. (1980) The development of difference turbidimetric analysis for monoclonal antibodies to human IgG. Molec. Immun. 17, 1315-I 318.
antibodies
to adenovirus
hexon
971
Wigand R., Bartha A., Dreizin R. S., Esche H., Ginsberg H. S., Green M., Hierholzer J. C., Kalter S. S., McFerran J. B., Petterson U., Russel W. C. and Wade11 G. (1982) Adenoviridae: second report. Intervirology 18, 169-176.