Download Differences in conformation of type 3 poliovirus antigenic sites on

Journal of General Virology (1990), 71, 1271-1274. Printed in Great Britain
1271
Differences in conformation of type 3 poliovirus antigenic sites on
non-infectious empty particles and infectious virus
Morag Ferguson* and Philip D. Minor
National Institute for Biological Standards and Control, Blanche Lane, South Mimms, Potters Bar,
Hertfordshire E N 6 3QG, U.K.
A panel of monoclonal antibodies which react with
empty non-infectious type 3 poliovirus particles (C
antigen) but not infectious virus (D antigen) were
characterized for their reactivity with C antigen particles
derived from neutralization-resistant virus strains
which had single amino acid substitutions at each of the
antigenic sites. Antibodies were identified which failed
to bind to variant viruses with modifications at each o f
antigenic sites 2b, 3b and 4 indicating that the same
amino acid sequences involved in the neutralization o f
infectious virus are also present on the surface of noninfectious particles but in different configurations.
Introduction
particles may be immature virions lacking viral R N A or
empty capsids formed during the process of attachment
and uncoating. On sucrose gradients the two particles
sediment at 155S and 80S respectively (FernandezTomas & Baltimore, 1973; Minor et al., 1980). These
particles have been designated D and C antigen
respectively (Mayer et al., 1957). Studies with polyclonal
and monoclonal antibodies have shown that, as well as
possessing distinct antigenic determinants, the 155S and
80S peaks obtained with type 3 poliovirus also share at
least one antigenic determinant (Schild et al., 1980;
Ferguson et al., 1984).
This paper describes studies with monoclonal antibodies and synthetic peptides which demonstrate that
the amino acid sequences of antigenic sites involved in
the neutralization of infectious virus are also present on
the surface of non-infectious empty particles, but in
different configurations.
Poliovirus is a member of the picornaviridae and has an
icosahedral structure composed of 60 copies of each of
four capsid proteins designated VP1, VP2, VP3 and
VP4. The three-dimensional structure of both types 1 and
3 poliovirus have been determined (Hogle et al., 1985;
Filman et al., 1989) and it has been shown that the amino
acid sequences from capsid proteins VP1, VP2 and VP3,
but not VP4, are exposed on the surface of infectious
virus.
The antigenic structure of types 1 and 3 poliovirus
has been analysed by means of the selection of mutant
viruses resistant to neutralization by monoclonal antibodies and the resulting amino acid substitutions
identified by primer extension sequencing (Minor et al.,
1983, 1985, 1986, 1987; Page et al., 1988; Wiegers et al.,
1989). Four antigenic sites have been identified on both
types 1 and 3 poliovirus. Some of these sites are complex,
comprising amino acid sequences from more than one
capsid protein. The sites identified so far are site 1, VP1
residues 89 to 100, 141 to 152, 166 and 253; site 2, VP1
residues 220 to 222, VP2 residues 164 to 172 and VP2
residue 270; site 3, VP3 residues 58 to 60, 70, 71 and VP1
residues 286 to 290; site 4, VP2 residue 72 and VP3
residues 76 to 79. There are differences in immunodominance in BALB/c mice of the different sites for
types 1 and 3 poliovirus (Minor et al., 1987) but
mutations have not yet been identified in all segments of
the complex sites for either type 1 or type 3 virus.
Following infection of tissue culture cells with poliovirus, both infectious and non-infectious virus particles
are present in the culture fluid. The non-infectious
0000-9397 O 1990 SGM
Methods
Preparation of [3SS]methionine-labelled virus. The Sabin strains of
poliovirus types 1 (P1/LSc), (P2/712ch) and 3 (P3/Leon 12a, b) and
antigenic mutants derived from them were grown in HEP2c cells and
purified on sucrose gradients as described previously (Minor et al.,
1980).
Single radial immunodiffusion antigen blocking tests. These were
carried out as described previously (Ferguson et al., 1984). Briefly,
dilutions of monoclonalantibodyor polyclonalsera were added to wells
in agarose gets containing low concentrations of hyperimmune antipoliovirus serum. After the wells had emptied, [3sS]methioninelabelled 155S D antigen or 80S C antigen were added to the wells.
Diffusion of radiolabelled antigen in the gel after 24 to 48 h was
detected by autoradiography.Test antibody is consideredto react with
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 08:28:12
1272
M. Ferguson and P. D. Minor
antigen if its diffusion into the gel is inhibited compared with control
antigen in a well to which phosphate-bufferedsaline had been added.
Antigen blocking titres are expressed as the dilution of antibody that
reduces the zone size by 75% in comparison with zones produced with
control antigen.
Preparation of monoclonal antibodies. Monoclonal antibodies were
prepared by fusion of splenocytes from immunized BALB/c mice and
myeloma cells as described previously (Ferguson et aL, 1984).
Hybridoma supernatants were screened using antigen blocking tests
and ascites were prepared in syngeneic mice.
Peptide and polyclonal antipeptide sera. Peptides containing amino
acid sequences located at antigenic sites 1, 3a and 3b of type 3 Sabin
vaccinevirus were synthesizedby the Fmoc polyamidemethod of solidphase peptide synthesis (Cambridge Research Biochemicals; Brown et
al., 1983). The sequences of the peptides were VP1 residues 89 to 104
(designated S10a), VP1 residues 282 to 292 (designated $21) and VP3
residues 53 to 68 (designated $20). An additional peptide, designated
W 1, which contains the amino acid sequence from VP2 residues 160 to
174 was kindly synthesized by Wellcome Biotechnology. Rabbit sera
were prepared against peptides coupled to carrier protein as described
previously (Ferguson et al., 1985).
Results
Antigenic sites recognized by C-specific monoclonal
antibodies
Monoclonal antibodies were identified which reacted
with D antigen, C antigen or both D and C antigen in
antigen blocking tests indicating that the two particles
contain both shared and distinct epitopes (Ferguson et
al., 1984). The sites of reactivity of monoclonal antibodies with neutralizing activity were identified by
selection of neutralization-resistant mutants or by
neutralization of viruses with amino acid substitutions
identified by primer extension sequencing. However
only one out of 29 C-specific antibodies blocked the
production of cytopathic effect. This antibody appeared
to be directed against site 1, VP1 89 to 100 (Minor et al.,
1986).
The remaining 28 C-specific antibodies were tested in
antigen blocking tests with the 80S peak of representative mutants with amino acid substitutions in each of the
antigenic sites. Eight of the 28 antibodies tested failed to
react with at least one virus with an amino acid
substitution in an antigenic site. The results obtained
with these antibodies are shown in Table 1. Antibodies
516 and 518 did not react with virus strains 4008 and
4025 which had amino acid substitutions at residues 166
and 164 on VP2 respectively. They did however react
with virus 4023 which has a mutation at residue 172 of
VP2. This suggests that these antibodies are directed
against antigenic site 2 and that, as with the neutralizing
D- or D + C-reactive antibodies, not all mutations within
an antigenic site will affect the reactivity of individual
antibodies. Similarly, antibodies 131 and 468 failed to
react with virus 4022 which has a mutation in site 3b at
VP3 residue 59, indicating that they are directed against
site 3b. Antibodies 182 and 475 did not react with virus
strains 4030 or 4031 which have mutations in site 4 at
VP3 residues 77 and 79 respectively. However antibodies
209 and 212 reacted with virus 4030 but not 4031. The
four main antigenic sites involved in neutralization of
infectious virus are therefore also present on the surface
of non-infectious C antigen but in a different
configuration.
All C antigen-specific antibodies were tested for
reactivity with separated poliovirus capsid proteins in
immunoblot experiments. Four antibodies were found to
react with VP1 and one with VP3 (Ferguson et al., 1984),
but none of these antibodies were found to be directed
against antigenic sites in the studies described above.
The lack of reactivity in immunoblots of antibodies
directed against the antigenic sites on non-infectious C
antigen particles indicates that the surface features of
empty particles, like those of infectious virus, have a
defined structure. Further information on the site of
reactivity of the unassigned C-specific antibodies m a y be
obtained when chimeric polioviruses, in which sequences of eight to 10 amino acids from type 3 virus are
inserted into the homologous region on type 1 virus,
become available.
Reactivity of monoclonal antibodies with the antigenic
sites of type 3 poliovirus D and C antigen
A total of 83 monoclonal antibodies against type 3
poliovirus have been characterized with respect to
reactivity with D and C antigens. The antigenic sites
which have been identified so far for type 3 virus are
given in Table 2 along with the number of antibodies
which have been identified as directed against each site
on D and/or C antigen. The antigenic sites against which
three out of 45 D-specific antibodies are directed have
yet to be identified whereas 20 of the 29 C-specific antibodies have still to be assigned to antigenic sites. Five out
of nine antibodies which react with both D and C
antigens are directed against site 1 at VP1 residues 89 to
100 indicating that at least part of this loop is in the same
configuration on both particles. In contrast, no D- and Creactive antibodies directed against sites 3 or 4 have been
identified. However only one antibody against site 4 on
D antigen has been produced suggesting that this site on
infectious virus is not well recognized by the mouse
immune system.
An antibody designated 200 reacted with both D and
C antigens in antigen blocking tests. This antibody
neutralized virus in standard microtitre assays but no
mutant viruses could be isolated as the titre was reduced
by only 90 % in plaque reduction assays. This antibody is
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 08:28:12
Poliovirus antigenic site conformation
1273
Table 1. Reactivity in antigen blocking tests of C antigen-specific monoclonal antibodies
with mutant virus strains
Virus strain with amino acid substitution
Monoclonal
antibody
Type 3 Sabin
vaccine virus
4023
VP2-172
516
518
131
468
182
475
209
212
4*
2
3
4
4
5
3
2
4
2
3
4
4
5
3
2
4008
VP2-166
< 1
< 1
4
4
4
5
2
2
4025
VP2-164
< 1
< 1
2
4
4
5
2
2
4022
VP3-59
4
2
<1
< 1
4
5
2
2
4030
VP3-79
4
2
4
4
< 1
< 1
3
2
4031
VP3-77
4
1
3
4
< 1
< 1
<1
< 1
* log10.
Table 2. Reactivity of poliovirus type 3 monoclonal
antibodies with infectious D antigen and non-infectious C
antigen
N u m b e r of antibodies reacting with
Antigenic site
D antigen
1 VPI 89-100, 166, 253
VP1 141-152
2 VP2 165-172
3 VP1 286--290"~
VP3 58-60 . J
4 VP3 77-79
Unknown
Total
C antigen
D + C antigens
19
0
4
1
0
2
5
2
1
19
1
3
45
2
4
20
29
0
0
1
9
Table 3. Reactivity of antipeptide sera with type 3 D and C
antigen in antigen blocking tests
Antigen blocking titre
Rabbit
Immunogen
237
238
239
240
266
267
268
269
$21"
$21
$21
$21
Wl t
Wl
Wl
Wl
Sabin type 3
D antigen
<
<
<
<
10
10
10
10
40
80
160
160
Sabin type 3
C antigen
160
320
160
80
160
160
320
320
* Sequence: VP1 residues 282-292, ( C ) G V D Y R N N L D P L C .
1" Sequence: VP2 residues 160-174, ( C ) F N K D N A V T S P K R E F C .
surface of the virus was in site 2 on VP2 at residue 165 (A.
Macadam, personal communication). In addition, although this antibody reacted with D antigen of site 2
mutants 4008 and 4025 which have amino acid substitutions in VP2 at residues 166 and 164 respectively (see
Table 1), it failed to react with C antigen of these viruses
in antigen blocking tests. This suggests that the change in
configuration caused by the amino acid substitutions was
greater in the empty C antigen particles than in the
infectious D antigen particles.
Reactivity of antipeptide sera with poliovirus particles
Antisera which were prepared against synthetic peptides
comprising amino acid sequences of antigenic site 1
(peptide S10a), site 2b (peptide W1), site 3a (peptide $21)
and site 3b (peptide $20) were tested in antigen blocking
tests with type 3 D and C antigens. Sera from animals
immunized with site 1 peptide S10a developed antibodies to both D and C antigens (Ferguson et al., 1985)
whereas no antiviral activity was obtained with site 3b
peptide $20. Site 2 peptide W 1 induced antibodies which
recognized both D and C antigens in antigen blocking
tests (Table 3) whereas antisera to site 3a peptide S21
reacted only with C antigen suggesting that the peptide
adopted the configuration found on C but not D antigen.
The data for site 3a confirm that this sequence is at the
surface of empty C antigen particles but in a configuration different to that found on D antigen.
Discussion
thought to be directed against site 2 on the basis of its
failure to react in antigen blocking tests with a strain,
DM 35, excreted by a primary vaccinee. The entire
capsid region of this strain has been sequenced and the
only amino acid substitution found which maps at the
Although the structure of infectious type 3 poliovirus is
well defined both at the atomic (Filman et al., 1989) and
antigenic (Minor et al., 1986) levels, little is known about
the structure and antigenic proper!ies of empty C antigen
particles. Monoclonal antibodies produced against type
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 08:28:12
1274
M. Ferguson and P. D. Minor
3 poliovirus antigens were initially used to characterize
the antigenic structure of infectious virus, D antigen and
non-infectious empty C antigen particles. Antibodies
were identified which reacted with D antigen, C antigen
or both D and C antigen indicating that the two particles
contain both shared and distinct epitopes (Ferguson et
al., 1984). In these studies we show that the amino acid
sequences which form the main antigenic sites on type 3
virus are also exposed on the surface of empty C antigen
particles but in different configurations. This conclusion
is supported by the reactions of antisera raised with
synthetic peptides.
The panel of type 3 monoclonal antibodies used in
these studies was the product of several fusions in which
different viruses and particles were used as immunogens
and for screening and do not therefore reflect the relative
dominance of the different sites. It is of interest that the
majority of D + C-reactive antibodies are directed
against site 1 suggesting that less conformational change
takes place within this exposed loop when RNA is lost.
However there are also a large number of D-specific antibodies directed against this site (Ferguson et al., 1984)
indicating that some regions are in different configurations. There does not appear to be a clear distinction
between amino acids recognized by D- or D + C-reactive
antibodies from studies involving the reactivity of
monoclonal antibodies with variant viruses containing
single amino acid substitutions within site 1 (P. D.
Minor, unpublished data).
Monoclonal antibodies against the complex antigenic
sites 2, 3 and 4 are D- or C-reactive with the exception of
antibody 200. Three polypeptide segments from type 1
virus contribute to antigenic site 2 although only two
have so far been identified on type 3 virus. The complex
nature of this site may therefore be responsible for the
unusual binding properties of antibody 200 in that
different mutations may result in different local conformational changes on D and C antigens.
The two segments of antigenic site 3 appear to be in
different configurations on D and C antigen particles on
the basis of reactivity with monoclonal antibodies and
with an antiserum against a peptide corresponding to
residues 286 to 290 on VP1. That the antipeptide serum
reacted only with empty C antigen particles demonstrates the problems of using peptides as a synthetic
vaccine for antigens with a defined structure.
The studies described here provide information on the
surface features of non-infectious type 3 C antigen
particles and on the structure of type 3 poliovirus. The
antibodies used in these studies may be useful tools for
investigating the structural transitions which occur
during adsorption to cells, uncoating, viral assembly and
during thermal inactivation.
References
BROWN, E., SHEPPARD, R. C. & WILLIAMS, B. J. (1983). Peptide
synthesis. Part 5. Solid-phase synthesis of (15-1eucine) little gastrin.
Journal of the Chemical Socie(y Perkins Transactions 1, 1161-1167.
FERGUSON,M., MINOR, P. D., MAGRATH,D. I., QuI, Y.-H., SPITZ, M.
& SCHILD,G. C. (1984). Neutralization epitopes on poliovirus type 3
particles: an analysis using monoclonal antibodies. Journal of
General Virology 65, 197-201.
FERGUSON, M., EVANS, D. M. A., MAGRATH,D. I., MINOR, P. D.,
ALMOND, J. W. & SCHILD, G. C. (1985). Induction by synthetic
peptides of broadly reactive, type-specific neutralizing antibody to
poliovirus type 3. Virology 143, 505-515.
FERNANDEZ-TOMAS,C. B. & BALTIMORE,D. (1973). Morphogenesis of
poliovirus. II. Demonstration of a new intermediate, the provirion.
Journal of Virology 12, 1122-1130.
FILMAN,n. J., SYED,R., CHOW, M., MACADAM,A. J., MINOR, P. D. &
HOGLE, J. M. (1989). Structural factors that control conformational
transitions and serotype specificity in type 3 poliovirus. EMBO
Journal 8, 1567-1579.
HOGLE, J. M., CHOW, M. & FILMAN, D. J. (1985). The three
dimensional structure of poliovirus at 2.9 A resolution. Science 229,
1358-1365.
MAYER, M. M., RAPP, H. J., ROIZMAN,B., KLEIN, S. W., COWAN,
K. M., LUKENS,D., SCHWERDT,C. E., SCHAFFER,F. L. & CHARNEY,
J. (1957). The purification of poliomyelitis virus as studied by
complement fixation. Journal of Immunology 78, 435-455.
MINOR, P. D., SCHILD, G. C., WOOD, J. M. & DANDAWATE,C. N.
(1980). The preparation of specific immune sera against type 3
poliovirus D-antigen and C-antigen and the demonstration of two Cantigenic components in vaccine strain populations. Journal of
General Virology" 51, 147-156.
MINOR, P. D., SCHILD, G. C., BOOTMAN, J., EVANS, D. M. A.,
FERGUSON, M., REEVE, P., SPITZ, M., STANWAY,G., CANN, A. J.,
HAUPTMANN,R., CLARKE,L. D., MOUNTFORD,R. C. & ALMOND,
J. W. (1983). Location and primary structure of a major antigenic site
for poliovirus neutralization. Nature, London 301, 674-679.
MINOR, P. D., EVANS, D. M. A., FERGUSON, M., SCHILD, G. C.,
WESTROP, G. & ALMOND, J. W. (1985). Principal and subsidiary
antigenic sites of VP1 involved in the neutralization of poliovirus
type 3. Journal of General Virology 66, 1159-1165.
MINOR, P. D., FERGUSON,M., EVANS, D. M. A., ALMOND,J. W. &
ICENOGLE, J. P. (1986). Antigenic structure of polioviruses of
serotypes 1, 2 and 3. Journal of General Virology 67, 1283-1291.
MINOR, P. D., FERGUSON, M., PHILLIPS, A., MAGRATH, D. I ,
HUOVILAINEN,A. & HOVI, T. (1987). Conservation in vivoof protease
cleavage sites in antigenic sites of poliovirus. Journal of General
Virology 68, 1857-1865.
PAGE, G. S., MOSSER,A. G., HOGLE, J. M., FILMAN,n. J., RUECKERT,
R. R. & CHOW, M. (1988). Three-dimensional structure of poliovirus
serotype 1 neutralizing determinants. Journal of Virology 62,
1781 1794.
SCHILD, G. C., WOOD, J. M., MINOR, P. D., DANDAWATE,C. N. &
MAGRATH, D. I. (1980). Immunoassay of poliovirus antigens by
single-radial-diffusion: development and characteristics of sensitive
autoradiographic zone size enhancement (ZE) technique. Journal of
General Virology 51, 157-170.
WIEGERS, K. J., UHLIG, H. & DERNICK, R. (1989). N-AglB of
poliovirus type 1 : a discontinuous epitope formed by two loops of
VP1 comprising residues 96-104 and 141-152. Virology 170,
583-586.
(Received 28 November 1989; Accepted 13 February 1990)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 08:28:12