227
J. gen. Virol. (1975), 26, 227-238
Printed in Great Britain
A Comparative Chemical and Serological Study of the Full and
Empty Particles of Foot-and Mouth Disease Virus
By D. J. R O W L A N D S , D. V. S A N G A R AND F. B R O W N
Animal Virus Research Institute, Pirbright, Woking, Surrey, U.K.
(Accepted 29 October I974)
SUMMARY
The chemical and serological properties of the full, naturally occurring empty
and artificially produced empty particles of foot-and-mouth disease virus, serotype
A (subtype IO, strain 6I) have been studied. The full I46 S particles comprised the
virus RNA, three polypeptides (VPI to VP3) mol. wt. about 3o x io z, one polypeptide (VP4) tool. wt. about I3"5 × IO3, and a small amount of a polypeptide
(VPo) mol. wt. about 43 x io 3. The naturally occurring 75 S empty particles contained no RNA and much less VPI and VP4 than were found in the full particles.
However they contained a much greater proportion of VPo than the full particles.
Dialysis of purified full particles against tris-EDTA, pH 7"6, produced artificial
75 S empty particles which contained only a small amount of RNA and no VP4;
otherwise the polypeptide composition was similar to that of the full particles.
Immunological and serological tests showed that the full particles were antigenically
similar to the naturally occurring empty particles but distinct from the artificial
empty particles. The latter particles, however, had serological properties similar to
those of the I2 S protein subunit of the virus. Both the full and naturally occurring
empty particles attached efficiently to susceptible ceils, whereas the artificial empty
particles attached only to a limited extent. The results are related to the function
of the individual polypeptides of the virus particle and compared with published
work on other picornaviruses.
INTRODUCTION
In contrast to full (D) poliovirus particles, the empty (i.e. RNA-free or (2) particles
occurring in harvests and the artificial empty particles produced by heating the full particles
at 56 °C or by irradiating them with u.v. light elicit the production of only low levels of
neutralizing antibody (Hummeler & Hamparian, I958; Hummeler & Tumilowicz, 196o;
Hinuma, Katagiri & Aikawa, I97O; Ghendon & Yacobson, I97I). This difference in the
antigenic activity of the different particles correlated with the serological differences demonstrated by complement fixation, immuno-diffusion and immune electron microscopy tests
(Le Bouvier, Schwerdt & Schaffer, I957; Mayer et al. 1957; Hummeler, Anderson & Brown,
I962). Furthermore, the D particles attach efficiently to susceptible cells whereas both the
naturally occurring and artificial empty particles do not attach (Katagiri, Hinuma & Ishida,
I968). The naturally occurring empty particles contain all the polypeptides present in the
full particles but two of the polypeptides (VP2 and VP4) are present as a single uncleared
molecule (VPo). The artificial empty particles differ from the full particles in not containing
any RNA or polypeptide VP 4. The polypeptide conferring D antigenicity on the full particle
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 21:43:44
228
D. ft. R O W L A N D S ,
D. V. S A N G A R
A N D F. B R O W N
is likely to be VP4 (Breindl, 1971). This polypeptide is also likely to carry the attachment
site in the closely related Coxsackie B3 virus (Crowell & PhiUipson, 197I).
In contrast, the situation with rhino- and foot-and-mouth disease viruses is somewhat
different. The naturally occurring empty particles of rhinovirus, type 2, possess D as well as
C antigenicity (Lonberg-Holm & Yin, 1973) and show limited attachment to susceptible
cells (Noble & Lonberg-Holm, I973). Similarly, as we show in this paper, the naturally
occurring empty particles of foot-and-mouth disease virus also possess D antigenicity and
attach as efficiently as the full particles to susceptible cells. Furthermore, our previous work
with foot-and-mouth disease virus has provided evidence that only polypeptide VPI is
concerned in the formation of neutralizing antibody and attachment to susceptible cells
(Wild & Brown, I967; Wild, Burroughs & Brown, I969). Trypsin destroys both these
properties but, as judged by polyacrylamide gel electrophoresis, only VPI is cleaved. The
polypeptide VP4 does not appear to be involved in either the formation of neutralizing
antibody or the attachment to cells. Furthermore, the antigenic site of VP4 appears to be
located internally (Talbot et al. 1973).
In this paper the polypeptide composition and antigenic properties of the full, naturally
occurring empty and artificially produced empty particles of foot-and-mouth disease virus
have been studied. The implications of the results are discussed in relation to the proposed
model for the virus.
METHODS
Virus growth and assay. Foot-and-mouth disease virus (serological type A, subtype lO,
strain 6 0 was grown in BHK zI cell monolayers in Eagle's medium. The same medium
was used for growing [3H]-uridine labelled virus. Methionine-free Eagle's medium was used
for labelling the virus with [35S]-methionine and Earle's medium when labelling with a
mixture of amino acids was required. The virus was titrated by intraperitoneal inoculation
of 7-day-old mice with Io-fold dilutions of the preparations.
Purification of full and empty particles. Virus particles were prepared by the method
described by Brown & Cartwright (1963). Purification of natural and artificial empty particles is described in the Results section.
Iodination of virus. Purified virus in 25/zl o-04 M-phosphate, pH 7"6, was mixed with
IOO/zc [125I]-Na in 5/A solution and Ioo/zg chloramine-T, in IO/zl, added. After I min, the
mixture was diluted with I ml o'04 M-phosphate and the virus separated on a Sephadex
G-Ioo column in 0"04 M-phosphate. The virus was mixed with SDS (to r ~) and re-cycled
in a sucrose gradient to ensure freedom from low tool. wt. radioactive contaminants.
Preparation ofantisera. Hyperimmune guinea pig antiserum was obtained by intradermal
inoculation of virus passaged in guinea pigs into the hind foot pads of groups of animals,
followed by an intramuscular inoculation of a similar preparation of virus iz weeks later.
The animals were exsanguinated IO days after the second inoculation. Sera were also
obtained from guinea pigs which had received subcutaneous inoculations of full and empty
particles, inactivated with 0"05 ~ acetylethyleneimine (AEI) for 6 h at 37 °C or o'o5
formaldehyde for 72 h at 2o °C followed by o'05 ~ acetylethyleneimine for I h at 37 °C.
These preparations were mixed with aluminium hydroxide gel prior to inoculation.
Determination of the activity of neutralizing antibody. Tenfold dilutions of virus were
mixed with equal vol. of guinea pig sera or the separated IgG fraction. After I h at
20 °C, o'03 ml vol. of the mixtures were inoculated into groups of 7-day-old mice.
The difference between the 5o ~ end point of an experimental mixture and that of the virus
alone was taken as the neutralizing activity of o'o15 ml of the serum or serum fraction.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 21:43:44
Full and empty particles in F M D V
229
Serum antibody blocking activity. Fivefold dilutions of antigen in 0"04 M-phosphate,
pH 7"6, were mixed with equal vol. of I/I25 hyperimmune antiserum for I h at zo °C. The
mixtures were then diluted with an equal vol. of water (to give a final phosphate concentration
of o.oI M) and passed through a column of DEAE-cellulose equilibrated with o-oI Mphosphate of pH 7"6. The unreacted IgG which passed unadsorbed through the column
was tested for virus neutralizing activity. Control mixtures of phosphate buffer alone and
I/I25 hyperimmune serum were examined similarly for neutralizing activity.
Immunodiffusion tests. The double diffusion method described by Ouchterlony (I949) was
used with minor modifications. Agarose (o'85 %) in tris-HC1, pH 7"6, and containing o'oI %
sodium azide, was used and the reactions were observed for 7 days.
Measurement of attachment to cells. Radioactive preparations of the full and empty
particles were mixed with primary pig kidney cells suspended in phosphate-buffered saline.
After I5 min at 37 °C, the suspensions were centrifuged at 6oog for 5 min and the radioactivity in the supernatant fluid and cells was measured.
Assay of radioactivity. Samples for scintillation counting were prepared as described in
detail in Talbot et aL 0973)- For autoradiography the polyacrylamide gels were sliced
longitudinally and the central portion dried under vacuum on to a wettable cellophane
sheet (Russell & Skehel, I972). The dried strips were placed in contact with X-ray film for
an appropriate period before developing the film.
Polyacrylamide gel electrophoresis. The virus polypeptides were analysed in SDS gels as
described by Burroughs et al. (I971).
RESULTS
Purification of full and empty particles
Full particles were prepared by the method described by Brown & Cartwright 0963).
The pellet obtained by ammonium sulphate precipitation and centrifuging at 8o ooo g for
1.5 h was resuspended in o'o4 M-phosphate, pH 7"6, and dispersed with I ~ SDS prior to
centrifuging in a I5 to 45 ~ sucrose gradient.
Natural empty particles are destroyed by SDS and to a lesser extent by sodium deoxycholate. Since they are stable in Nonidet P4o, however, the last step in the purification
scheme for full particles was modified by mixing the resuspended virus pellet with o.oI
pancreatic ribonuclease at 37 °C for r5 min to destroy ribosomes and then mixed with I ~o
Nonidet P4o before centrifuging in a sucrose gradient. Using this purification method with
virus grown in the presence of pH]-uridine and [14C]-amino acids, the distribution of radioactivity and infectivity showed that the empty particles contained no detectable RNA
(Fig. I a). Fractions 14 and 15 contained less than I ~oo of the infectivity of the full particles
(Fractions 7 to 9).
The method of K. Strohmaier (personal communication) was used to prepare artificial
empty particles. Purified full particles were dialysed against o.oi ~ EDTA, o-o2 M-tris,
pH 7"6, for 2"5 h and then separated from any residual full particles by centrifuging in a
sucrose gradient (Fig. I b). In most experiments yields of 75 S particles were about z5 ~, based
on radioactive protein recoveries. The empty particles contained 4 ~ of the uridine and
o.ooi ~ of the infectivity of the full particles from which they were prepared.
The morphology of typical preparations of the three particles is shown in Fig. 2. Our
preparations of artificial empty particles always appeared to be aggregated, as shown in the
lower panel, but occasionally separate particles were observed (upper panel). Since the
particles were isolated from the 75S region of the gradient, the aggregation must have
occurred during the preparation of the samples for microscopy.
I5
v I R ~6
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 21:43:44
230
D.J. ROWLANDS, D. V. SANGAR AND F. BROWN
I
I
I
I
(a)
20
20
o
"
[3H].uridine
10
10
j
7
_ [14C]-amino
/ ~
acids
t",
)<
._=
X
.E
0--0
f / w / acids f
6
6
4
4
~"
[3H]'uridine
2
2
"0""0"0"
"01
5
,,
"'0
'
10
15
20
Bottom
Top
Fraction number
Fig. I. Sucrose gradient sedimentation of (a) virus pellets after treatment with I ~ Nonidet P4o,
showing the separation of full and empty particles and (b) full particles after dialysis against
EDTA. 0-------0, [14C]-aminoacids; O . . . . O, laH]-uridine.
Polypeptide composition of full and empty particles
SDS-polyacrylamide gel electrophoresis in lo ~ gels at p H 7"6 of the proteins of [14C]amino acid labelled full particles gave two main peaks of radioactivity and a very small
amount of a third (Fig. 3 a). Using bovine serum albumin and myoglobin as internal markers,
the major peaks had mol. wt. of 3° and I3"5 × Io ~ and the minor peak had a mol. wt. of
43 x io 3. The peak with mol. wt. 43 x io 3 consisted of VPo and that with tool. wt. I3"5 × Io ~
consisted of VP4. Although we were unable to resolve the major peak in IO ~ gels, it was
considered to contain the three polypeptides VPI, VP2 and VP3 found in foot-and-mouth
disease virus of serotypes o and Asia I. Similar difficulties in resolving this peak have also
been encountered with several other strains of the virus (Talbot et al. I973)- By increasing
the gel concentration to I2. 5 ~ , the band was resolved into two peaks which were readily
detected by Coomassie blue staining or by autoradiography (Fig. 3 b). The intensity of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 21:43:44
Full and empty particles in F M D V
Full
particles
Natural
empty particles
231
Artificial
empty particles
Fig. 2. Electron micrographs of full, natural empty and artificial empty particles stained with t %
phosphotungstic acid.
staining of the more slowly migrating band (VPt) was about one-half that of the faster
moving band (VP2 and VP3).
Following trypsin treatment of the full particles, VPo, VPI and VP4 were unchanged
but the band containing VP2 and VP3 was reduced in intensity by about 5o % and additional
peaks were obtained at positions corresponding to mol. wt. of I9 and Io × IO3 (Fig. 3 c).
Polypeptides of these sizes are also found when other serotypes of the virus are treated
with trypsin. Since the enzyme removed only about 50 % of the material in the VP2 and
VP3 peak, it is clear that there are at least two polypeptides migrating to this position.
Polypeptides VPI, VP2 and VP3 of serotypes o and Asia I can be clearly separated, even
in IO % gels, but as these serotypes produce unstable empty particles they were not used in
this study. Trypsin treatment of types o and Asia I converts VPI into polypeptides of tool.
wt. I9 and Io x ~o3. With the virus used in the present work, VP2 or VP3, but not VPI, is
clearly the trypsin-sensitive polypeptide.
The natural and artificial empty particles gave the polypeptide profiles shown in Fig. 4.
Natural empty particles gave two major peaks (VPo and VP2 +VP3). Minor peaks corresponding to VP~ and VP 4 were seen on the autoradiographs but were of such a low intensity
that they did not appear in the photograph. These results are in agreement with the observations made by Vande Woude, Swaney & Bachrach 0972). In contrast, artificial empty
particles gave a polypeptide profile similar to that of full particles except for the absence
of VP 4.
The constituent potypeptides of the major peak of full and artificial empty particles were
compared by using iodinated virus. More than 95 % of the iodine label of full particles was
associated with the major peak containing VPI, VP2 and VP3 (Fig. 5a). After trypsin
treatment, most of the label was then present in the polypeptide with mol. wt. 19 x lO3
(Fig. 5 b). Since VP2 or VP3 is the trypsin-sensitive polypeptide, this means that only one of
15 "2
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 21:43:44
D. J. R O W L A N D S , D. V. SANGAR AND F. BROWN
232
(a)
(b)
Fulls
~o~
Fulls
Fulls + trypsiT~
Natural
Artificial
12.5~,
12.5~o
empties
empties
(c)
(a)
(b)
VP0
VPO
VPI
VP2
VP3
PO
, ~ VP1
VP0
VPI
P1
VP2
-VP3
P2
P3
VP2 +
VP3
VP2
+ VP3
P4
~sln
luct 1
P4
VP4
Trypsin
product II
Fig. 3
Fig. 4
Fig. 3. Polyacrylamide gel electrophoresis of [35S]-methionine labelled full particle polypeptides
in SDS gels of different concentrations. The dried gels were autoradiographed on X-ray film:
(a) full particles in a 1o ~ gel; (b) full particles in a Iz'5 ~ gel; (c) trypsin-treated full particles
in a 12"5 ~ gel.
Fig. 4. Polyacrylamide gel electrophoresis of [35S]-methionine labelled natural and artificial empty
particle polypeptides in 12.5 ~o gels, The dried gels were autoradiographed on X-ray film.
(a) Natural empty and (b) artificial empty particle polypeptides.
these polypeptides is iodinated in the intact virus. Conversion o f iodinated full particles
into empty particles did not lead to any loss in specific activity, showing that the trypsinsensitive polypeptide is retained in the empty particles. Polyacrylamide gel electrophoresis
showed that the label was still associated with the polypeptide which is sensitive to trypsin
(i.e. VP2 or VP3 - see Fig. 5 ¢).
Serological properties of the three particles
Immunodiffusion
The full and natural empty particles gave a line o f identity when allowed to diffuse t h r o u g h
o'85 ~o agarose gels towards hyperimmune guinea pig serum. Each precipitin b a n d crossed
the b a n d formed by the I2 S subunit obtained by heating the full or natural empty particles
at 56 °C (Fig. 6a). This result differs in one respect f r o m that obtained by Graves, C o w a n &
T r a u t m a n 0 9 6 8 ) ; these authors f o u n d that the natural empty particles p r o d u c e d a precipitin
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 21:43:44
Full and empty particles in F M D V
I
40
I
I
I
I
233
I
I
I
50
60
70
(a)
30
10
f
........
8
~
.......
=m . . . . . . .
J
(b)
7 61
×
E
4
............................ L . .....J
15
(c)
I0
1
10
20
30
40
Fraction number
Fig. 5. Polyacrylamide gel electrophoresis of the polypeptides of untreated particles, trypsintreated particles and artificial empty particles all of which were iodinated with I~H.The gels were
sliced into I turn sections and counted in a Packard scintillation counter: (a) virus; (b) trypsin virus;
(c) artificial empty particles.
line which fused with those produced by both the full and i2 S particles, although the region
of fusion with the I2 S particle was 'vague and did not give the impression of strict identity'
(Cowan, I973).
in contrast to the results with the natural empty particles, the precipitin line obtained
with the artificial empty particles showed no relationship with the full particles but fused
with the band produced by the I2 S particles (Fig. 6/?).
Serum blocking
In view of the identity of the full and naturally occurring empty particles in immunodiffusion tests, serum blocking tests were made to determine whether the empty particles
would react with the neutralizing antibody present in hyperimmune serum. Fivefold dilutions
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 21:43:44
234
D.J. ROWLANDS,
D. V. S A N G A R
A N D F. B R O W N
Fig. 6. Immunodiffusion of full, natural empty, artificial empty and I2 S protein subunit preparations with hyperimmune guinea pig serum. A, full particles; B, natural empty particles; C, natural
empty particles heated at 56 °C; D, 12 S protein subunit prepared by heating full particles at 56 °C;
E, artificial empty particles; F, antiserum. The faint line seen in (a) well A, bottom right, is due to
spontaneous breakdown of 146 S particles to I 2 S particles.
Table I. Antibody blocking activity oJfull, natural empty and artificial empty
particles of FMD V
Dilution of
antigen
I/I
I/5
1/25
I / 125
Log reduction of neutralizing activity in o'o15 ml of IgG
,
~
Fulls
Natural empties Artificialempties
2.2
2-7
2-8
2"6
I"3
Nil
2"2
0"8
Nil
0"4
O'2
Nil
The mixtures of particles and 7 ~ hyperimmune serum were diluted twofold and filtered through DEAEcellulose in o.o1 M-phosphate, pH 7"6. The residual neutralizing activiW was measured by titration with
tenfold dilutions of virus.
of the preparations were mixed with a constant dilution of antiserum and the unreacted
IgG separated by filtration through DEAE-cellulose. The virus-neutralizing activity of the
filtrates was then determined. The natural empty particles had the same serum blocking
activity as the full particles; however, artificial empty particles absorbed much lower levels
of neutralizing antibody (Table 0.
Immunogenic activity
Since the natural empty particles and the full particles absorbed the same amount of
neutralizing antibody, their ability to elicit the production of neutralizing antibody was
compared. Only low levels of neutralizing antibody were produced in guinea pigs inoculated
with natural empty particles treated with o'o5 ~ acetylethyleneimine to destroy any residual
infective particles (Table 2). Cowan 0973) has also shown that natural 75 S particles similarly
inactivated produce much lower levels of neutralizing activity than an equal amount of the
full particles.
This finding, which was unexpected in view of the results of the serum blocking test, was
explained by the fact that inactivation with AEI resulted in considerable breakdown of the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 21:43:44
Full and empty particles in F M D V
235
Table 2. Immunogenic activity of full, natural empty and art~cial empty particles
of FMD V in guinea pigs
Preparation
Method of inactivation
Fulls
Neutralizing
activity* of serum
(log ID50/o'oi5 ml)
AEI alone
Formalin plus AEI
AEI alone
Formalin plus AEI
AEI a/one
Formalin plus AEI
Natural empties
Artificial empties
3-8
3"5
o'9
3"I
0"8
o'7
* Tenfold dilutions of virus were mixed with an equal volume of serum and 0'03 ml vol. inoculated intraperitoneally into groups of 7-day-old mice. The values in the table give the difference between the 5o ~ end
points of the virus alone and virus plus serum.
75S
(a)
75S
1
(b)
20
4
15 7
X
×
.= 3
2
5
10
15
20
5
10
15
Bottom
20
Top
Fraction number
Fig. 7. Sucrose gradient of sedimentation natural empty particles after inactivation with acetylethyleneimine (a) and after fixing with formaldehyde prior to inactivation with acetylethyleneimine
(b).
natural empty particles into slowly sedimenting material (Fig. 7a). I f the particles were firs
fixed with 0"05 ~ formaldehyde at 20 °C for 72 h before inactivation of residual infectivity
with o'05 ~ AEI, they then sedimented at 75 S (Fig. 7 b) and in most experiments produced
as much neutralizing antibody as the full particles (Table 2). However, in a few experiments
using similar conditions, only low levels of neutralizing antibody were obtained. At present
we have no explanation for the variable nature of the response to the natural empty particles.
The artificial empty particles always produced low levels of neutralizing antibody, irrespective of whether they were treated with acetylethyleneimine or formaldehyde (Table 2).
Attachment of the particles to susceptible cells
The ability of full particles to react with neutralizing antibody or to produce this antibody
when inoculated into guinea pigs appears to be related to their ability to attach to susceptible
cells, since treatment with trypsin destroys all three properties while only affecting one of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 21:43:44
236
D . J . R O W L A N D S , D. V. S A N G A R A N D F. B R O W N
Table 3. Attachment of full, natural empty and artificial empty particles
to pig kidney cells
Preparation
Fulls
Natural empties
Artificial empties
Radioactivity
attached to cells
(ct/min)
Radioactivity
remaining in
supernatant fluid
(ct/min)
Attachment
(~)
I4593
I452I
683
2319
2301
I0074
86
86
6
the capsid polypeptides (Wild et al. I969). Both full and natural empty particles attached
efficiently to pig kidney cells. However, artificial empty particles attached only to a limited
extent (Table 3) despite the presence of the trypsin-sensitive polypeptide.
DISCUSSION
All the picornaviruses so far examined contain three polypeptides with tool. wt. of approx.
3o × Io 3 and one polypeptide with a mol. wt. of approx. IO × IO3 (for list of references see
Talbot & Brown, I972). From the relative proportions of the different polypeptides, models
have been suggested for the virus particle (e.g. Rueckert, Dunker & Stoltzfus, 1969; Johnston
& Martin, I97I; Talbot & Brown, I972; Philipson, Beatrice & Crowell, I973). In the case
of foot-and-mouth disease virus, the capsid has been depicted as consisting of 20 units
comprising 3 molecules each of VPI, VP2 and VP3, together with 3° molecules of VP4;
this latter protein may be located internally. The I2S subunit, which can be derived from
the virus by reducing the pH below 7, comprises VPI, VP2 and VP3 in the same proportions
as they are present in the virus, but does not contain any VP4 (Burroughs et al. I97I).
Trypsin treatment cleaves one of the 3o × IO3 tool. wt. polypeptides into units with mol.
wt. of I9 and IO × Io ~ but does not affect the morphology of the virus particles. However,
the treated particles no longer attach to susceptible cells and produce only low levels of
neutralizing antibody in cattle or guinea pigs. These results have led to the conclusion that
the immunizing antigen and cell attachment site are associated with the trypsin-sensitive
polypeptide.
Natural empty particles, which contain the trypsin-sensitive polypeptide, also produce
neutralizing antibody provided they are fixed with formaldehyde prior to inoculation.
However, artificial empty particles and the I2 S protein subunit, both of which contain the
trypsin-sensitive polypeptide, fail to produce neutralizing antibody. Since neither of these
particles contain VP4, these observations lead to the conclusion that the immunizing activity
is expressed only in a structure of which VP4 is also a member, either as an individual polypeptide (in full particles) or as part of the larger polypeptide VPo (in natural empty particles).
The situation with poliovirus is different in that neither the naturally occurring nor the
artificially produced empty particles produce neutralizing antibody. Breindl (I97I) showed
by immunodiffusion tests that the neutralizing antibodies in poliovirus antisera attached
to VP4 of the virus particles, suggesting that the immunizing antigen of the virus is located
on this polypeptide. The absence of VP4 from artificial empty particles would thus provide
a ready explanation for their lack of immunizing activity. However, the lack of immunizing
activity of natural empty particles is more difficult to account for but it is possible that the
configuration of VP4 when it forms part of the uncleaved protein VPo prevents it from
producing neutralizing antibody. If the cell attachment site of poliovirus is located on VP4,
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 21:43:44
Full and empty particles in FMD V
237
as it appears to be in the case of the closely related enterovirus Coxsackie B3 (Crowell &
Philipson, I97I), this would suggest that neutralizing antibody reacts with the cell attachment site on the virus particle.
This fundamental difference between poliovirus and foot-and-mouth disease virus
demonstrates the dangers inherent in postulating a single unifying model for the picornaviruses. These dangers are emphasized by the work of Halperen, Eggers & Tamm 0964),
which showed that the full and natural empty particles of Echo ~2 virus were antigenically
similar and attached with equal efficiency to susceptible cells. Noble & Lonberg-Holm
(I973) have also shown that only 20 % of the population of natural empty particles in human
rhinovirus type 2 attached to susceptible cells, although there was no difference in the polypeptide compositions of the particles which attached and those which failed to attach to
the cells. Clearly, more information is required before we can attempt to explain the apparent
fundamental difference in the location of the antigenic and cell attachment sites of various
members of the pieornavirus group.
REFERENCES
LE BOUVIER, G. L., SCHWERDT, C. E. & SCHAFFER, E. L. (1957). Specific precipitates in agar with purified poliovirus. Virology 4, 590-593.
BREINDL, M. (1970. VP4, the D-reactive part o f poliovirus. Virology 46, 962--964.
BROWN, F. & CARTWRIGHT, B. (1963). Purification o f radioactive f o o t - a n d - m o u t h disease virus. Nature,
London x99, I i 6 8 - I 17o.
BURROUGHS, J.N., ROWLANDS, D. J., SANGAR, D. V., TALBOT, P. & BROWN, F. (197I). F u r t h e r evidence for
multiple proteins in the f o o t - a n d - m o u t h disease virus particle. Journal of General Virology x3, 73-84.
COWAN, K. M. (1973)- A n t i b o d y response to viral antigens. Advances in Immunology x7, 195-253.
CROWELL, R. L. & PHILIPSON, L. (197I). Specific alterations o f Coxsaekie virus B3 eluted f r o m H e L a cells.
Journal of Virology 8, 5o9-515.
GHENDON, Y. Z. & YACOBSON, E.A. (I97I). Antigenic specificity o f poliovirus-related particles. Journal o f
Virology 8, 589-59o.
GRAVES, J. H., COWAN, K. M. & TRAUTMAN, R. (I968). I m m u n o c h e m i c a l studies o f f o o t - a n d - m o u t h disease.
2. Characterization o f R N A - f r e e virus-like particles. Journal of Immunology 34, 269-274.
HALPEREN, S., EGGERS, H. J. & TAMM, I. (I964). C o m p l e t e a n d coreless h e m a g g l u t i n a t i n g particles p r o d u c e d in
Echo IZ virus-infected cells. Virology 23, 81-89.
rlINUMA, Y., KATAG1RI, S. & AIKAWA, S. (I970). I m m u n e response to H particles o f poliovirus. Virology 4o,
773-776.
HUMMELER, K., ANDERSON, T.F. & BROWN, R.A. (1962). Identification o f poliovirus particles o f different
antigenicity by specific agglutination as seen in the electron microscope. Virology x6, 84-9o.
/-ILTMMELER, K. & HAMPARIAN, V. V. (1958). Studies on the c o m p l e m e n t fixing antigens o f poliomyelitis.
I. D e m o n s t r a t i o n o f type a n d g r o u p specific antigens in native a n d heated viral preparations. Journal
of Immunology 8I, 499-505HUMMELER, K. & TUMILOWICZ, J. J. (1960). Studies on the c o m p l e m e n t fixing antigens o f poliomyelitis.
2. Preparation o f type-specific a n t i - N a n d a n t i - H indicator sera. Journal of Immunology 84, 630-634.
JOHNSTON, M. D. & MARTIN, S. J. (1971). Capsid a n d procapsid proteins o f a bovine enterovirus. Journal of
General Virology xx, 71-79.
KATAGIR1, S., I-IINUMA,Y. & ISHIDA, N. (I968). Relation between the adsorption to cells a n d antigenic properties
in poliovirus particles. Virology 34, 797-799.
LONBERG-HOLM, I¢. & VIN, F. H. (1973). Antigenic determinants o f infective a n d inactivated h u m a n rhinovirus,
type 2. Journal of Virology x9, 369-380.
MAYER, M. M., RAPP~ H. J., ROIZMAN, B., KLEIN, S. W., COWAN, K. M.~ LUKENS, D., SC/.IWERDT~ C. E.~ SCHAFFER,
F. L. & CHARNEY, J. (/957)- T h e purification o f poliomyelitis virus as studied by c o m p l e m e n t fixation.
Journal of Immunology 78, 435-455.
NOBLE, J. & LONBERG-HOLM~K. (I973)- Interactions o f c o m p o n e n t s o f h u m a n r h i n o v i r u s type z with H e L a
cells. Virology 51 , 270-278.
OUCHTERLONY, O. (I949). A n t i g e n a n t i b o d y reactions in gels. drkiv fSr Kemi, Mineralogi och Geologi
268, 1-9.
P/.IILIPSON, L., BEATRICE, S. T. & CROWELL, R. L. (I973). A structural m o d e l for picornaviruses as suggested
f r o m a n analysis o f urea-degraded virions a n d procapsids o f Coxsackie B3. Virology 54, 69-79.
RUECKERT, R. R., DUNKER, A. K. & STOLTZFUS, C. M. (1969). T h e structure o f mouse-Elberfeld virus; a model.
Proceedings of the National Academy of Sciences of the United States of America 62, 912-919.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 21:43:44
238
D.J.
ROWLANDS,
D. V. S A N G A R
AND
F. B R O W N
RUSSELL, W. C. & SKEHEL,J. J. (I972). The polypeptides of adenovirus-infected cells. Journal of General Virology
xS, 45-57.
TALBOT, e. & BROWN, F. (t972). A model for foot-and-mouth disease virus. Journal of General Virology xS,
I63-I7o.
TALBOT, P., ROWLANDS, D. J., BURROUGHS, J. N., SANGAR, D. V. & BROWN, F. (I973). Evidence for a g r o u p
proteins in foot-and-mouth disease virus particles. Journal of General Virology I9, 369-38o.
WILD, T. F. & BROWN, F. (I967). N a t u r e of the inactivating action o f trypsin on f o o t - a n d - m o u t h disease virus.
Journal of General Virology i, 247-25o.
WILD, T. F., BURROUGHS, J. N. & BROWN, t r- (1969). Surface structure of f o o t - a n d - m o u t h disease virus. Journal
of General Virology 4, 313-32o.
VAND~ WOUDE, C. F., SWANEY,J. B. & BACHRACH, rI. L. (I972). Chemical and physical properties of foot-andm o u t h disease virus: a comparison with Maus-Elberfeld virus. Biochemical and Biophysical Research
Communications 48, 1222-I 229 .
(Received 3 September I 9 7 4 )
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 21:43:44