J. gen. Virol. (1981), 55, 415427.
Printed in Great Britain
415
Key words:foot-and-mouth disease virus/neutralization~sensitization
Neutralization of Foot-and-Mouth Disease Virus. I. Sensitization of the
140S Virion by Antibody Also Reactive with the 12S Protein Subunit
By M. M. H A R D Y t
AND D. M. M O O R E *
U.S. Department of Agriculture, Science and Education Administration, Agricultural
Research, Plum Island Animal Disease Center, P.O. Box 848, Greenport, New York 11944,
U.S ,,I.
(Accepted 13 March 1981)
SUMMARY
The in vitro interaction of foot-and-mouth disease virus (FMDV) with an immune
serum resulted in a fraction of virus which failed to be neutralized. This inability of
antibody to neutralize the entire population of a virus preparation was studied with
emphasis on the antigenic specificity of the antibody-virus reaction. Antibody to
FMDV recognized multiple antigenic determinants. Immunoadsorbent fractionation
of the serum into 12S subunit cross-reactive and 140S virion-specific antibody
revealed that these multiple antigenic determinants are factors in determining the
neutralizing ability of the antibody. Antibody specific to the infective 140S virion
neutralized virus effectively, whereas antibody reactive with both the 140S virion and
12S non-infective component did so ineffectively. Neutralization was independent of
viral aggregation, strain, or type heterogeneity, dissociation of the immune complex,
heterogeneity of antibody class, or incubation time. The non-neutralized fraction of
virus was not due to insufficient antibody in the system, was demonstrated to be
complexed with antibody ('sensitized') and could be neutralized with anti-globulin
serum. The findings demonstrate the heterogeneity of antibody specificity to FMDV
in serum preparations and relate the importance of antibody specificity to the
neutralization of virus in vitro.
INTRODUCTION
The interaction of virus, antiviral antibody and host cell is one of the most extensively
studied, fundamental and yet controversial phenomenon of viral immunology. The in vitro
neutralization reaction ideally results in complete inhibition of viral infectivity. However,
non-neutralized fractions resulting from the incomplete neutralization of animal virus
populations by homologous antibody have frequently been reported for many viruses,
including foot-and-mouth disease virus (FMDV) (Dulbecco et al., 1956; Bradish et al., 1962;
Philipson, 1966; Fiala, 1969; Lewenton-Kriss & Mandel, 1972; Mandel, 1976; Rweyemamu
et al., 1977; Booth et al., 1978). Non-neutralized fractions have been identified for such a
large variety of virus-antiserum-cell systems (Daniels, 1975; Della-Porta & Westaway,
1978) that their existence should perhaps be considered a general characteristic.
The exact mechanism of neutralization, as well as the nature of the non-neutralized
fraction, has yet to be elucidated. Properties of each of the three components of the
neutralization system (virus, antibody, cell) have been implicated by various studies as the
responsible factor in incomplete neutralization. It is generally accepted, however, that the
t Present address: Biology Department, Brookhaven National Laboratory, Associated Universities, Inc.,
Upton, New York 11973,U.S.A.
Downloaded from www.microbiologyresearch.org by
0022-1317/81/0000-4514 $02.00 © 1981 SGM
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:38:47
416
M. M. H A R D Y AND D. M. M O O R E
non-neutralized fraction of most systems consists of infectious immune complexes
('sensitized' virus) (Majer, 1972; Della-Porta & Westaway, 1978). The existence of sensitized
virus as the cause of the non-neutralized fraction reflects a heterogeneity of virus-antibody
complexes. This heterogeneity has been related to the differences in affinity, the net charge of
the antibodies, or both (Svehag, 1968). The present study proposes the hypothesis that
antibody specificity is also an important factor. The investigation of a non-neutralized
fraction of virus in FMDV-antibody reaction mixtures demonstrated that the fraction was
associated only with antibody cross-reactive with virus and the 12S subunit and was absent
from antibody-virus mixtures specific for the 140S virion.
METHODS
Viruses. The large-plaque antigenic variant 'ab' (Cowan, 1969) of FMDV type A12 strain
119 (A12) was the principal virus used in this study. The virus was isolated by two successive
plaque isolations (Martinsen, 1970). A stock suspension of virus was prepared for
neutralization tests by two passages in secondary bovine kidney (2 ° BK) cells with Hanks'
balanced salt solution containing 10% lactalbumin hydrolysate (HLH) (I.C.N. Pharmaceutical, Cleveland, Ohio, U.S.A.) plus 10% (v/v) foetal bovine serum. Aliquots of virus
stock were diluted in 10% (v/v) normal bovine serum and filtered through a 200 nm
membrane filter before use. Plaque-isolated virus was propagated for purification by two
passages in baby hamster kidney (BHK2) cell cultures in HLH (Polatnick & Bachrach,
1964). In some tests, progeny of virus that escaped neutralization was harvested and tested
for neutralization resistance. The virus was obtained by aspirating several isolated plaque
zones through the overlay medium, suspended in HLH, and used in a neutralization test.
Type A24 of FMDV (A24), used in confirming assays, was produced from infected bovine
tongue tissue by four passages in BHK cell cultures with HLH as described for A12.
Inactivation and purification of virus. The A~2 BHK2 virus was inactivated with 3
mM-ethylenimine (BEI; Bahnemann, 1975) at 25 °C for 72 h, after which the reaction was
stopped by addition of a molar excess of sodium thiosulphate (Brown & Cartwright, 1963).
The BEI-inactivated virus was concentrated 100 times by precipitation with polyethylene
glycol 6000 (PEG) and resuspended in barbital-glycine buffer containing 0.15 M-NaC1
(Cowan et al., 1974). The PEG-concentrated virus was purified by ultracentrifugation into
CsC1 gradients (Wagner et al., 1970) and dialysed against NaC1 solution buffered with tris
(TBS; 0.05 M-tris-HC1 pH 7-5, 0.3 M-NaC1). Virus concentration was determined
spectrophotometrically (Bachrach et al., 1964).
Preparation of antiserum. Antiserum to A12 and A24 was produced by subcutaneous
inoculation of five guinea-pigs (Cowan, 1968) with 10 /~g inactivated purified 140S virus
antigen emulsified in Freund's-type incomplete oil adjuvant (Morgan & McKercher, 1977).
Sera (GP-a-A12, GP-a-A24 were collected and pooled at 31 days post-immunization. All sera
were heat-inactivated at 56 °C for 30 min.
Production of l 2 S protein antigen and preparation of l 2 S immunoadsorbent column. The
12S virus subunit-protein antigen was prepared by acid degradation of purified 140S FMDV
antigen (Cowan, 1968) and dialysed against TBS pH 7.5. The capsid polypeptides from the
purified 12S antigen and the 140S whole-virus antigen from which it was made were resolved
by 12.5 % polyacrylamide-SDS gel electrophoresis, stained by Coomassie Brilliant Blue and
examined to ensure that the polypeptides were not degraded and, therefore, antigenically
changed (Moore & Cowan, 1978).
The 12S protein antigen used in the immunoadsorbent column was dialysed against
borate-NaC1 buffer pH 8.5 (0.12 M-borate buffer, 0.2 M-NaCl). Fifteen mg of 12S antigen
were mixed with 1 g of washed and reswollen cyanogen bromide-activated Sepharose 4B
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:38:47
F M D V sensitized by anti-12S/140S antibody
417
(Pharmacia) equilibrated with borate-NaC1 buffer pH 8-5 (described above) and gently
tumbled at 4 °C overnight to allow coupling. The gel was washed twice with the borate-NaC1
buffer to remove unreacted 12S subunit antigen and the remaining amino-reactive gel groups
were blocked with 0.4 M-tris-HC1 buffer pH 8. The column was rinsed with 1 M-sodium
thiocyanate (NaSCN) to elute non-covalently bound 12S antigen and reequilibrated with
borate-NaC1.
Fractionation of antiserum by immunoadsorbent chromatography. Immune serum (10 ml)
in 0.2 M-NaC1 was applied to the 12S antigen-Sepharose 4B column (1 g gel, reswollen),
which was then rinsed with borate-NaC1 buffer. The column was eluted of bound
immunoglobulin, first with 1 M-NaSCN in 0.15 M-NaC1, 0.02 M-potassium phosphate
(PO4-NaC1) buffer pH 6.5, and second with 2 M-NaSCN in PO4-NaC1 buffer pH 6.5.
Normal guinea-pig serum (NGP; 5%, v/v) was added to the NaSCN fractions as carrier
protein. Fractions were compared by immunodiffusion analysis for 140S and 12S antigen
cross-reactivity, pooled into 12S antigen-adsorbed and 1 M-NaSCN and 2 M-NaSCN 12S
antigen cross-reactive serum eluates and dialysed under negative pressure against TBS pH
7-5. The recovered, dialysed, concentrated serum volumes for the 12S antigen-adsorbed,
1 M-NaSCN and 2 M-NaSCN eluates were 5 ml, 2.5 ml and 2.2 ml respectively.
Solution fractionation of 140S-12S component-reactive serum. The equivalence point of
the 12S-anti-12S subunit system was determined by mixing a twofold 12S antigen dilution
series (1600 to 50/~g range) with equal volumes of serum and comparing the dilutions by
immunodiffusion analysis (Sutmoller & Cowan, 1974). Antibody reactive with the 12S
subunit was subsequently batch precipitated with equal volumes (1 ml) of purified 12S
antigen at equivalence (400/tg/ml) in TBS and serum. The reaction was carried out at 37 °C
for 1 h and at 4 ° C for 18 h. The immune precipitate was centrifuged at 12 000 g for 20 min
at 4 °C and the supernatant fluid aspirated. The 12S subunit-reactive antibody was recovered
by dissolving the 12S-anti-12S subunit immune complex with 4 M-NaSCN and then dialysing
the preparation with TBS pH 7.5. Concentrations of NaSCN above 3 M were found to
degrade the 12S subunit antigen, enabling recovery of free antibody (data not shown).
Immunodiffusion and immunoelectrophoretic analysis. Ouchterlony-type immunodiffusion
tests were performed in 1% agar, barbital-glycine buffer with 0.15 M-NaC1 (Cowan et al.,
1974). Immunoelectrophoretic (IEP) analyses were performed in 1% agar, sodium-barbital
buffer (ionic strength 0.03, pH 8.6) according to the method of Hirschfeld (1960). Serum
samples (25 /A) were subjected to electrophoresis and purified 140S virus antigen or
rabbit-anti-normal guinea-pig (Ra-a-NGP) serum (200/21) was applied for precipitation.
Cell culture and neutralization tests. The Mengeling and Vaughn foetal porcine kidney cell
line (MVPK-1 clone 7) (Dinka et al., 1977) was grown and maintained in Eagle's minimum
essential medium (MEM; F-15, Grand Island Biological, Grand Island, N.Y., U.S.A.)
containing 1 raM-sodium pyruvate, 5 % (v/v) foetal bovine serum, penicillin (100 units/ml)
and streptomycin (50 gg/ml). Cell cultures were prepared in four-well (35-mm) plastic tissue
culture dishes (Falcon) and incubated at 37 °C in a humidified atmosphere of 2 % CO 2 in air.
Neutralizing activities of the antisera were measured in a plaque reduction-neutralization
test (PRNT) adapted from McVicar et al. (1974). Serial fivefold dilutions of antisera were
prepared in HLH, mixed with equal volumes of stock virus and incubated at 37 °C for 1 h.
The virus control was mixed with an equal volume of HLH and treated in the same manner.
Cell cultures were inoculated with 0.1 ml of reaction mixture, adsorbed for 1 h at 37 °C,
overlaid with 1% methyl cellulose (4000 centipoise) in F-15 and incubated for 48 h at 37 °C
in 2% CO2 in air. The number of plaque-forming units (p.f.u.) was plotted versus antibody
dilutions based on original serum volume before fractionation. Neutralizing titres and slopes
were calculated by computer by use of the logit-log transformation method (Trautman &
Harris, 1977).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:38:47
418
M . M . HARDY AND D. M. MOORE
Secondary neutralization with anti-immunoglobulin G serum. Secondary neutralization
was accomplished by the addition of anti-globulin (Majer, 1972). Rabbit-anti-guinea-pig
immunoglobulin G (Ra-a-GPIgG) was calibrated for equivalence with NGP serum by
radioimmunoprecipitation of ~25I-labelled guinea-pig immunoglobulin G (M. M. Hardy & D.
M. Moore, unpublished results). Mixtures of virus and antiserum were prepared and
incubated in the usual manner except that dilutions of antiserum of 1 : 250 and higher were
supplemented with 1 : 50 NGP serum to maintain equivalence. After incubation, each mixture
was divided, mixed with an equal volume of H L H or 1:5 Ra-a-GPIgG, incubated at 37 °C
and assayed for survivors. The virus controls, in 1 : 50 NGP serum, were also subdivided and
mixed with HLH or Ra-a-GPIgG as above.
RESULTS
In vitro neutralization reaction
An example of the type of neutralization response curve obtained by assaying anti-A12
serum with A12 virus in a standard PRNT in this study is shown in Fig. 1, line (a). After an
initial reduction in plaque counts with increasing serum concentration, infectivity again
increased sharply. At very high serum concentration ( < 1 : 5 0 dilution; -log10 = 1.7), the
infectivity again decreased, approaching zero at 1:2 (-log~0 = 0.3) antiserum dilutions. The
apex of this non-neutralized peak fluctuated in different tests between serum dilutions 1 : 50
and 1:250 (-log~0 = 2.4). Assays were done in duplicate with each point on the response
curves representing the mean value. The responses were consistent and reproducible in more
than 30 tests performed on different passages of the same cell line. Increasing the virus
concentration in the assay failed to alter the shape of the curve. As can be seen from the high
standard error and low slope values (insert, Fig. 1) for the response curve, the non-neutralized
fraction greatly interfered with an accurate measure of the neutralizing ability of the
antiserum. Fig. 1, line (b), represents the computer-predicted response for the 60%, 70% and
80% neutralization endpoints of Fig. 1, line (a), calculated by the logit transformation.
Ideally, this line should closely parallel the actual response. Fig. 1, line (b), demonstrates that
the empirical application of the endpoint calculation method would be invalid.
Effect of serum fractionation based on specificity on the neutralization of virus
Immunoadsorbent eolumn fractionation
The starting serum and the serum fractions collected from the 12S subunit immunoadsorbent column were compared by immunodiffusion with 12S and 140S purified antigens (Fig.
2). The starting serum formed precipitin lines with both the 12S and 140S antigens, but a
specific spur line formed with the 140S virion over the 12S subunit line. The 12S
subunit-adsorbed serum was specific for the 140S virion and failed to precipitate 12S antigen
(Fig. 2). The antibody eluted with 1 M-NaSCN precipitated both 140S and 12S antigens, but
with an apparent 140S virion-specific spur line. The antibody eluted with 2 M-NaSCN also
precipitated both 140S and 12S antigens, but with a 12S subunit-specific spur line.
The neutralization characteristics of the starting serum and column-fractionated serum
fractions were compared on MVPK cell monolayers by the P R N T method. Neutralization
results are shown in Fig. 3. The 12S subunit cross-reactive 1 M-NaSCN eluate displayed a
biphasic, non-linear dose-response curve in the area of high antibody concentration which
was almost identical to, but somewhat displaced to the left of, the response curve of the
starting serum. The 12S subunit cross-reactive 2 M-NaSCN eluate demonstrated a very
shallow dose-response curve. The response curve obtained for the 140S virion-specific serum,
however, was sigmoidal. Furthermore, computer-calculated 70% and 50% plaque reduction
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:38:47
FMD V sensitized by anti-12S/140S antibody
I
1
70 - v.c. (ax)
I
I
I
419
t
70% reduction std. slope
?
log titre error
/
(la) [ 2.58 1.81 0 . 0 9 /
6O -
-
/
50
(a)
40
1
30
/
g
~- 20
. . . . 2t-
10
I
|
I
I
I
2
2.58
3
-log~0 Serum dilution
I
I
4
4.72
Fig. 1. Neutralization of A12 strain 119 large-plaque 'ab' variant FMDV by guinea-pig antiserum,
incubated for 1 h at 37 °C: (a) typical response curve demonstrated by this antiserum; (b)
computer-predicted_response calculated by logit-log transformation of response (a). The mean virus
control value [v.c.(x)] is indicated for response curve (a); 60%, 70% and 80% = computer-calculated
plaque reduction endpoints.
Fig. 2. Effect of 12S subunit i,mmunoadsorbent column fractionation on specificity of antiserum to A12
strain 119 large-plaque 'ab' variant FMDV. Immunodiffusion analysis of the starting serum (ser), 12S
subunit column-adsorbed (12S ADS) and t M- and 2 M-NaSCN eluate (EL) fraction pools for 140S and
12S antigen reactivity, are shown.
e n d p o i n t s exactly s u p e r i m p o s e d o n the experimental line (Fig. 3). T h e 140S virion-specific
s e r u m pool d e m o n s t r a t e d calculated slopes r a n g i n g f r o m 0 . 7 5 to 1.1; the slope values
calculated for the 12S cross-reactive fractions a n d starting s e r u m were low, r a n g i n g from
0 . 0 9 to 0 . 3 4 .
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:38:47
420
M. M. H A R D Y
I
AND
i
D . M. M O O R E
i
i
I
70
6O
50
v
.- 40
E
3o
g
~ 20
50 %
10
0
I
I
I
2
l
3
logt0 Serum dilution
I
I
4
5
Fig. 3. Neutralization response curves of the starting and 12S subunit column fractionated serum pools
incubated for 1 h, 37 °C with A12 strain 119 large-plaque 'ab' variant FMDV: Q, starting serum; A,
140S virion-specific fraction; [3, 1 M-NaSCN eluate; 0 , 2 M-NaSCN 12S subunit cross-reactive eluate.
v.c. (~) = mean virus control value; 50 % and 70 % = computer-calculated plaque reduction endpoints.
The starting serum and all three fractionated pools completely neutralized the virus at high
antiserum concentrations ( < 1 : 2 dilution). However, because of the concentration of the
fractionated samples during dialysis, the relative volume of serum added to virus was greater
than that possible for unconcentrated serum. The equivalent of undiluted 1 M- and 2
M-NaSCN eluate serum pools demonstrated a non-neutralized infectivity of 0.5 to 1% when
allowed to react with virus. This non-neutralized fraction was not seen with the virion-specific
pool. The non-neutralized peak of the starting serum was associated only with the 12S
subunit-reactive antibody after fractionation.
The starting serum and the 1 M-NaSCN eluate 12S subunit-reactive pool occasionally
markedly enhanced infectivity in the non-neutralized peak region of the neutralization
response. Also noticeable was a decrease in plaque size in very high serum concentrations
( < 1 : 5 0 dilution) of the starting serum and 1 M- and 2 M-NaSCN eluate pools. The 140S
virion-specific serum pool produced no such plaque-size reduction.
Solution fractionation
Immunodiffusion tests on the serum fractions attained by solution fractionation
substantiated that the 12S subunit cross-reactive and 140S virion-specific serum pools had
the same specificity as the corresponding column-fractionated pools.
When the solution-fractionated pools were tested for neutralizing ability, the aberrant
neutralization response was again absent with the 140S virion-specific pool and present with
the 12S cr0ss-reactive fraction (data not shown). The column-fractionated serum, therefore,
was used in all subsequent tests in this investigation.
Viral aggregation and viral type or antibody class heterogeneity as factors influencing the
neutralization response
Previous investigations (Wallis & Melnick, 1967) have proposed that the persistence of
Downloaded from www.microbiologyresearch.org by
infectivity in reactions with low dilutions of serum could be due to the presence of large
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:38:47
FMD V sensitized by anti-12 S /140 S antibody
421
Fig. 4. Immunoelectrophoresisof anti-A~2 strain 119 large-plaque 'ab' variant FMDV starting serum
and 12S subunit column-fractionatedserum pools: (a) starting serum; (b) 140S virion-specificfraction
(12S absorbed); (c) I M-NaSCN eluate; (d) 2 M-NaSCN eluate. Top troughs contained rabbitanti-normal guinea-pig serum (Ra-a-NGP). Bottom troughs contained A12 strain 119 large plaque 'ab'
variant FMDV purified 140S antigen (140S).
aggregates of virus that, when used in a neutralization test, become complexed with antibody
only on the periphery and dissociate when inoculated on to the cell culture surfaces. This
hypothesis was tested by filtering the virus stock before reaction with antibody to remove
large aggregates. The filtration had no effect on the shape of the neutralization response
curve, indicating that aggregation was not responsible for the non-neutralized fraction.
Because of the possible interference from aggregation, however, all data presented in this
investigation were obtained with prefiltered virus preparations.
Although the virus stock was obtained by plaque isolation, the appearance of a
neutralization-resistant population of virus suggested the possible existence of virus
heterogeneity. However, a P R N T with non-neutralized peak-derived virus progeny and the
fractionated serum pools demonstrated the same general trend as with the parent population
(i.e. the non-neutralized peak was associated solely with 12S subunit-reactive antibody) (data
not shown).
Characterization by IEP analysis of the antibody classes present in the four serum
fractions indicated that only the immunoglobulin G (IgG) fraction reacted with the 140S
virus antigen (Fig. 4). The presence of serum proteins other than IgG in the 1 u- and 2
M-NaSCN 12S-reactive Downloaded
eluates is from
duewww.microbiologyresearch.org
to the addition of N G P byserum as carrier which
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:38:47
422
M. M. HARDY
I
I
70 ,- v.c.(x)
Ra-a-GPlgG v.c. (:~)
I
AND
I
i
D. M. MOORE
-v.c,(~)
(a) - +Ra
I
I
I
. ~/¢'""
1
~ (b)
O
50
0" /
30
/
,,'
,'
" 0
•~ 70
I......
1
/' 1 : 3 1 2 5 0
t.-~ f
'I
1:1250
I
1
.'"
!
(c) - -,-- v .c. (i)
"-- v.c.(x)
: . + R a a G P I g G v.c. (:~)
(d) -
~a
,o
¢
~- 5o
• _
/
/
/
/
30'
/
/
,i
/
/
I0
0
i
2
I
3
.¢'i '~ 1:6250 I
4
5
I
I
I~
6
2
3
,/
Xf
4
1:6250
I
5
--log~o Serum dilution
Fig. 5. Primary (
) and secondary ( - - - ) neutralization of A~2 strain 119 large-plaque 'ab' variant
F M D V by: (a) starting serum; (b) 140S virion-specific fraction; (c) 1 M-NaSCN 12S subunit
cross-reactive eluate; and (d) 2 M-NaSCN 12S subunit cross-reactive eluate. Secondary neutralization
of the_ primary reaction was accomplished by the addition of rabbit-anti-guinea-pig I g G (Ra-a-GPIgG).
v.c. (x) = mean virus control value.
permitted the relative comparison of serum immunoglobulin classes present in the purified
immune serum fractions. The heavy Ra-a-NGP precipitin lines appearing in Fig. 4 (a) and
4 (b) are due to an overload of anti-A~2 serum in the centre well which was necessary to
produce a virus precipitin line. The overload of 2 M-NaSCN eluate antibody produced only a
light precipitin line upon reaction with 140S antigen which is barely visible in Fig. 4(d).
Dependence of neutralization on reaction time
Comparison of the neutralization responses of the starting and fractionated pools after
incubation for 1 h (37 °C) and 18 h (4 °C) intervals showed that although the height of the
non-neutralized peak was slightly diminished by the increased reaction time, the shape of the
response was not greatly altered (data not shown). The neutralization response of the 140S
virion-specific antibody fraction was completely unaffected.
Demonstration of infectious immune complexes by secondary neutralization
Most non-neutralized fractions found in virus populations have been attributed to the
presence of virus that has reacted with specific antibody (is sensitized) but has not been
neutralized by that antibody (Daniels, 1975; Della-Porta & Westaway, 1978). Whether such
infectious immune complexes were present and whether their presence affected the shape of
the dose-response curve was studied by exposing the primary neutralization mixture to
Ra-a-GPIgG for secondary neutralization. The similarity of response of the two virus control
means indicated that the virus infectivity was not affected by the anti-globulin (Fig. 5).
However, secondary neutralization markedly altered the shape of the three aberrant
dose-response curves and enhanced the antibody neutralizing titre for all four serum samples.
The responses of the primary reaction mixtures remained unchanged from previous results.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:38:47
F M D V sensitized by anti-12S/140S antibody
423
Table 1. Primary and secondary* 70% plaque reduction log neutralization titrest for
anti-A ~ strain 119 large-plaque 'ab' variant FMD V starting serum and 12S subunit columnfractionated serum pools
Primary neutralization
Secondary neutralization
&
.&
f-
Starting serum
140S virion-specific
fraction
1 M-NaSCN eluate
12S-reactive fraction
2 M-NaSCN eluate
12S-reactive fraction
Standard
error
0.10
<0.10
Titre
1.45
2.33
Standard
error
1.40
0.14
Slope
0.12
0,61
Titre
5.00
3-89
0.13
9.55
0.10
4.62
0.02
1.85
0.68
0.74
0-29
4-13
0-42
0.75
Slope
2.07
t-69
* Secondary neutralization of the primary reaction was accomplished by the addition of rabbit-anti-guinea-pig
IgG.
t Titres represent those calculated from the responses presented in Fig. 5.
This observed alteration in curve shapes upon second neutralization served to normalize the
responses into sigmoid-type curves, allowing the logit calculation of titres with low error and
steep dose-response curves (Fig. 5 & Table 1). Enhancement of neutralizing titres by
secondary neutralization demonstrated that virus was sensitized with all four serum fractions.
The removal of this sensitized virus through secondary neutralization effectively eliminated
the non-neutralized peak.
Non-neutralized fractions with FMD V A 24
Several sera to FMDV type A24 w e r e tested to confirm the occurrence of non-neutralizing
aberrant responses with FMDV. A non-neutralized fraction was found and the presence of
sensitized A24 virus was detected, through the use of second neutralization, in a PRNT with a
guinea-pig (GP) anti-Az4 hyperimmune (infected) serum, a GP anti-A24 140S purified antigen
immune serum, and a GP anti-A24 12S purified antigen immune serum. Filtration of the A24
strain virus stock also did not alter the responses.
DISCUSSION
An in-depth study of the neutralization of animal viruses by antibody was first described by
Dulbecco et al. (1956). They found that neutralization was a first-order reaction, the slope of
which was proportional to the concentration of the antibody. The theoretical neutralization
curve, therefore, as proposed in the study, was sigmoid or occasionally linear. This type of
response curve has come to be known as the expected or usual neutralization response.
Svehag (1968) contended, however, that the neutralization plot cannot be expected to be
curvilinear if the antibody population is heterogeneous in affinity. The results of this study
demonstrate that, at least for this system, efficiency of neutralization is directly linked to the
specificity of the antibody, and a non-sigmoid neutralization plot to the heterogeneity of
antibody specificities. When the 12S subunit cross-reactive (sensitizing) antibody was
removed from the serum, leaving only the 140S virion-specific (neutralizing) antibody, the
neutralization reaction was sigmoidal, with the amount of neutralization proportional to
antibody concentration. FMDV has been demonstrated to have at least three different
antigenic sites that stimulate the production of antibodies (Cowan, 1968; Wild et al., 1969;
Brown & Smale, 1970). Brown & Smale (1970), in particular, showed these three
immunogenic sites to be a 140S virion/12S subunit shared site, a 140S virion-specific
trypsin-sensitive site and a 140S virion-specific site that was not trypsin-sensitive. Rowlands
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:38:47
424
M . M. H A R D Y
AND
D . M. M O O R E
et al. (1971) demonstrated the latter two sites to produce neutralizing antibody, although in
different concentrations. The number of different antigenic determinants on the virus particle
and whether all of them are critical are, therefore, important factors.
Two fractions of 12S subunit/140S virion cross-reactive antibody separated by column
elution were shown to be of differing specificities by immunodiffusion and neutralization,
suggesting at least four different antigenic sites on the 140S intact virion. With the 1
M-NaSCN eluate, the 140S antigen-specific spur line that appeared over the 12S antigen line
may indicate the presence of antibody that has a greater affinity or complementarity for the
140S antigen than the 12S component, even though it reacted with the 12S subunitconjugated column. The absorption of 12S subunit reactivity from the serum in solution, its
elution, and subsequent neutralizing results, which paralleled those found for the column
fractionation, indicated that the solid-phase 12S subunit column had not been distorted or
aggregated to mimic intact virus. The presence of neutralizing activity in the 2 M-NaSCN
fraction suggests the presence of at least one type of 140S-12S cross-reacting antibody that
has higher affinity for 12S subunits and is different from the 140S-12S cross-reactivity of the
1 M-NaSCN antibody.
Since the neutralization response curves for all serum fractions varied in reactivity, ranging
from good to very poor neutralization, the possibility that the effect was due to aggregation or
heterogeneity of virus assay stock, as proposed by Wallis & Melnick (1967), is doubtful. In
addition to this report, prior filtration was reported as incapable of preventing the formation
of a non-neutralized fraction by Ashe et al. (1969); Majer & Link (1970), Lewenton-Kriss &
Mandel (1972), Booth et al. (1978). The non-neutralized fraction did not represent a resistant
subpopulation of virus for the progeny displayed neutralization patterns identical to those of
the parental stock virus (Dulbecco et al., 1956; Ashe & Notkins, 1967; Majer & Link, 1970).
The non-neutralized fraction was not a result of insufficient antibody, since the shape of the
neutralization curves was unaltered by increasing virus concentration. Excess antibody was
shown by the decrease in plaque size at high concentrations of serum (Kjellen & Schlesinger,
1959; Wecker, 1960). Washing of the inoculated cell cultures before application of the
overlay allowed normal plaque growth and size. The numbers of plaques remained
unchanged, showing that the excess antibody did not affect the response and that the
non-neutralized fraction was not caused by dissociated neutralized virus in the overlay.
Neutralization results with different host cells showed an absence of the non-neutralized peak
(M. M. Hardy & D. M. Moore, unpublished data), further demonstrating that dissociation did
not occur. Dissociation of virus-antibody complexes is highly improbable, as shown by the
work of Mandel (1961) and other investigators (Dulbecco et al., 1956; Ashe & Notkins,
1967; Svehag, 1968; Radwan & Burger, 1973). This result also demonstrated that the
non-neutralized fraction was not caused by dissociation of viral aggregates which may have
passed the filter. Further investigations have indicated that the MVPK cells are more sensitive
in discriminating between FMDV infectivity and neutralization (M. M. Hardy & D. M.
Moore, unpublished data).
It is widely accepted that the non-neutralized fractions of most viruses are not particles that
have failed to react with antibodies, but consist of virus-antibody complexes that can be
eliminated by addition of anti-globulin (Mandel, 1958; Hahon, 1970; Wallis & Melnick,
1970; Lewenton-Kriss & Mandel, 1972; Majer, 1972; Booth et al., 1978). Rweyemamu et al.
(1977) demonstrated that treatment of FMDV neutralization mixtures with anti-rabbit-IgG
enhanced neutralization without a significant alteration in the relationship between virus and
attached antibody. In this study, the degree of sensitization of the virus varied for the four
serum fractions tested. The 140S virion-specific serum-reacted virus, although it did not
display a non-neutralized peak, was sensitized. This finding supports Majer's (1972)
conclusions that sensitized virus is not restricted to the non-neutralized fraction and that
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:38:47
FMD V sensitized by anti-12S/140S antibody
425
sensitization with neutralizing antisera always precedes neutralization. This sensitization is
not perceived by the authors as the first step in a two-step stabilization like that favoured by
Svehag (1968). In that theory, the initial monovalent neutralizing antibody-virus reaction is
reversible (and sensitizing), while a secondary two combining-site antibody reaction is
stabilizing (neutralizing). This study does not support this stabilization theory, because
increased incubation time had no major effect on the responses. It is more likely that, as in a
frequency distribution, there were not enough antibody-reacted critical sites to neutralize
these virus particles. Indeed, Trautman & Harris (1977) proposed such a model for the
FMDV neutralization response. This sensitization is different from that caused by sensitizing
antibody in that it is due to a population or concentration effect, rather than to antigenic
specificity.
The 12S subunit cross-reactive serum fractions contained mostly, if not only, sensitizing
antibodies. The 2 M-NaSCN 12S subunit cross-reactive fraction demonstrated the least
neutralizing ability but the same degree of sensitization (5 times less than the start) as the 1
M-NaSCN 12S subunit cross-reactive antibody in the secondary reaction. This finding would
indicate that they contained the same relative concentrations of antiviral antibody. The fact
that the 1 M-NaSCN 12S subunit cross-reactive fraction produced a neutralization curve
identical in shape to that produced by the starting serum may reflect a mixture of similar
antibody specificities. The presence of the 140S virion-specific antibody and cross-reactive
antibody with higher affinity for the 140S virion evidently contribute to the shape of the curve
because the 2 M-eluate, which contains neither of these, had the shallowest response curve and
lowest titre.
A non-neutralized fraction is more evident with immunoglobulin M (IgM), because of its
low affinity for and high dissociability with virus (Gard, 1957; Lafferty, 1963). LewentonKriss & Mandel (1972) were unable to isolate a solely sensitizing antibody and found that
IgG and IgM both sensitized and neutralized. The results of this study also demonstrate that
IgG both neutralized and sensitized. Brown & Smale (1970) showed that 12S subunit does
not react with IgM indicating that IgM would not have been present in the 12S-reactive
fractions and would not have been an influencing factor in this study.
Sensitization of virus has been found for several other FMDV systems. R. C. Knudsen
(Plum Island Animal Disease Center, unpublished data) found a very poor in vitro
neutralization response for a GP-adapted anti-A12 convalescent serum with MVPK, BHK
and GP tongue explant cultures. Secondary neutralization detected the presence of sensitizing
antibody and resulted in neutralization. Non-neutralized aberrant response curves similar to
those of the Al2 system have been observed for FMDV type O t, strain Casaros
(B. Gametchu, unpublished data). These aberrant responses were observed with both
monolayer and suspension cell-derived virus, with infected and hyperimmunized GP sera, and
with infected hyperimmunized bovine sera. When these sera were adsorbed on a
Sepharose-12S subunit column, the 12S subunit-adsorbed sera displayed 'normal' responses,
while the responses of the 12S subunit-reactive eluate remained atypical (B. Gametehu,
unpublished data). Additionally, antisera to an isolated FMDV capsid protein, VP 3 (the
external virus protein, also called VP~ or VP-threonin), demonstrated sensitization in in vitro
and neutralization in in vivo assay. Guinea-pig antiserum to VP 3 displayed protracted
non-neutralized atypical neutralization response curves. When this serum was 12S-subunit
column adsorbed, it was found to contain only 12S subunit cross-reactive and no 140S
virion-specific antibody (D. M. Moore, unpublished observations). In general, non-neutralized
fractions and sensitized virus have been reported for such a wide variety of viruses that
reports of the absence of sensitization are perhaps more unusual.
This report proposes that the 140S virion-specific antibody is directed against, or specific
to, the critical site(s) of the virion. This antibody is directly neutralizing in high enough
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:38:47
426
M. M. HARDY
AND
D. M. MOORE
concentrations to react with or complete the required number of critical sites. The I2S
subunit cross-reactive antibody is primarily sensitizing antibody that is unable to neutralize
directly, in vitro, because of its specificity for a non-critical site(s). Any neutralization
occurring with antibody of this specificity is due to a blocking of critical sites through steric
hindrance.
The authors thank Drs Ellen M. Duffy and Gerald Green for their careful review and constructive
criticism of this manuscript. The authors also are grateful for the excellent photographic assistance of
Mr Tracy Durham (deceased).
REFERENCES
ASHE, W. K. & NOTKINS, A. L. (1967). Kenetics of sensitization of herpes simplex virus and its relationship to the
reduction in the neutralization rate constant. Virology 33, 613-617.
ASHE, W. K., MAGE, M. & NOTKINS, A. L. (1969). Kenetics of neutralization of sensitized herpes simplex virus with
antibody fragments. Virology 37, 290-293.
BACHRACH, H. L., TRAUTMAN, R. & BREESE, S. S. (1964). Chemical and physical properties of virtually pure
foot-and-mouth disease virus. American Journal of Veterinary Research 25, 333-342.
BAHNEMANN, n. G. (1975). Binary ethyleneimine as an inactivant for foot-and-mouth disease virus and its
application for vaccine production. Archives of Virology 47, 47-56.
BOOTH, J. C., RWEYEMAMU,M. M. & PAY, T. W. (1978). Dose-response relationships in a micro-neutralization test for
foot-and-mouth disease viruses. Journal of Hygiene 80, 31-42.
BRADlSH, C. J., FARLEY, J. O. & FERRIER, H. E. N. (1962). Studies on the nature of the neutralization reaction and the
competition for neutralizing antibody between components of the virus system of foot-and-mouth disease.
Virology 18, 378-400.
BROWN, F. & CARTWRIGHT, B. (1963). Purification of radioactive foot-and-mouth disease virus. Nature, London
199, 1168-1170.
BROWN, F. & SMALL, C. J. (1970). Demonstration of three specific sites on the surface of foot-and-mouth disease
virus by antibody complexing. Journal of General Virology 7, 115-127.
COWAN, K. U. (1968). Immunochemical studies of foot-and-mouth disease. IV. Preparation and evaluation of
antisera specific for virus, virus protein subunit and the infection-associated antigen. Journal of Immunology
101, 1183-1191.
COWAN, K. M. (1969). Immunological studies of foot-and-mouth disease. V. Antigenic variants of virus
demonstrated by immunodiffusion analysis with 19S but not 7S antibodies. Journal of Experimental
Medicine 129, 333-350.
COWAN, K. M., EROL~N. & WHITELAND, A. P. (1974). Heterogeneity of type Asia, foot-and-mouth disease virus and
BHK-21 cells and the relationship to vaccine production. Bulletin de l'Office International des Epizooties 81,
1271-1298.
DANIELS, C. A. (1975). Mechanisms of viral neutralization. In Viral Immunology and Immunopathology, pp.
79-97. Edited by A. L. Notkins. New York: Academic Press.
DELLA-PORTA, A. J. & WESTAWAY,E. G. (1978). A multi-hit model for the neutralization of animal viruses. Journal
of General Virology 38, 1-19.
DINKA, S. K., SWANEY, L. M. & McVICAR, J. W. (1977). Selection of a stable clone of the MVPK-1 fetal porcine
kidney cell for assays of foot-and-mouth disease virus. Canadian Journal of Microbiology 23, 295-299.
DULBECCO, R., VOGT, R. M. & STRICKLAND,A. G. R. (1956). A study of the basic aspects of neutralization of two
animal viruses, Western equine encephalitis virus and poliomyelitis virus. Virology 2, 162-205.
FIALA, Mo (1969). A study of the neutralization of rhinoviruses by means of a plaque-reduction test. Journal of
Immunology 103, 107-113.
GARD, S. (1957). Immunoinactivation of poliovirus. Archivfffr die Gesamte Virusforschung 7, 449-460.
HAHON, N. (1970). Neutralization of residual infectivity of Venezuelan equine encephalomyelitis virus by
anti-gamma globulin. Journal of General Virology 6, 361-372.
HIRSCHFELD, J. (1960). Immunoelectrophoresis: procedure and application to the study of group-specific
variations in sera. Science Tools (LKB Produkter) 7, 18-25.
KJELLEN, L. E. & SCHLESINGER, R. W. (1959). Influence of host cell on residual infectivity of neutralized vesicular
stomatitis virus. Virology 7, 236-239.
LAFFERTY, K. J. (1963). The interaction between virus and antibody. II. Mechanism of the reaction. Virology 21,
76-90.
LEWENTON-KRISS, S. & MANDEL, B. (1972). Studies on the non-neutralizable fraction of poliovirus. Virology 48,
666-678.
McVICAR, J. W., SUTMOLLER, P. & ANDERSEN, A. A. (1974). Foot-and-mouth disease virus: plaque reduction
neutralization test. Archiv fu'r die Gesamte Virusforschung 44, 168-172.
MAJER, M. (1972). Virus sensitization. Current Topics in Microbiology and Immunology 58, 69-84.
MAJER, M. & LINK, F. (1970). Studies on the non-neutralizable fraction of vaccinia virus. Clinical and Experimental
Immunology 7, 283-291.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:38:47
FMD V sensitized by anti-12S/140S antibody
427
MANDEL, B. (1958). Studies on the interactions of poliomyelitis virus, antibody, and host cells in a tissue culture
system. Virology 6, 424-447.
MANDEL, B. (1961). Reversibility of the reaction between poliovirus and neutralizing antibody of rabbit origin.
Virology 14, 316-328.
MANDEL, B. (1976). Neutralization of poliovirus: a hypothesis to explain the mechanism and the one-hit character
of the neutralization reaction. Virology 69, 500-510.
MARTtNSEN, J. S. (1970). The effect of diethylaminoethyl dextran and agar overlay pH on plaque formation by two
plaque size variants of foot-and-mouth disease. Canadian Journal of Comparative Medicine 34, 13-19.
MOORE, D. M. & COWAN, K. M. (1978). Effect of trypsin and chymotrypsin on the polypeptides of large and small
plaque variants of foot-and-mouth disease virus: relationship to specific antigenicity and infectivity. Journal
of General Virology 41,549-562.
MORGAN, D. O. & McKERCHER, P. D. (1977). Immune response of neonatal swine to inactivated foot-and-mouth
disease virus vaccine with oil adjuvant. I. Influence of colostral antibody. Proceedings of the 81st Annual
Meeting of the U.S. Animal Health Association, Minneapolis, pp. 244-255.
PmLIPSON, L (1966). Interaction between poliovirus and immunoglobulins. II. Basic aspects of virus-antibody
interaction. Virology 28, 35-46.
POLATNICK, J. & BACHRACH, H. L. (1964). Production and purification of milligram amounts of foot-and-mouth
disease virus from baby hamster kidney cell cultures. Applied Microbiology 12, 368-373.
RADWAN, A. I. ~, BUROER, D. (1973). The role of sensitizing antibody in the neutralization of equine arteritis virus
by complement or anti-lgG serum. Virology 53, 366-371.
ROWLANDS, D. J., SANGAR, D. V. & BROWN F. (1971). Relationship of antigenic structure of foot-and-mouth disease
virus to the process of infection. Journal of General Virology 13, 85-93.
RWEYEMAMU, M. M., BOOTH, J. C. & PAY, T. W. F. (1977). Neutralization kinetics studies with type SAT-2
foot-and-mouth disease virus strains. I. Factors that influence the rate and pattern of neutralization. Journal
of Hygiene 78, 99-110.
SUTMOLLEa, O. & COWAN, K. M. (1974). The detection of foot-and-mouth disease virus antigens in infected cell
cultures by immunoperoxidase techniques. Journal of General Virology 22, 287-291.
SVEHAG, S. E. (1968). Formation and dissociation of virus-antibody complexes with special reference to the
neutralization process. Progress in Medical Virology 10, 1-63.
TRAUTMAN, R. & HARRIS, F. (1977). Modeling and computer simulation approach to the mechanism o f
foot-and-mouth disease virus neutralization assays. Scandinavian Journal of Immunology 6, 831-841.
WAGNER, t3. G., CARD, S. L. & COWAN, K. M. (1970). Immunochemical studies of foot-and-mouth disease. VII.
Characterization of foot-and-mouth disease virus concentrated by polyethylene glycol precipitation. Archly
ffir die Gesamte Virusforschung 30, 343-352.
WALLIS, C. ~ MELNICK, J. L. (1967). Virus aggregation as the cause of the non-neutralizable persistent fraction.
Journal of Virology 1,478-488.
WALLIS, C. & MELNICK, J. L. (1970). Herpesvirus neutralization: induction of the persistent fraction by insufficient
antibody. Virology 42, 128-137.
WECKER, E. (1960). A simple test for serodifferentiation of poliovirus strains within the same type. Virology 10,
376-379.
WILD, T. F., BURROUGHS, J. N. & BROWN, F. (1969). Surface structure of foot-and-mouth disease virus. Journal of
General Virology 4, 313-320.
(Received 26 November 1980)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:38:47