J. gen. Virol. 098o), 5I, 157-17o
Printed in Great Britain
t57
Immunoassay of Poliovirus Antigens by Single-Radial-Diffusion:
Development and Characteristics of a Sensitive Autoradiographic
Zone Size Enhancement (ZE) Technique
By G. C. S C H I L D , J. M. W O O D , P. D. M I N O R ,
C. N. D A N D A W A T E * AND D. I. M A G R A T H
Division of Viral Products, National Institute for Biological Standards and Control,
Holly Hill, Hampstead, London NW3 6RB, U.K.
(Accepted 20 June 198o)
SUMMARY
The reactions of polioviruses in single-radial-immunodiffusion (SRD) tests
were investigated with a view to developing accurate and sensitive antigen assay
systems. In direct SRD tests, employing high concentrations of immune poliovirus serum in agarose gels, poliovirus D-antigens produced clear reaction zones
demonstrated by protein staining. The reactions were type-specific for polioviruses
of types I, 2 and 3 but the tests were of low sensitivity, being applicable only to the
assay of virus concentrates.
A novel autoradiographic zone size enhancement (ZE) test was developed which
increased the sensitivity of the SRD assay 40- to Ioo-fold. The ZE test was dependent upon the ability of unlabelled poliovirus to co-migrate with the radioactive
marker virus and so enhance the zone size detected autoradiographically. The areas
of the autoradiographic zones were directly proportional to the concentration of
unlabelled antigen. The ZE test was capable of detecting poliovirus D antigens in
diluted cell culture fluid harvests in amounts corresponding to IO3.3 to Io4"8TCIDs0
of infectious virus.
Studies with poliovirus type 3 strains indicated that the ZE test was narrowly
strain-specific for the D-antigen of poliovirus type 3 strains when homologous
type 3 D-antigen was used as radioactive marker, but broadly cross-reactive for the
D-antigen of type 3 viruses when heterologous poliovirus type 3 D-antigen was
used as marker.
INTRODUCTION
The ability to perform precise, sensitive and specific in vitro assays of virus antigens is
relevant to the understanding of the immunological characteristics of viruses, and in certain
circumstances has practical applications in the standardization of virus vaccines. Singleradial-diffusion (SRD) techniques have been widely applied for antigen assays and have
characteristics which enable reproducible and accurate quantification (Rumke & Thung,
I964; Mancini et al. I965). These methods depend upon the formation of visible zones of
immunoprecipitate when antigens diffuse radially from wells into agarose gels in which
specific antibody is uniformly distributed. They have been applied to a limited number of
virus antigens, including adenovirus hexon (Pereira et al. I972) and influenza virus haemagglutinin (Schild et al. I975) and foot-and-mouth disease virus (Cowan & Wagner, I97O).
* Present address: Virus Research Centre, 2oA Dr Ambedkar Road, P.O. Box i I, Pane 4I tOOl, India.
Downloaded
oo22-I317/8o/oooo-4248 $o2.oo
© I98Ofrom
SGMwww.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26
158
G. C. S C H I L D A N D O T H E R S
Despite the importance of polioviruses in human disease and as antigens for prophylactic
use in man, little progress has been made in the development of sensitive in vitro assay
systems for poliovirus antigens. The present paper describes studies on the use of conventional (direct) SRD tests for the assay of poliovirus antigens and also describes the
development and characteristics of a novel, autoradiographic zone size enhancement (ZE)
system for antigen assay which was of greatly increased sensitivity compared to SRD tests
employing protein staining.
METHODS
Polioviruses. Poliovirus type I (Mahoney strain, PI-Mh), poliovirus type 2 (MEF strain,
Pz-MEF) and poliovirus type 3 [3o, 19o and Saukette (P3-3o, P3-I9o and P3-Sauk)] were
• wild' strains from the collection of this laboratory. Other polioviruses employed were the
Sabin vaccine strains of poliovirus types ~, 2 and 3 (PI-Sab, P2-Sab, P3-Sab) and poliovirus
type 3 strains I I5 and 119 (P3-1 r5, P3-1 I9) isolated from recipients of live Sabin vaccine,
having antigenic and biochemical characteristics resembling those of the Sabin type 3
vaccine virus (Minor, 198o; A. L. Van Wezel & P. D. Minor, unpublished results).
Poliovirus stocks were prepared by infecting monolayer cultures of HEP2 (Cincinatti)
cells at high virus input multiplicities. The cells were maintained at 35 °C in Eagle's basal
medium (BME) with lactalbumin hydrolysate (o'5 ~) and harvested after 24 to 48 h when
there was extensive cell destruction.
For some studies, virus was concentrated and partially purified from clarified (centrifuged
~ooo g for 3o rain) tissue culture fluids by treating the fluids with polyethylene glycol 6ooo to
precipitate the virus particles (Esposito, I976). Sodium chloride (final concentration,
o'5 M) and polyethylene glycol (7 ~o, w/v) were added to the clarified tissue culture fluids and
the mixtures held at 4 °C for I8 h before centrifuging at 8o0o g for 60 min. The resultant
virus-rich precipitates were resuspended in phosphate-buffered saline (PBS) pH 6"4, to
to 2 ~ of the original volume, homogenized and finally clarified by further low speed
centrifugation. The infectivity titres of the final concentrates ranged from IO1°'° to ro 11"5
TCIDs0/ml.
Virus #!feetivity and infectivity neutralization (VN) tests. These were performed on HEP2
cell cultures using microtitre techniques as described by D6m6k & Magrath (I979).
Preparation of purified 35S-labelled poliovirus. 35S-labelled poliovirus antigens were
prepared as described by Minor et al. 098o). Radioactive D- and C-antigens (Mayer et aL
~957; Roizman et al. I957, I958), harvested as the I55S and 8oS radioactive peaks from
sucrose density gradients and shown to be antigenically pure by autoradiographic SRD
tests with anti-D and anti-C sera (Minor et al. 198o), were used as markers in the ZE tests.
Preparation of immune sera
Antisera to unfractionated potiovirus antigen (anti-V sera). Rabbits or guinea-pigs were
immunized with o.2 ml vol. of concentrated poliovirus antigens. Animals were given two or
three intramuscular injections of antigen at intervals of 2 weeks. Sera were collected 7 days
after the last injection. Antisera selected for use in the studies possessed virus neutralization
(VN) titres of I : 5000 against the homologous virus and gave a single well-defined precipitin
line when tested against concentrated virus in double-immunodiffusion (DID) tests (Van
Wezel & Hazendonk, I979).
Antisera to purified poliovirus D- and C-antigens. The preparation and properties of
these specific anti-D and anti-C sera for poliovirus type 3 antigens are described elsewhere
(Minor et al. I98o). Antisera to D-antigens were prepared in guinea-pigs and antisera to
C-antigens were prepared in rabbits.
Direct single-radial-diffusion (SRD) tests. SRD tests were carried out as described for
influenza virus haemagglutinin by Wood et al. (I977) except that Sarcosyl detergent was
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26
Sensitive assay for poliovirus antigen
159
added to the agarose gels in order to facilitate uniform diffusion of poliovirus antigens.
Briefly, immunoplates were prepared using agarose gels (I ~ Seakem, medium EEO, Marine
Colloids, Rockland, U.S.A. : prepared in PBS pH 7q, containing o.oI ~ sodium azide and
o - o z 5 ~ sodium sarcosyl, NL 97) to which was added immune poliovirus serum. The poliovirus antigens for assay were added to 4 mm diam. wells in the immunoplates. After a
diffusion period of 2 to 7 days, the immunoplates were washed, dried and stained with
Coomassie brilliant blue G-25o (Serva Feinbiochemica, Heidelberg, Germany), and
diffusion zones were measured using a Transidyne Calibrating Viewer (Transidyne General
Corp., Ann Arbor, U.S.A.) coupled to Autodata digital caliper equipment (Autodata,
Hitchin, U.K.).
Autoradiography of immunoplates. Immunoplates for autoradiography were washed in
PBS for 24 h to remove free 35S-labelled antigen, pressed, dried (Wood et al. I977, I98o)
and contact exposures prepared on X-ray film (Fuji Photo Film Co., Japan) using exposure
times of z to Io days. The diameters of the autoradiographic zones indicating limits of
diffusion of the 35S-labelled antigens were measured as described above for stained SRD
zones.
Zone size enhancement (ZE) assays. Immunoplates for ZE tests were prepared as for direct
SRD tests (above) but with lower concentrations of immune poliovirus sera (anti-V, anti-D
or anti-C; final serum dilution in gel lO -4 to Io-6). Twenty/d vol. of appropriate dilutions
of unlabelled poliovirus antigen in PBS were added to each well, followed 3o to 6o min
later by I o / d of a standard dilution of 35S-labelled poliovirus D- or C-antigen as marker.
Standard 'challenge' dilutions of the labelled antigens were selected which, in the absence
of added unlabelled antigen, produced small (5 mm diam.) autoradiographic zones. Such
dilutions had radioactive counts of 2o0 to 4oo ct/min per Io/A vol. The plates were left at
room temperature for 2 to 3 days to allow for diffusion of the antigens and the antibodyantigen reaction zones detected autoradiographically. Zones surrounding wells containing
unlabelled assay antigen and marker, which were o'5 mm greater in diam. than the zones
surrounding control wells containing radioactive marker and PBS only, were considered to
be enhanced.
RESULTS
Direct SRD assays
The sensitivity and specificity of direct SRD tests for the detection of poliovirus antigens
were examined using Coomassie blue staining to demonstrate the zones of antibodyantigen reaction. Potent rabbit antisera prepared against concentrated, unfractionated
poliovirus antigens (anti-V sera) for polioviruses of types I, 2 and 3 or guinea-pig antisera
to purified D-antigen of poliovirus type 3 (anti-D sera) were incorporated into agarose gels
at a range of dilutions from lo -~ to IO-4 and poliovirus concentrates added to wells in the
immunoplates.
The plates were examined at intervals over a 7 day period unstained, by dark ground
illumination and after staining. Table I shows the minimum concentration in agarose gel of
each of the sera required for the production of well-defined, stained zones in SRD tests with
homologous antigen and the homologous VN titres of the sera. The minimal concentrations
of immune rabbit sera required for SRD tests were between 2o- and 5o-fold higher than the
homologous VN end-point titres. A guinea-pig anti-D serum (Table I) with potent VN
activity was relatively poor in SRD tests. The most potent serum with a VN titre of I : IOOOOO
gave clear SRD reactions when used at a concentration of o'5 to o'7/zl serum/ml gel. In
tests with limiting concentrations of antisera the SRD zones were not visible before staining.
However, well-defined opalescent zones were seen with homologous antigens on unstained
immunoplates when concentrations of sera two- to fourfold higher than these minimal
6
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26
VIR 5 r
160
G.
C. SCHILD
AND
OTHERS
Antigen
P1-Mh
O
P1-Sab
{ 43
" "
' -'
(~
o 0 :o ,
,
P2-MEF
0
0
0
6
©
O-
Fig. i. Specificity o f direct S R D reactions with poliovlrus antigens. I m m u n e rabbit sera (anti-V)
to poliovirus type I ( M a h o n e y and Sabin), type 2 ( M E F ) and type 3 (Sabin) were incorporated
into agarose gels at concentrations indicated in Table I. Concentrates o f poliovirus antigens o f
type I (Mahoney and Sabin), type 2 ( M E F and Sabin) and type 3 (30 and Sabin) were added to
wells. The reaction zones were stained with Coomassie blue. Zones were produced only by h o m o typic antigens and their intensity was greater for homologous strains than for heterologous strains.
Relationship between virus neutralization titre and
minimal concentration of antiserum required for SRD tests
T a b l e I.
Immune sera
Virus
neutralization
titre*
Minimal dilution
o f serum required
for S R D testst
Anti-V sera (rabbit)
PI-Sabin
3000
i : 150
Pi-Mahoney
18o00
l : 800
P2-Sabin
38000
l : 800
Pz-MEF
zooo
I : 70
P3-Sabin
2000
i : 90
P3-115
iooooo
1 : 2000
P3-3o
28000
I : 750
P3-Saukette
6000
I : 300
P3-I I9
I7o00
t :500
P3-19o
8500
I : 5oo
P3-3o
5000
I : 200
Anti-D serum (guinea-pig)
P3-3o
> 30000
l : 200
* Neutralization titre in microtitration assays with homologous poliovirus strain.
t Minimal final dilution o f antiserum in gel required to give clearly visible stained zones in S R D tests
with homologous or closely related poliovirus antigen.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26
Sensitive assay for poliovirus antigen
Dilution
1 t 5 virus 1
115 vkus 2
2:3
1" 1
t6t
1:2
1:4
1:8
@0-@@
@ O @ o
o.
Antiserum. anti-I 15
Fig. z. Direct S R D assay o f poliovirus type 3, strain 115, concentrates. Twenty/,1 vol. o f dilutions o f
two preparations (I and z) o f virus were added to 4 m m diam. wells in immunoplates containing
o'7/tl/ml o f rabbit antiserum (anti-V) to strain l I5. After allowing 5 days for diffusion, the
reaction zones were clearly visible after staining with Coomassie blue.
g-.
E
I
I
I
I
I
I
I
I
80
60
i
40
S
2O
0
1:16 1.8
I
I
I
I
1.4
t:2
Antigen dilution
3.4
1"1
Fig. 3- S R D dose-response curves for poliovirus type 3, strain 115, virus concentrates tested as
indicated for Fig. 2.
concentrations were used. The zones produced by the most potent antigens with infectivity
titres to 11'5 TCIDso/ml reached a maximum diam. of approx. Io mm after allowing 4 to 5
days for diffusion.
Studies on the specificity of the SRD reactions with anti-V sera (Fig. l) and anti-D sera
(not shown) indicated they were poliovirus type-specific; no intertypic reactions were
observed. Within a poliovirus type the intensities of the stained zones were greater for
homologous than for heterologous antigens, and for heterologous antigens multiple reaction
rings were sometimes apparent (Fig. i and 5).
Studies of the dose-response characteristics of direct SRD assays were performed with
several preparations of antisera and homotypic antigens. A representative assay is illustrated
in Fig. 2. With decreasing concentrations of antigen, reduction in zone size but not intensity
was observed. Linear dose-response curves (Fig. 3) were obtained when antigen dilution was
plotted against the square of zone diameter (d~). Curves for different preparations of an
antigen tested on the same immunoplate converged to a common intercept on the d ~ axis
as has been observed for similar assays of influenza haemagglutinin (Wood et al. I977).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26
6-2
16z
G. C. S C H I L D AND OTHERS
20
15
E
,~ 10
5
0
~' 125
x
100
75
~.
25
0
?
8
x
.= 4
"-- 2
0
I
1
5
5
~
(c)
,,-,~
~
~
~~
l
J
J
I
I
I
I
I
1
i
1
2
3
4
5~ 6
7
8
9
10
I
I
11 12
VP2
VP3
I
I
I
13
t4
15
Gradient fraction number
Fig. 4. Sucrose gradient fractionation of asS-labelled poliovirus P3-3o antigens (Minor et al. I98O)
from HEP2 cell harvests. Virus was centrifuged in a t5 to 45~ sucrose gradient in Io mM-tris-HCl
pH 7"4, 50 mM-NaCI at z5ooo rev/min for 4 h and fractions of 2 mI were collected by bottom
puncture. Undiluted gradient fractions were tested for (a) SRD direct activity on immunoplates
containing antiserum to poliovirus type 3 (anti-V) and the zone diameters measured, (b) infectivity
in HEP2 cell monolayers, (c) radioactivity counted in 5/~1 vol. and (d) polypeptide profiles by
PAGE on lo to 3o~ linear gradient SDS-polyacrylamide slab gels (Minor, I98O). The antigen
sedimenting at I55S (D-antigen) showed SRD activity, was infectious and contained the polypeptides associated with infectious virus. The 8oS fraction (C-antigen) did not give stainable
SRD reactions in the system used but was found to react in immunoplates containing high concentration of anti-C antibody (Minor et al. I98o).
Studies were p e r f o r m e d to d e t e r m i n e the nature o f the poliovirus antigen reacting in S R D
tests. F r a c t i o n s from sucrose density gradients on which harvests o f asS-labelled poliovirus
P3-3 o h a d been centrifuged ( M i n o r et al. 198o) were tested on S R D i m m u n o p l a t e s containing h o m o l o g o u s anti-V serum. The i m m u n o p l a t e s were examined after staining for
reaction zones. R a d i o a c t i v e counts a n d virus infectivity assays were also p e r f o r m e d on the
g r a d i e n t fractions. A peak o f S R D activity was detected which coincided with the ~55S
radioactive peak a n d with the p e a k o f infective virus (Fig. 4). N o S R D activity c o r r e s p o n d i n g
with the 8oS r a d i o a c t i v e p e a k was detected. T h e I55S and 8oS c o m p o n e n t s were further
characterized by analysis o f their p o l y p e p t i d e c o m p o s i t i o n s by p o l y a c r y l a m i d e gel electrophoresis ( P A G E ) (Fig. 4)- The patlerns o f p o l y p e p t i d e s detected were characteristic o f Dand C-antigen respectively ( M i n o r et aL" i 98o). It was concluded t h a t the S R D zones detected
on i m m u n o p l a t e s . c o n t a i n i n g anti-V o r a n t i - D sera were p r o d u c e d by antigeffs associated
with intact infections virus, i.e. D-antigen ( M a y e r et al. 1957; R o i z m a n e t al. ~957). This
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26
Sensitive assay for poliovirus antigen
163
was supported by the observation (data not shown) that antigens of all three types of poliovirus lost all activity in SRD tests with anti-V or anti-D sera after heating at 56 °C for
45 rain, a procedure known to convert D-antigen to C-antigen (Le Bouvier, ]955; Mayer
et aL I957).
Autoradiographic SRD tests: effect of addition of unlabelled poliovirus antigen on
autoradiographic zones produced by 35S-labelled antigen
As reported elsewhere (Minor et al. I98o) SRD reaction zones could be demonstrated by
autoradiography when zsS-labelled poliovirus was added to immunoplates containing
specific antisera. Studies were performed to determine the effect on autoradiographic SRD
zones when the labelled antigen was added to wells which contained unlabelled homotypic
poliovirus antigens. The mixtures were tested in immunoplates containing concentrations
of serum suitable for use in direct SRD tests so that stained and autoradiographic zones
could be detected for the same immunoplate.
The results of a study with several poliovirus type 3 antigens are illustrated in Fig. 5. The
immunoplates contained antiserum to potiovirus P3-II5 (Sabin vaccine-like). Unlabelled
concentrates o f ' wild' polioviruses P3-3o and P3-I9O, and vaccine-like strains P3-Sabin and
P3-I I5, were added to wells in the immunoplates followed by 3sS-labelled P3-Sabin Dantigen (homologous marker system) or 25S-labelled P3-3o D-antigen (heterologous marker
system). Comparisons were made of the size of the stained zones with the autoradiographic
zones detected on autoradiographs of the same plates. Identity of the stained and autoradiographic zones indicated co-migration of the unlabelled and labelled antigen, i.e. that
the unlabelled antigen had enhanced the size of the radioactive zone produced by the labelled
antigen (a positive ZE response). Where the size of an autoradiographic zone surrounding a
well which contained both reagents was the same as that produced by labelled virus alone,
this indicated that co-migration of the two antigens did not occur, i.e. the unlabelled antigen
had produced no ZE response. For the homologous marker system, only the unlabelled
viruses antigenically homologous with the serum and marker virus (P3-Sabin and P3-I15)
produced ZE activity whilst P3-3o and P3-~9o produced no ZE activity. In contrast, in the
heterologous system employing 3~S-labelled P3-3o D-antigen, each of the unlabelled viruses
produced a clear ZE response. In studies with the same unlabetled antigens and markers,
but employing immunoptates containing anti-V serum for P3-3o, the converse pattern of
reactions was observed. For the homologous (P3-3o D-antigen) marker system P3-3o and
P3-I9O antigens gave ZE responses, indicating their antigenic similarity, but P3-Sabin and
P3-~ 15 did not produce ZE activity. It was concluded from these investigations with other
antisera and a larger collection of poliovirus 3 strains (G. C. Schild, unpublished results)
that ZE reactions with polioviruses were narrowly strain-specific when a homologous
35S-labelled D-antigen marker was used and broadly reactive within a subtype when
heterologous D-antigen marker was used.
Dose-response characteristics and sensitivity of ZE assays
It was established from studies with poliovirus type 3 antigens that maximal sensitivity
for ZE assays, consistent with suitable dose-response characteristics, was achieved using
antiserum concentrations approx. I oo-fold lower than the minimal concentrations necessary
for direct SRD assays (Table I) and using concentrations of 35S-labelled, purified D-antigen
marker which contained 20o to 400 ct/min per ~o/~1. Fig. 6 illustrates ZE assays with serial
dilutions of several poliovirus type 3 antigens (cell culture fluid harvests and high dilutions,
from Io -2, of virus concentrates) performed on immunoplates containing anti-V serum to
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26
t64
G. C. SCHILD AND OTHERS
0
~o~
~._~ E._-~.~
~
©
~ E .o > o . 9 -
o
o
~
o
~
0
0
U~
..~
~0
z
@
o
r~
0
i
P3-I I5, and using a "heterologous marker system' 35S-labelled P3-3o D-antigen. All poliovirus type 3 antigens tested, including P3-Sauk (data not shown), produced ZE responses
whilst poliovirus type I and type 2 antigens gave no ZE response. The most potent cell
culture fluids could be diluted I : 64 before extinction of ZE activity was reached whilst the
most potent poliovirus type 3 concentrate produced ZE activity at a dilution of I:32oo.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26
Sensitive assay for pofiovirus antigen
165
Antigen dilution
1:1
1'2
1:4
1"8
t"16
1"32 1%4
(a)
P3.30to
C: 0
0
P3"3Ooonon-lO-2
"4
P3-115 (2)concn. 10 '~
Pl'Mhc°ncn'lO-2
-IF
I
I
220 _(b)
0
I
0
0
0
0 0 0 0
0
0
0
0
0
0
0
0
0
0
0
0
10-
<
0
.
.
.
.
I'"
/
/
200 -
o
180
160
140
o
[]
_
12o
"~
N
i00
80
60
40
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
20 I I
I
1"32 1 16 1-8
I
l 4
I
1
12
1.t
Antigen dilution
Fig. 6. Dose-response studies of ZE assays of poliovirus type 3 antigens. Twenty/d vol. of serially
diluted preparations of poliovirus P3-3o cell culture fluid (P3-3otc, © - - © ) ; strain 30 concentrate,
diluted io -~ (P3-3o concn. IO-~, 0 - - 0 ) ; IO-z dilutions of two strain 115 concentrates [P3-II5
O) concn, l o - L 5 - - D ; P3-II5 (2) concn. I o - L I I - - m ] and poliovirus strain Mahoney concentrate diluted io -~ (PI-Mh concn, io -~, - . . . . ) were added to wells in immunoplates containing rabbit antiserum (anti-V) to poliovirus P3-I15 (o.oI/tl/ml gel). This was followed by
addition of Io pl of 1:50 dilution of ssS-labelled strain 30 D-antigen marker. The autoradiographic ZE responses (a) and the resulting dose-response graph (b) are illustrated. Each type 3
preparation gave a linear dose-response with all dose-responses meeting at a common intercept
on the d z axis. Type I virus showed no ZE response.
The diameters of the autoradiographic zones were measured and graphs plotted of d ~ against
antigen dilution (Fig. 6b). Each poliovirus type 3 antigen showed a linear dose-response and
the response lines for different antigen preparations converged to a common intercept on the
d * ordinate. Thus ZE potency assays could be analysed in a similar manner to direct SRD
assays for which the slopes of the dose-response curves are proportional to the antigen concentration (Schild et al. I975; Wood et al. I977). For similar ZE assays but using homologous D-antigen as marker (P3-I I5), only Sabin vaccine-like strains gave ZE responses.
The relative sensitivities of direct SRD and ZE assays for several poliovirus type 3
preparations are shown in Table 2. The viruses were titrated for infectivity and were assayed
for antigen by direct SRD and ZE tests in comparison with a commercial poliovirus type 3
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26
Virus concentrate
I I 5 strain
Cell culture fluid
30 strain
Antigen preparation
D - a n t i g e n reference
9"5
l I.O
Assay
system
employed
Direct S R D
ZE
Direct S R D
ZE
Direct S R D
ZE
Antiserum
dilution used
in assay*
5 × lO -4
3 × IO-~'
5 x io -4
3 x to -~
5 x io -4
3 x lO -6
Limiting dilution
at which significant
antigenic activity
detectedt
1:4
I : 16o
1:64
1:64oo
< I : I§
1:256
A n t i g e n concn.
in D units at
limiting
dilution;~
ioo
2'5
I4o
3'5
3'5
Relative sensitivities of direct SRD and ZE assays for poliovirus type 3 antigens
Infectivity
titre
(log TCIDs0/ml)
Inactivated virus
T a b l e 2.
Infectivity titre at
limiting dilution
(lOgl0 TCIDs0/ml)
9"2
7'2
7' i
* R a b b i t s e r u m (anti-V) to poliovirus P3-115 (Sabin vaccine-like strain), h o m o l o g o u s V N titre o f 1:80000.
I" Highest antigen dilution s h o w i n g significant activity in direct S R D (stained zones) or Z E (autoradiographic) tests.
D - a n t i g e n units determined by c o m p a r i s o n with poliovirus type 3 (Saukette) reference antigen, with a n assigned potency of 4oo D - a n t i g e n u n i t s / m l (Glaxo Laboratories,
G r e e n f o r d , U.K.).
§ < I : 1 n o detectable activity with u n d i l u t e d antigen.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26
09
rn
0
;>
Z
u
~3
O~
Sensitive assay f o r poliovirus antigen
I(a)
I
I
I67
I
155s
90
70
50
30
E
t
140 -(b)
120
t
,oo
80
60
40
20
l
l
i
5
10
Fraction number
t
15
Fig. 7. An unlabelled harvest of HEPz cells infected with poliovirus type 3, strain 3o was centrifuged in a 15 to 45 ~ sucrose density gradient (25 ooo rev/min for 4 b) and 2 ml fractions collected by
bottom puncture. The fractions were tested for ZE activity in (a) employing ~S-labelled, strain
3o, D-antigen as marker in immunoplates containing homologous immune rabbit serum (anti-V)
and (b) employing 35S-labelled, strain 30, C-antigen as marker in immunoplates containing specific
anti-C rabbit serum. The figure shows the distribution of ZE activity in the gradient in each of
the immunoplates.
reference antigen (arbitrary potency of 4oo D-antigen units/ml; Beale & Mason, I962). For
the purposes of these studies a direct SRD or ZE response was judged to be significant when d ~
was greater than 2 mm 2above the common intercept on the d 2axis of the dose-response graph.
ZE assays were shown to be 4o- to Ioo-fold more sensitive than direct SRD tests. The most
potent cell culture fluid harvests, which had no detectable activity in direct SRD tests, had
limiting ZE activity at a 1:64 to I : I28 dilution. The limits of sensitivity in terms of infectious
virus, for detection of D-antigen corresponded to approx, ro 9 TCIDs0/ml for direct SRD
assays, approx. Io 7 YCIDso/ml (io 4'3 TCIDso/ZO/d) for ZE assays.
D- and C-antigen specificity of Z E assays
To determine the nature of the poliovirus antigen reactive in ZE tests, unlabelled harvests
of HEP2 cell cultures infected with poliovirus type 3 were centrifuged in sucrose gradients
and gradient fractions were examined for ZE activity on immunoplates containing homologous anti-V serum with *sS-labelled, purified D-antigen as marker and in immunoplates
containing anti-C serum for poliovirus 3 with 35S-labelled, purified C-antigen as marker.
The distribution of ZE activity in the gradient for the 'D-antigen' and the 'C-antigen'
immunoplates is shown in Fig. 7. Only a single peak of ZE activity was detected on each
immunoplate. The D-antigen immunoplate detected a band of ZE activity with a sedimentation coefficient of approx. I55S whilst the C-antigen immunoplate detected a band of
activity at approx. 8oS. These findings provided evidence that the ZE tests with anti-V serum
(or anti-D serum, data not shown) using D-antigen as the radioactive marker were Dantigen specific. This finding was supported by the fact that ZE activity in this system was
not detected when the assay antigen was heated at 56 °C for 45 rain to denature D-antigen.
The 'C-antigen system' detected C- but not D-antigen as evidenced by the sedimentation
characteristics of the peak of ZE activity detected in this system and the finding that the ZE
activity was stable on heating at 56 °C for 45 min.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26
I68
G . C. S C H I L D
1
AND OTHERS
I
I
J
2-6
~
2.4
'~ 2-2
r',
o 2.0-
~
•
-
1.8
~ 1-6
"~ 1.4
~ 1.2
1.0
I
I
6.0
7.0
8-0
9.0
Virus infectivity titre (log TCIDso )
Fig. 8. Relationship between infectivity titre (TCIDso/ml) and D-antigen activity detected in ZE
assays for poliovirus type ~ preparations. Poliovirus type I strain 3o, concentrate diluted I o-2 ([]),
strain Mahoney cell culture fluid (A), strain Mahoney concentrate (a) diluted 2× ~o -1 ( I ) ,
concentrate (b) diluted 2 × IO -1 ( I ) , and Sabin strain, cell culture fluid ((3) were assayed for ZE
activity using rabbit antiserum to poliovirus I (Mahoney) and 35S-labelled type ~, strain 1/67
D-antigen as marker. ZE activity was expressed in D-antigen units by comparison with the ZE
activity of type I D-antigen reference preparation (88o D-antigen units/ml) supplied by Glaxo
Laboratories. Virus infectivity assays were performed in HEP2 cell cultures.
Z E assays with poliovirus type I antigens
Studies with poliovirus type I antigens employing ZE techniques produced similar results
in terms of antigenic specificity and dose-response characteristics, to those described for
type 3 viruses. Several preparations of different type I virus strains of known infectivity were
examined for ZE antigen activity which was measured in relation to a poliovirus type I
D-antigen reference preparation (arbitrary potency of 88o ' D' units/ml). The results indicate
a good correlation between virus infectivity (log TCIDso) and ZE activity (log D units/ml)
(Fig. 8). Virus preparations with infectivity titres of IOv'9 TCIDs0/ml (approx. IOO D units/
ml) and above showed good ZE reponses, whereas preparations with titres (TCIDs0 IO6"~)
gave minimal ZE activity. The extinction end-point for poliovirus type I D-antigens was
approx. IO6 TCIDs0/ml (io 3'3 TCID~0/2o pl) and corresponded to about 5 D-antigen units.
DISCUSSION
A disadvantage of SRD assays for virus antigens, which are often present in low concentration in biological materials, is their relative insensitivity when reaction zones are
detected by protein staining. For influenza haemagglutinin, the limit of sensitivity is I ~tg
antigen/ml (Wood et al. I977). The studies described here indicated that intact poliovirus
particles diffused well in agarose gels and gave well-defined SRD reactions which were typespecific and had satisfactory dose-response characteristics. The tests were, however, of
limited use being insufficiently sensitive to detect antigen in non-concentrated virus preparations. In the studies using immunoplates containing immune (anti-V) sera to unfractionated poliovirus antigen or with specific anti-D sera, it was shown that the SRD
reactions corresponded to D-antigen. Direct SRD studies using potent specific anti-C sera
(G. C. Schild, unpublished observations) have shown that only when very high concentrations of anti-C antibody are used in immunoplates are stainable zones corresponding to
C-antigen detectable. The lack of such reactions with anti-V sera may thus be related to their
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26
Sensitive assay for poliovirus antigen
169
relatively low content of antibody to C antigen (G. C. Schild, unpublished results). Sucrose
density gradient analysis reported here and in the accompanying paper (Minor et al. I98o)
failed to demonstrate antigenic activity (SRD or ZE) in the slow sedimenting region of the
gradient. This is in contrast with previous reports of an antigenic I4S component in poliovirus preparations (Ghendon & Yakobson, x97I).
The ZE assay system for poliovirus antigens described here appears to offer considerable
potential for the sensitive quantification of poliovirus D antigens, applicable to concentrations of antigen present in conventional cell culture fluids and inactivated poliovirus
vaccines. It is also of potential value as a means of fine antigenic characterization of polioviruses. The narrow strain specificity of the homologous ZE system for poliovirus D
antigens demonstrated in the present studies is in contrast to the broad type specificity
exhibited by direct SRD tests with poliovirus D-antigens and the broad subtype specificity
exhibited by such tests for influenza haemagglutinin antigens.
The ZE assay depends essentially on competition for antibody in the agarose gel between
the unlabelled and radioactively labelled poliovirus antigen and upon the availability in the
gel of antibody molecules capable of binding with and halting the diffusion of the labelled
antigen. When it is antigenically homologous with the antiserum and the labelled antigen
(homologous marker system) the unlabelled antigen is capable of reacting with antibody
molecules in the gel which are specific for the labelled antigen thus effectively reducing their
concentration. Under these conditions it is shown that the labelled and unlabelled antigen
co-migrate in the gel since the autoradiographic and stained SRD zones are identical in size
(Fig. 5)- The degree of enhancement of autoradiographic zone size, compared to zones
surrounding wells containing labelled antigen only, is proportional to the amount of unlabelled antigen.
When the labelled virus and antiserum are antigenically homologous and the unlabelled
(homotypic) virus is antigenically heterologous, the unlabelled virus would be expected to
react with only a proportion of antibody molecules in the gel, leaving other strain-specific
molecules to react with the labelled antigen so arresting its diffusion. This is consistent with
the finding that in the homologous marker system only homologous unlabelled antigens
give ZE responses.
For ZE systems in which the labelled virus and antiserum are antigenically heterologous
all homotypic poliovirus D-antigens produce ZE responses. An explanation for this finding
is that in the heterologous system only antibody molecules of cross-reactive specificity
within a poliovirus type are involved. Such cross-reactive antibody is depleted by binding
with all poliovirus D-antigens within a type. This suggested mechanism for the ZE test is
supported by studies described in the present paper and by studies of influenza haemagglutinin using similar ZE techniques (Wood et al. I98O).
An interesting feature of the ZE assay for poliovirus antigens is that, under appropriate
conditions, it may be adapted for the detection of either the D- or the C-antigens of polioviruses. However, detailed information on the sensitivity or strain specificity of the ZE
assay for poliovirus C antigen is not yet available.
An important potential application of ZE assays is the standardization of potency of
inactivated polio vaccines for use in man. Immunological potency of polio vaccines is
related to D-antigen content which has been evaluated for vaccine concentrates by quantitative immunodiffusion techniques (Le Bouvier, I959, Beale & Mason, I96e). However, the
lack of sensitivity of these methods has precluded their use for final vaccines which are
routinely evaluated by assessment of antibody responses in experimental animats (Gard et al.
I956). Studies are in progress in our laboratory to evaluate the use of ZE assays for poliovaccine standardization and satisfactory assays have been performed on D-antigens present
in conventional formalin-inactivated vaccines.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26
I70
G.C.
SCHILD
AND
OTHERS
A special feature of the poliovirus ZE system is its high degree of specificity which is
largely independent of the properties of the antibody employed but directed by the marker
antigen. This is not the case for other aptigen assay systems such as SRD, radioimmunoassay
and enzyme-linked immunosorbence where the assay is dependent upon the specificity of
the antibody employed and which generally shows broad antigenic reactivity.
The authors wish to thank Ms Una Dunleavy, Ms Ann John and Ms Janet Bootman for
their able technical assistance.
REFERENCES
BEALE, A. J. & MASON, P. J. (1962). T h e m e a s u r e m e n t o f the D - a n t i g e n in poliovirus preparations. Journal
of Hygiene 60, I I3-12o.
COWAN, K. M. & WAGNER, G. G. (1970). I m m u n o c h e m i c a l studies of foot a n d m o u t h disease. VIII. Detection
a n d q u a n t i t a t i o n of antibodies by radial i m m u n o d i f f u s i o n . Journal of Immunology Io5, 557-566.
D~SM(SK, L & MAGRATH, D. L (1979). Poliovirus isolation a n d serological techniques. W o r l d Health Organization publication, no. 46.
ESPOSlTO, J. J. (1976). Detection of poliovirus antigens a n d antibodies: micro-indirect h a e m a g g l u t i n a t i o n
a n d h a e m a g g l u t i n a t i o n inhibition tests for poliovirus types I, 2 a n d 3. Microbios ~6, 29-36.
GARD, S., WESSLEN, T. EAGRAEUS, A., SVEDMYR, A. & OLIN, G. (I956). T h e use o f guinea pigs in tests for
i m m u n o g e n i c capacity o f poliomyelitis virus preparations. Archiv .fiir die Gesamte Virusforschung 6,
4 o i - 4 I I.
GHENDON, Y. Z. & YAKOBSON, E. A. (1971). Antigenic specificity of poliovirus - related particles. Journal o f
Virology 8, 589-59o.
LE BOUVlER, G. L. (I955). T h e modification o f poliovirus antigens by heat a n d ultraviolet light. Lancet ii,
IOI3-Ioi6.
LE BOUVlER, G. L. (1959). Poliovirus D a n d C a n t i g e n s - their differentiation a n d m e a s u r e m e n t by precipitin in agar British Journal of Experimental Pathology 4 o, 452-46o.
MANCIN1, G., CARBONARA, O. A. & HEREMANS, J. V. (1965). I m m u n o c h e m i c a l q u a n t i t a t i o n of antigens by
single-radial-immunodiffusion, lmmunochemistry z, 235-254.
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). 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.
MINOR, P. D. (1980). C o m p a r a t i v e biochemical studies o f type 3 poliovirus. Journal of Virology 34, 73-84.
MINOR, P. D., SCHILD, G. C., WOOD, J. M. & DANDAWATE, C. N. (I980). T h e p r e p a r a t i o n o f specific i m m u n e
sera against type 3 poliovirus D - a n t i g e n a n d C-antigen a n d the d e m o n s t r a t i o n of two C-antigen
c o m p o n e n t s in vaccine strain populations. Journal of General Virology $x, 147-I56.
PEREIRA, H. G., MACHADO, R. D. & SCHILD, G. C. (I972). Study o f adenovirus h e x o n antigen-antibody reactions by single-radial-diffusion techniques. Journal of Immunological Methods z, I 2 I - I 2 8 .
ROIZMAN, B., r~OPKEN, W. & MAYER, M. M. (1957). I m m u n o c h e m i c a l studies o f poliovirus. II. Kinetics of
f o r m a t i o n of infectious a n d non-infectious type I poliovirus in three cell strains o f h u m a n derivation.
Journal of lmmunology 8o, 386-395.
ROIZMAN, B., MAYER, M. M. & RAPP, r~. (I958). I m m u n o c h e m i c a l studies o f polioviruses. VI. F u r t h e r studies
o n the i m m u n o l o g i c a n d physical properties o f poliovirus particles p r o d u c e d in cell culture. Journal
of Immunology 81, 419-425.
RtJMKE, Pn. & TnUNG, P. J. (1964). I m m u n o l o g i c a l studies o n the sex-dependent p r e a l b u m i n in m o u s e urine
a n d its occurrence in the serum. Acta Endocrinologica 47, 156-164 •
SCribED, G. C., WOOD, J. M. & NEWMAN, R. W. (1975)- A single-radial-immunodiffusion technique for the
a s s a y o f influenza h a e m a g g l u t i n i n antigen. Bulletin of the Worm Health Organization 52, 223-230.
VAN WEZEL, A. L. & rIAZENDONK, A. G. (1979). Intratypic serodifferentiation of poliomyelitis virus strains
by specific antisera. Intervirology xx, 2-8.
WOOD, J. M., SCHILD, G. C., NEWMAN, R. W. & SEAGROATT, V. (1977). A single-radial-immunodiffusion techn i q u e for the assay of influenza haemagglutinin. Bulletin of the World Health Organization 52, 223-230.
WOOD, J. M., SCHILD, G. C. & NEWMAN, R. W. (I980). A sensitive, single-radial-diffusion a u t o r a d i o g r a p h i c
zone size e n h a n c e m e n t technique for the assay o f influenza haemagglutinin. Journal of General
Virology 47, 355-363.
(Received 6 May I98o)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 12:37:26