J. gen. ViroL
(I972), x6, 231-236
23I
Printed in Great Britain
A Quantitative, Single-radial-diffusion Test for Immunological
Studies with Influenza Virus
(Accepted lO M a y 1972)
Antibodies for influenza haemagglutinin and neuraminidase are assayed routinely in
haemagglutination-inhibition (Hirst, I942) and neuraminidase-inhibition (Ada, Lind &
Laver, 1963) tests which involve reaction systems of three components. More recently,
influenza antibodies have been detected by gel immuno-double-diffusion tests with influenza
viruses disrupted by detergents (Styk & Hanna, I966; Schild & Pereira, I969). Although the
latter method has the advantage that it detects antibodies for each of the major antigens of
the virus in a single test system (Schild et aL 1972a; Schild, I972 ) the quantitation of antibody is not readily achieved. In the present report we describe a quantitative, single-radialdiffusion test which is a sensitive, rapid and convenient method for the estimation of antibodies to the haemagglutinin and neuraminidase antigens of the influenza virus. A unique
feature of this test is that only one component, the antibody, diffuses during the test; the
other component, the antigen in the form of intact influenza virus particles, does not diffuse.
The method does not depend on the formation of immune precipitates. In contrast, in the
radial-immunodiffusion method described by Mancini, Carbonara & Heremans (1965)
for the assay of serum protein antigens, both antibody and antigen are in diffusible form. The
test we describe has many potential applications in immunological, diagnostic and epidemiological studies with influenza and other viruses.
For our own studies, influenza viruses were cultivated in embryonated eggs and purified
and concentrated using a continuous-flow centrifuge (Skehel & Schild, 197I), or by conventional methods on sucrose density gradients (Schild & Pereira, 1969). Virus concentrates
contained approximately 4 x I o 6 haemagglutinin units (H.A.U.)/ml. and lO to 15 mg. protein•
ml. Before use in single-radial-diffusion tests, virus concentrates were treated by ultrasound
(Burndept Ultrasonic cleaning bath, 8o kHz) for 3 to 5 sec. to break up aggregates of virus.
This did not disrupt virus particles.
Suspensions of purified, intact influenza virus were prepared in agarose gels. An agarose
with a low (37 °) setting temperature (Indubiose A-37 from Hndustrie Biologique Fran~aise)
was employed routinely and avoided the exposure of the virus antigen to the relatively
higher temperatures (45°+) necessary to maintain other types of agar or agarose in a molten
state. Agarose gels (1"5 %, w[v, Indubiose A-37 in o'I5 M-NaCI+o'oI M-phosphate buffer,
final pH 7"I + o . 1 % sodium azide) were me/ted at IOO° and cooled to 420 before the addition
of virus to a final concentration of 4 x 104 H.A.U./ml. (o-x rag. protein/ml.). The agarose and
virus were mixed by vigorous shaking and uniform, I'5 mm., layers of gel were made as
described by Mancini et al. (I965) using photographic glass-plates. Alternatively, plastic
immunoplates (Hyland Laboratories, California, U.S.A.) with 2"5 ml. agarose per plate were
used to give a gel thickness of 2"5 mm. Circular wells (2-o mm. diameter) were cut in the
agarose to accommodate the antisera. Volumes of 3/zl. of sera, diluted where appropriate,
were added to the wells and the plates kept in a moist chamber at room temperature.
The presence in test sera of antibodies to the haemagglutinin or neuraminidase of the virus
was detected by the appearance of zones of opalescence surrounding the wells. The zones
were readily visible within 3 to 4 hr after setting up the tests and reached maximum size at
7 to 8 hr. For convenience, tests were read after leaving overnight for I6 to I8 hr. Accurate
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 07:37:34
232
Short Communications
Figs. i, 2. Single-radial-diffusion reactions of anti-haemagglutinin and anti-neuraminidase sera in
1.5 % A-37 agarose gels containing a suspension of 4 x io 4H.A.U./ml. of purified, intact A/HON~
KONG/I/68 (I-I3 N2) virus (X-31 strain). Volumes of 3/~1. of antisera were added to the wells.
Fig. I. Reactions of antisera to purified H 3 haemagglutinin and purified N 2 neuraminidase. Wells
I, 2, 4, 8 and I ~ contain samples of different rabbit sera against H 3 haemagglutinin, and wells 5, 13
and 14 contain rabbit sera against purified N2 neuraminidase. Well Io contains mixtures of
anti-H3 and anti-N2 sera and wells 3, 6, 7 and 9 contain rabbit antisera against purified A/HON6
KONC/I/68virus (anti H 3 + anti N 2). Well 12 contains potent antiserum to influenza A ribonucleoprotein. Antisera against haemagglutinin gave intense opalescent zones with distinct margins.
Anti-neuraminidase sera gave less intense zones with diffuse margins. Zones with intense central
areas and less dense haloes were observed when sera contained both antibodies.
measurements of zone diameters were made under dark-ground illumination using an eyepiece micrometer scale ( x 8 magnification, Matchless Machines Ltd., U.K.) or directly
from photographic enlargements ( x 3 magnification). Once formed, the zones were stable
for several weeks and their diameters were constant in size when the plates were stored by
immersion for several weeks in phosphate buffered saline.
Fig. I shows the reactions of various antisera in agarose gels containing X-3I virus
(Kilbourne et al. I 9 7 0 , a high-yielding recombinant influenza A virus antigenically identical
to A/HOYG KONG/I/68 (H3Na) (WHO, I970- Rabbit antiserum prepared against purified
H 3 haemagglutinin derived from A/noNe KONG/I/68 (H3 N2) virus (Schild et al. 1972 b) gave
intense and sharply defined zones of opalescence, up to ~o ram. diameter. In contrast, antibody prepared against purified N a neuraminidase derived from A/noN6 KON(;/I]68 virus
(Schild, in preparation) gave less intense zones of up to I3 mm. diameter with less well
defined margins. The difference in appearance of the zones produced by anti-neuraminidase
and anti-haemagglutinin antibodies was consistently observed with a number of different
strains of influenza virus and with different sources of antisera (rabbit, guinea pig, mouse,
ferret). Rabbit antisera against preparations of whole influenza virus (A/MoNa Konc/i/68)
which contained antibodies to both surface antigens (H 3 and N2) of the virus gave zones
with a dense central area (anti-haemagglutinin antibody) and a less dense ' h a l o ' (antineuraminidase antibody). Zones of similar appearance appeared when mixtures of antihaemagglutinin and anti-neuraminidase sera were used (see Fig. 0.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 07:37:34
Short Communications
233
Fig. 2. Effect o n zone size o f dilution of a n t i - h a e m a g g l u t i n i n s e r u m . Wells x, 2, 3, 4, 5,r6 a n d 7
c o n t a i n I ; i, I "2, I • 4, 1 "8, I : I6, I "32 a n d I : 64 dilutions o f r a b b i t a n t i s e r u m against purlfled H 3
h a e m a g g l u t i n i n ( H I titre ~ : 15oo).
Potent antiserum to influenza A ribonucleoprotein antigen (Schild & Pereira, I969) or to
influenza A internal matrix protein antigen, a recently characterized, type-specific internal
antigen of the influenza virus (Schild, I972), failed to give visible zones in gels containing
intact X-3I virus. In addition, sera known to contain non-specific inhibitors of haemagglutination for influenza X-31 virus failed to give opalescent zones: these tests included the
sera of man, rabbit, horse, guinea pig, swine, mouse and chicken.
In the present studies with intact influenza virus particles only antibodies to the virus
envelope proteins gave opalescent zones. However, in gels containing detergent-disrupted
influenza virus, antibodies to influenza A ribonucleoprotein and internal matrix protein
(Schild, I972) gave opalescent zones which allowed the quantitation of these antibodies
(Schild, in preparation).
To determine the relationship between zone size and antibody concentration, a series of
dilutions of anti-haemagglutinin and anti-neuraminidase sera were tested against X-3 ~ virus.
A representative experiment with dilutions of anti-haemagglutinin (H 3) serum is shown in
Fig. 2. The linear relationship between zone area and antibody concentration is shown by the
plot of annulus area of the opalescent zone (omitting the diameter of the well) against
relative antibody concentration (Fig. 3). Thus zone areas give a direct measure of the relative
concentrations of either anti-haemagglutinin or anti-neuraminidase antibodies.
The relationship between the levels of antibody to influenza envelope antigens measured
by zone area and by conventional techniques (haemagglutination-inhibition (HI), virus
neutralization and neuraminidase-inhibition) is under investigation. However, in tests with
immune rabbit and guinea-pig sera, the single-radial-diffusion test was slightly less sensitive
than the HI test (WHO Expert Committee on Influenza, I953). Sera with HI titres of S0-gave
small but distinct zones whilst sera with lower HI titres of-~- or less failed to give visible
reactions. For anti-neuraminidase (rabbit) sera the radial-diffusion methods was about as
sensitive as neuraminidase-inhibition tests (Schild & Newman, 1969 a). Sera with NI 5o titres
of ~o- gave small zones of 2-I mm. diameter.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 07:37:34
Short Communications
234
50
40 .
°i°0I
100
75
50
25 12.5
Relative antibody concentration
0
Fig. 3. Dose-response curves for anti-haemagglutinin and anti-neuraminidase serum. The plots of
annulus area (excluding area of the well) against relative antibody concentration give a straight line
for each type of antiserum. 0 - - 0 ,
Rabbit antiserum in well tested against purified N2 neuraminidase (NI5o titre 1:3oo) in gel; •
• , rabbit antiserum in well tested against purified
H3 haemagglutinin (HI titre I : Iooo) in gel.
Table I. Specificity of single-radial-diffusion reactions of
anti-haemagglutinin and anti-neuraminidase sera
Immune rabbit sera in wells
r
Strain of intact virus in gels
(4 x ~o~ H.A.U./ml.)*
A/BEL/42 (HoNI)
A/SINGAPORE/I/5 7 (H2 N2)
A/BEL/42(Ho)-A/SlNG/57(N 2)
X-31 (H3 N2)
Anti-BEL
Anti-SING Anti-HONGKONG Anti-SING
haemagglutinin haemagglutinin haemagglatinin neuraminidase
(Ho)
( H 2)
( U 3)
( N 2)
5'9ti
< 2"0
5"9 i
< 2.o
< 2.0
4'8 i
< 2.o
2.2 d
< 2"0
2'2 d
< z-o
4"9 i
< 2.0
6"5 d
6"4 d
3"8 d
* The origin and character of these recombinants have been described (Schild et al. 1970; Kilbourne et al.
I97I).
t Zone diameter (ram). i, Intense zone with sharp margins; d, Relatively diffuse zone. Sera giving no
reactions are recorded as < 2.0 (diameter of well).
Statistical analyses of the results of dilutions anti-H3 and anti-N 2 sera multiply tested
by single-radial-diffusion on immunoplates containing X-3I virus have indicated that
differences in relative antibody concentrations as small as 15 to 2o % could be detected
and were of statistical significance. With conventional methods of detection of antibody to
influenza virus (HI, NI and CF tests) only differences in titre of 4oo% or greater (4-fold)
are regarded as significant.
In studies with paired human sera from individuals with clinical A/HONG KONG/I/68
(H3N2) infections, the single-radial-diffusion tests appeared as sensitive as conventional
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 07:37:34
Short Communications
235
serological diagnostic tests for influenza infection (complement fixation or haemagglutination-inhibition tests) and the results were unambiguous. In samples of sera from 70 individuals
no zones were detected in tests with X-3 t virus on the' acute' serum samples collected during
the first one or two days of illness. Zones ranging from 2-3 ram. diameter up to Io ram.
diameter were detected in convalescent sera taken :o to 14 days later.
The specificity of single-radial-diffusion reactions was examined with a number of
influenza A virus strains (Table t). Antiserum against the purified haemagglutinin, H o
(Schild, ~97o), of A[BEL[42 ( H o N I ) virus reacted equally well with BEL and with a recombinant virus A/~EL/42(Ho)-A/sINo/57(N2) known to contain H o haemagglutinin and
N2 neuraminidase. This antiserum gave no reaction with A[StNG/57(H2N2) parent virus
or with X-3:(H3N2). Antiserum against SINGAPORE neuraminidase (N2) reacted equally
well with A/SINGAPORE and with the BEL(Ho)-sING(Nz) recombinant; it a/so gave a reaction
with X-3I(H3N2 ) virus which contains neuraminidase of the same subtype but gave no
reaction with A/BEL/4z(HoNI ). As in the other studies, the zones produced by antiserum to
neuraminidase were relatively less intense and had less distinct margins than those produced
by anti-haemagglutinin sera. The single-radial-diffusion tests in general indicated a lack of
cross-reaction between different haemagglutinins (Ho with H z or H3) or neuraminidase
(N I and N 2) subtypes. The significance of the minor cross-reactions between H z and H 3 is
not clear. The ability of radial-diffusion tests to distinguish between variants within a given
H or N subtype is under investigation.
In an attempt to elucidate the mechanism of the single-radial-diffusion test, electron
micrographs of thin sections ofagarose gels containing 4 x ~o4 H.A.U./ml. of X- 3 ~ virus were
examined. These indicated that the virus was present in the gel as single particles or as small
clumps of up to eight particles. Intact particles or clumps were separated by at least 5o times
the particle diameter (5o00 nm.). Sections through zones produced by anti-haemagglutinin
or anti-neuraminidase antibody showed virus particles with haloes of antibody but did not
indicate any change in the distribution of virus particles. It seems therefore that the test
depends upon an increase in the light scattered by virus after primary attachment of antibody
and is independent of secondary agglutination or precipitation. This conclusion is supported
by the finding (Schild & Pereira, I969) that intact influenza virus particles do not diffuse in
gels of the present type.
The single-radial-diffusion test appears to have considerable advantages over conventional tests for antibodies to influenza envelope proteins. The test is simple to perform and
is not susceptible to the effects of non-specific inhibitors. Very small volumes of antisera are
required for the test and assays of antibody have been made with small samples of whole
blood obtained from finger pricks. This has obvious practical advantages for routine diagnostic and epidemiological studies since venepuncture may be avoided. In addition, by the
choice of appropriate antigenic hybrid influenza viruses, antibody reacting with a particular
envelope antigen may be assayed separately. In studies employing a recombinant influenza
virus containing A/HONG KON(~/I/68 neuraminidase (N2) and haemagglutinin (Have)
derived from fowl plague virus (Schild & Newman, I969 b) antibody to A2/HON(; KONG
neuraminidase in human and animal sera were assayed independently of anti-haemagglutinin antibody. Using the reverse type of recombinant containing A/HONG KONG/68
haemagglutinin (H3), antibody to haemagglutinin may be assayed. Techniques previously
available have not enabled the independent assay of haemagglutinin and neuraminidase
antibodies in a single system.
Since agarose gels containing X- 31 virus have been stored at 4 ° for up to 6 months before
use and have given reproducible results, and since the opalescent zones once formed
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 07:37:34
236
Short Communications
are stable f o r several weeks, the s i n g l e - r a d i a l - d i f f u s i o n test seems to be well suited to field
studies o n influenza. I n a p i l o t study, i m m u n o p l a t e s c o n t a i n i n g X - 3 I virus w e r e d i s t r i b u t e d
by m a i l to l a b o r a t o r i e s in H o n g K o n g , J o h a n n e s b u r g a n d T h e G a m b i a f o r t h e a d d i t i o n o f
p a i r e d h u m a n sera f r o m i n d i v i d u a l s w i t h H o n g K o n g influenza. O n r e t u r n o f the plates to
this l a b o r a t o r y 6 w e e k s later t h e r e a c t i o n s o f t h e test sera w e r e r e a d easily a n d s h o w e d n o
d e l e t e r i o u s effects d u e to t r a n s p o r t a t i o n . A c o l l a b o r a t i v e s u r v e y is in p r o g r e s s in w h i c h the
s i n g l e - r a d i a l - d i f f u s i o n test will be u s e d for s e r o l o g i c a l e p i d e m i o l o g i c a l studies o f i n f l u e n z a
in a n u m b e r o f a r e a s o f the w o r l d .
W e w i s h to t h a n k M r s S u s a n P i d d i n g t o n a n d M r R. N e w m a n f o r e x c e l l e n t t e c h n i c a l
assistance. M r D . H a r v e y k i n d l y c a r r i e d o u t the c o n t i n u o u s - f l o w u l t r a c e n t r i f u g a t i o n a n d
D r M . V. N e r m u t the e l e c t r o n m i c r o s c o p y . D r W. K . C h a n g , M i s s R. H a r w i n a n d M r K .
W i l l i a m s k i n d l y c o l l a b o r a t e d in the studies w i t h t r a n s p o r t a b l e i m m u n o p l a t e s .
W H O Worm Influenza Centre
National Institute f o r Medical Research
Mill Hill, London N W 7 I A A
England
G . C. SCHILD,
MICHELE HENRY-AYMARD,
H. G. PEREIRA
REFERENCES
ADA, G. L, LIND, P. E. & LAVER,W. ~. (I963). An immunological study of avian, viral and bacterial neuraminidase based on specific inhibition of enzyme by antibody. Journal of General Microbiology 32, 225.
HIRST, G. K. (t942). The quantitative determination of influenza virus and antibodies by means of red cell
agglutination. Journal of Experimental Medicine 75, 47.
ICILBOLmNE,E. D., SCHULMAN,J. L., SCHILD, O. C., SWANSON,J. & BUCKER,D. (I971). Correlated studies of a
recombinant influenza virus vaccine: i. derivation and characterization of the virus. Journal oflnJectious
Diseases 124, 449.
MANCINI, G., CARBONARA, A.O. & HEREMANS, J.F. (I965). Immunochemical quantitation of antigens by single
radial immunodiffusion. Immunoehemistry 2, 235.
SCHILD,O. C. 097O). Studies with antibody to the purified haemagglutinin of an influenza A o virus. Journal
of General Virology 9, I9I.
SCHILD, G. C. (I972). Evidence for a new, type-specific internal antigen of the influenza virus. Journal of
General Virology IS, 99.
SCHILD, G. C., MCCOHAN, D. & KENDAL, A. P. 097o). Immunological and structural studies on a recombinant
influenza A virus. In The Biology of Large RNA Viruses. Ed. by R. D. Barry and B. W. J. Mahy, p. 638.
London: Academic Press.
SCHILD, G. C. & NEWMAN,R. W. (19692). Antibody against influenza A2 virus neuraminidase in human sera.
Journal of Hygiene Cambridge 67, 353.
SCHILD, G. C. & NEWMAN,R. W. (I969b). Immunological relationships between the neuraminidases of human
and aiximal influenza viruses. Bulletin of the World Health Organization 4x, 437SCHILD, G. C. & PEREIRA, H.G. 0969). Characterization of the ribonucleoprotein and neuraminidase of
influenza A virus by immunodiffusion. Journal of Virology 4, 355.
SCHILD, G. C., WINTERS, W. D. & BRAND, C. M. (1972 a). Serological diagnosis of human influenza infections by
immuno-double-diffusion techniques. Bulletin of the World Health Organization 45,465.
SCHILD, G. C., BRAND, C. M., HARKNESS, J. W. & LAMONT, P. (I972b). Studies on relationships between porcine
and human influenza. I I. Immunological comparisons of human Aa/Hong Kong/68 virus with influenza
A viruses of porcine origin. Bulletin of the Worm Health Organization (in the Press).
SKEHEL, J. J. & SCHILD, G. C. (I97I). The polypeptide composition of influenza A viruses. Virology 44, 396.
STYK, B. & HANA,L. 0966). Immunodiffusion studies on the reaction of influenza virus with specific antibody
and non-specific serum beta inhibitor. Acta Virologica xo, 28I.
WHO EXPERTCOMMITTEEON INFLUENZA(I953). World Health Organization Technical Report Series, no. 64.
WHO CROUPON INFLUENZANOMENCLATURE(1971). A revised system of nomenclature for influenza viruses.
Bulletin of the Worm Health Organization 45, 1I9.
(Received 7 March 1972)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 07:37:34