Vol.2, No.2, 97-100 (2010) Health
SciRes Copyright © 2010 Openly accessible at http://www.scirp.org/journal/HEALTH/
Comparative study of haemagglutination inhibition,
Agar gel precipitation test, Serum neutralization and
Enzyme linked immunosorbent assay for detection to
avian influenza viruses
Shahid Faraz, Muhammad Abubakar 1, Mohammad Farooque2, Sarfaraz Ali Fazlani, Ghluam
Hussain Jaffar
Faculty of Veterinary and Animal Sciences, Lasbela University of Agriculture, Water and Marine Sciences, Uthal.
1National Veterinary Laboratory, Park Road, Islamabad.
2Quarantine Officer, Animal Quarantine Department, Islamabad.
Received 22 October 2009; revised 7 December 2009; accepted 9 December 2009.
The sensitivity, specificity and reproducibility of
the serological tests for detection of avian in-
fluenza viruses were carried-out by using Ham-
agglutination inhibition (HI), Agar gel precipita-
tion test (AGPT), and Enzyme linked immu-
nosorbent assay (ELISA) and Serum neutraliza-
tion test. The geometric mean titre (GMT) of hae-
magglutination inhibition antibodies recor- ded
as log2 indicated that the post vaccination titres
in the field were on higher side i.e., 7.9 for H7
and 5.9 for H9. The correlation between HI titre
and AGPT affirmed that for the AGPT test need
high antibody titre for positive reaction. The
pooled sera were also used to correlate the se-
rum neutralization test and enzyme linked im-
muno-sorbent assay. The serial two fold dilute-
ons were tested for the serum neutralization ac-
tivity and concluded that the HI titre log2 4 pro-
vided 100% protection, than 52% and 45% pro-
tection in 1:2 and 1:4 dilution was recorded,
respectively. Similarly, the ELISA test showed
positive results up to 1:16 HI titre, i.e. log2 4 and
confirmed the linear relation between these two
serological tests. In HI test, the concentration of
antigen can influence the result. It also needs
careful preparation of concentration of eryth-
rocyte suspension. Agar Gel immuno-diffusion
is basically a qualitative test as it can not de-
termine the quantity of antigen or antibody with
the help of this test. It lacks the level of sensi-
tivity as offered by other test. If serum neutrali-
zation test is performed on a pooled serum sam-
ples, then it could lead to a false conclusion on
antibodies status. ELISA is most sensitive, spe-
cific and accurate as compare to all other sero-
logical tests.
Keywords: Serology; HI; ELISA and Avian Influ-
enza Virus.
Infectious viral diseases are a major threat to poultry.
Avian influenza is one of most important among them
which inflicts heavy economical losses. It is caused by a
virus that belongs to family Orthomixoviridae, genus
influenza A virus [1].
Avian influenza virus is classified into subtypes on the
basis of antigenic differences in their surface glycopro-
tein hemagglutinin (HA; H) and neuraminidase, (NA; N).
To date 16H subtypes (H1-H16) and 9N subtypes (N1-
N9) have been recognized [2]. It rapidly infects the poul-
try when heavy outbreak occurs. Many species of birds
are found to be susceptible to avian influenza virus.
Avian influenza viruses are circulating periodically
among the domestic poultry over 100 years [3]. Wild
waterfowl and shorebirds are considered to be reservoir
of influenza A virus because the species harbor all 16
HA subtypes [4]. Although chicken and turkeys are not
natural host species for avian influenza but these viruses
routinely cross over from wild birds reservoir to infect
poultry birds [5].
During 2003 and 2004, there were quite a few out-
breaks of avian influenza throughout the Asia, including
South Korea, Japan, Indonesia, Vietnam, Thailand and
China. The outbreak resulted not only tremendous eco-
nomic loses in the poultry industry but also claimed dea-
th in human in Vietnam and Thailand in 2004.
98 S. Faraz et al. / HEALTH 2 (2010) 97-100
SciRes Copyright © 2010 Openly accessible at http://www.scirp.org/journal/HEALTH/
Table 1. Calculation of the 50% endpoint of a neutralization test by the method of Reed and Muench.
Serum Dilution Response Accumulated Values
Value Log Infection
Ratio Infected
Non In-
fected em-
Non In-
/ Total
% Infected
Undiluted - 0/5 - - - - - -
1:2 10-0.3 0/5 5 0 18 0 18/18 100%
1:4 10-0.6 1/5 4 1 13 12 13/25 52%
1:8 10-0.9 2/5 3 2 9 11 9/20 45%
1:16 10-1.2 3/5 3 2 6 9 6/15 40%
1:32 10-1.5 3/5 2 3 3 7 3/10 30%
1:64 10-1.8 4/5 1 4 1 4 1/5 20%
Control Sera
- 05 - - - - - -
Control Sera
- 00 - - - - - -
Table 2. Correlation among HI (log2) AGP and ELISA.
H7 H9
00 00 - 2 - 0.0
00 00 - 4 - 0.1
00 00 - 8 - 0.2
16 - 0.4 16 - 0.5
32 - 0.8 32 - 1.01
64 - 0.9 64 - 1.32
128 + 1.02 128 - 1.28
256 + 1.43 256 + 1.42
512 ++ 1.72 512 ++ 1.59
1024 ++ 1.78 1024 ++ 1.62
2048 ++ 1.82 2048 ++ 1.80
Shows positive and negative Pattern reactivity of AGPT
- Negative
+ Positive
++ Strong Positive
Avian Influenza viruses have been reported to cause
high mortality in Pakistan. The first outbreak of avian
influenza virus in Pakistan was recorded in 1995 when
highly pathogenic influenza of H7 subtype was identi-
fied Naeem and Hussain [6]. In 1998 Avian Influenza
subtype H9N2 was isolated from breeder flock showed
a reduction in egg production along with respiratory
infection Naeem at el. [7]. The H7 and H9 strains were
still being isolated in Karachi in 2004.
New diagnostic test for emerging avian diseases are
developed using molecular biology techniques. These
mainly rely on detection of nucleic acid (either RNA or
DNA) unique to that pathogen. Analysis is being used to
identify the species, subspecies, type and subtype and
even in some cases individual strains. These tests are
based on molecular techniques such as Restriction frag-
ment length polymorphism (RFLP), Polymerase chain
reaction (PCR) and nucleic acid sequencing. The sero-
logical detection of antibodies for avian Influenza virus
in the poultry birds is of the great importance in pre-
venting and controlling the avian influenza. This study
revealed standard sero-diagnostic techniques used to
evaluate sensitivity, specificity and accuracy of the tech-
niques in identification of avian influenza in birds.
S. Faraz et al. / HEALTH 2 (2010) 97-100
SciRes Copyright © 2010 Openly accessible at http://www.scirp.org/journal/HEALTH/
Day-old broiler chicks were purchased and used for the
experiment. The broiler chickens were divided into three
groups A, B and C. Group A and B were vaccinated with
H7N3 and H9N2, respectively, while group C was kept
as control. Before vaccination, the blood samples were
collected randomly from 5% birds prior to vaccination at
3rd week of age to determine the status of maternal anti-
bodies in the sera of chicks.
2.1. Serum Samples
To assess the susceptibility of chickens the serum sam-
ples were checked prior to vaccinate and found zero titre
against avian influenza. Blood samples were allowed to
coagulate than serum was collected, marked, centrifuged
and stored at 20 oC prior to use.
2.2. Source Virus
A local virus isolate chicken/Pakistan/23/99 (H9N2)
virus was used with ELD 50 of 109.26/0.1ml. Another vi-
rus isolate H7N3 was also received from Sindh Poultry
Vaccine Centre (SPVC) with ELD 50 of 107.56/0.1ml.
Both serotypes were confirmed from southeast poultry
research institute Georgia.
2.3. Haemagglutination Inhibition Test
The (HI) test was performed by adopting the technique
described in (OIE) manual of diagnostic tests using 4HA
units of H7 and H9 viruses with 1% suspension of
washed chicken RBCs. Serum was diluted while antigen
was constant.
2.4. Agar Gel Precipitation (AGP) Test
This test was performed to adopt the method described
by Beard in 1998. Agar was prepared in borate buffer,
than allowed to solidify. Cylinder of gel were cut with a
gel cutter, than antigen was placed in the central wells
while the sera were in the peripheral wells than the Petri
plates were kept in moist chamber to avoid drying of gel.
2.5. Virus Neutralization Test
The virus neutralization test was performed to titrate the
antibodies against avian influenza virus; both strains
were checked and confirmed for sterility. Test was done
by using the embryo inoculation, ß-method (constant
virus diluted serum) and followed the technique de-
scribed by Beard [8], while endpoint titre was calculated
as log2 exponent that was 50% neutralization endpoint or
PD50 by the formula of Reed and Munch.
2.6. ELISA Test
The ELISA test was run in accordance as described by
Terry and Tony in 1991 [9] and as described in the
monograph of Australian Centre for International Agri-
cultural Research (ACIAR 1995), in which plates were
read at 405nm absorbancy in a microplate ELISA reader.
The geometric mean titer (GMT) in log2 of serotype
H7 was recorded as 7.9±0.23 (66) and for H9 was cal-
culated as 5.7±0.29 (66). The mean H7 titre was sig-
nificantly (p<0.05) higher than H9.
3.1. Agar Gel Precipitation Test
The pattern of reactivity of both strains is similar and no
remarkable difference has been observed. However, the
precipitation line can be seen in those samples which ha-
ve HI titres more than 7 log2 in H9 subtype. Moreover,
there is a linear relation between precipitates of antibod-
ies and antigen. Results also showed that H7 subtype
was not positive till the HI titre was 9 log2.
3.2. Virus Neutralization Test
Before the using of strains for serum Neutralization Test,
the ELD 50 was determined and recorded as 1011.26/ml
for H9 and 108.31/ml for H7subtype. The results in (Table
1) indicated that undiluted and 1:2 dilution of serum
showed 100% protection and than gradually decreased
i-e 52% and 45% protection in 1:4 and 1:8 dilution
respectively. However, negative control sera showed no
protection while positive showed 100%.
3.3. ELISA Test
The average optical density value (OD405nm) of posi-
tive sera showed 0.4 OD up to 1:16 dilution, i.e, 4 on log
2. Both serotypes showed positive correlation between
HI titre and ELISA at GMT Log24. However there is a
linear relation between HI titre, ELISA and AGPT
shown in Table 2.
Present study revealed that ELISA is better than HI in
identifying sera with low antibody titer. HI titer within
range of 1:2 to 1:4 are considered as suspected whereas
1:8 may be considered as positive and according to the
manufacturer of ELISA kit, 0.2 value considered as a
suspected positive value if value was recorded more than
0.6 considered as positive Ewing et al.[10].
Meilinjin et al. [11] conducted the comparative study
on newly developed ELISA technique, in which they
used nucleoprotein as antigen for detecting the antibod-
ies to Avian Influenza Virus. They compared this techni-
que with (HI) and ELISA test. Comparative study indi-
cated that these two tests had a high agreement ratio and
S. Faraz et al. / HEALTH 2 (2010) 97-100
SciRes Copyright © 2010 Openly accessible at http://www.scirp.org/journal/HEALTH/
no statistically significant difference.
Haem-agglutination inhibition test and titration of an-
tibodies by geometric mean titer (GMT) is the convenient
and best technique to measure the level of protection in
vaccinated chickens as well as to check the efficacy of
vaccine, such types of results were also reported by Meu-
lemans, et al. [12] who carried out a study to run HI,
AGP and ELISA for measuring the antibodies against
avian influenza virus infection. His studies showed a lin-
ear relation among these three techniques, when the chic-
kens were exposed with Avian Influenza Virus the anti-
body status was measured 157 day post infection. It was
suggested that AGP is a type-specific; the HI detects only
haemagglutinin subtypes, however only ELISA is most
sensitive test to detect the antibodies.
The findings of present study showed a degree of cor-
relation between HI, AGP, ELISA and virus Neutraliza-
tion test, however the degree of accuracy depends on the
standardization of reagents, buffers and all other pa-
rameters related to the standard procedures of the test.
This statement correlates with the availability of equip-
ments and expertise in different laboratories as they per-
form the same test. Thats why the repeatability and re-
producibility are two key features that a test must have
to be accepted for different laboratories.
[1] Swayne, D.E., Senne, D.A. and Beard, C.W. (1998) Avi-
an Influenza in: A laboratory manual for the isolation and
identification of avian pathogens, 4, 150-155.
[2] Micheal L.P. and David E.S. (2005) Public health risk
from Avian Influenza viruses. Avian Dis. 49, 317-327.
[3] Swayne, D.F. and Suarez, D.L. (2000) Highly pathogenic
avian influenza. Rey. Sci. Tech. Off. Int. Epizoot. 19, 463
[4] Suss, J., Schafer, J., Sinnecker, H. and Webster, R.G.
(1994) Influenza virus subtypes in aquatic birds of east-
ern Germany. Arch. Virol. 135, 101-114.
[5] Suarez, D.L. (2000). Evolution of avian influenza viruses.
Vet. Microbiol., 74, 15-27.
[6] Naeem, K, and Hussain, M. (1995) An outbreak of avian
influenza in poultry in Pakistan Vet. Rec., 137, 439.
[7] Naeem, K, Ullah, A., Manvell, R.J. and Alexander, D.J.
(1999) Avian influenza subtype H9N2 in poultry in Paki-
stan, Vet. Rec., 145, 560.
[8] Beard, C.W. (1998) Isolation and identification of avian
pathogens. In: Hitchner, S.B., Domermuth, C.H., Pur-
chase, H.G. and Williams, J.E. (ed.). 67-69.
[9] Terry, S. and Tony, D. (1991) Australian center for inter-
national agricultural research (ACIAR). University of
Pertanian Malaysia.
[10] Ewing, M.L., Kleven S.H. and Brown, M.B. (1994)
Comparison of enzyme linked immunosorbent assay
(ELISA) and hemagglutination-inhibition for detection of
antibody to Mycoplasma gallisepticum in commercial
broiler. Avian Diseases, 40, 13-22.
[11] Meilinjin, A.B, Wang, G., Zhang, A.R., Zhao, S., Li, H.,
Tan, Y. and Chen, H. (2004) Development of enzyme-
linked immunosorbent assay with nucleoprotein as anti-
gen detection of antibodies to avian influenza virus. Lab
or animal infectious diseases.
[12] Meulemans, M, Carlier, C., Gonza, M. and Petit (1987)
Comparasion of hemagglutination inhibition, agar gel
precipitation, Enzyme linked Immunosorbent assay for
measuring the antibodies against influenza virus in
chicken. Avian Dis. 31, 560-563.