The protective efficacy of a subunit avian influenza virus H5 vaccine based on recombinant baculovirus expressed H5 haemagglutinin antigen and an inactivated H5N2 avian influenza vaccine combined with a marker antigen (tetanus toxoid) was compared with commercially available inactivated H5N2 avian influenza vaccine in young ducks. Antibody responses, morbidity, mortality, and virus shedding were evaluated after challenge with a Vietnamese clade 1 H5N1 HPAI virus [A/VN/1203/04 (H5N1)] that was known to cause a high mortality rate in ducks. All three vaccines, administered with water-in-oil adjuvant, provided significant protection and dramatically reduced the duration and titer of virus shedding in the vaccinated challenged ducks compared with unvaccinated controls. The H5 subunit vaccine was shown to provide equivalent protection to the other two vaccines despite the H5 antibody responses in subunit vaccinated ducks being significantly lower prior to challenge. Ducks vaccinated with the H5N2 marker vaccine consistently produced antitetanus toxoid antibody. The two novel vaccines have attributes that would enhance H5N1 avian influenza surveillance and control by vaccination in small scale and village poultry systems.
Control of the H5N1 highly pathogenic avian influenza (HPAI) epizootic in village communities in Southeast and East Asia since 2003 has been difficult. Conventional control methods used for HPAI, including quarantine, enhanced biosecurity, and stamping-out are often not logistically possible in these villages where nutrition and livelihoods depend on low intensity poultry production. Use of vaccination against H5N1 avian influenza has become an important control tool in these settings [
Ducks and other members of the Anatidae family are natural host species of influenza A viruses [
Recent field and experimental vaccines for HPAI in poultry include inactivated, conventional whole virus vaccines [
Use of H5 vaccination in ducks complicates serological surveillance using H5 antibody testing, as both infected and vaccinated ducks produce antibody to H5 haemagglutinin. Strategies to differentiate infected from vaccinated animals (DIVA) have been considered and developed for incorporation into surveillance programs for avian influenza in poultry [
The recombinant baculovirus-expressed H5 vaccine was developed and produced at the Temasek Life Sciences Laboratory, Singapore (TLL vaccine) as described [
The inactivated H5N2 whole virus vaccine with tetanus toxoid marker antigen (TT/H5N2 vaccine) was prepared as a small pilot batch by Intervet International (Boxmeer, The Netherlands) using the standard H5N2 inactivated whole virus used in their Nobilis Influenza H5 vaccine combined with a commercial tetanus toxoid (TT) vaccine antigen (Pfizer, Melbourne, Australia) diluted to contain a final concentration of TT of 0.3 mg total protein per dose in the vaccine. This concentration had been shown to be highly immunogenic in previous studies when combined with inactivated H6N2 avian influenza virus in chickens and ducks [
The above vaccines were compared with a commercially available inactivated H5N2 whole virus vaccine (Nobilis Influenza H5 vaccine, Intervet International Boxmeer, The Netherlands) (H5N2 vaccine). Both TT/H5N2 and H5N2 vaccines were formulated in a water-in-oil adjuvant. The dose of these vaccines given to the ducks was 1.0 mL which is twice the dose given to chickens on the manufacturers recommendation.
The ducks used for these studies were young hybrid Pekin ducks (
The ducks were transferred to the high biocontainment BSL3+ animal house facility of the State Key Laboratory of Emerging Infectious Diseases at HKU for the H5N1 virus challenge. The ducks were housed as vaccine groups in cages with absorbent bedding inside Class 3 negative-pressure, flexible film isolators with HEPA-filtered inlet air and exhaust air. Food and water were replenished daily and room lighting was on a 12-hour cycle. Animals were individually leg banded for identification. All animal experimentation was carried out with the approval of the institutional animal ethics committee and in compliance with the facility biosafety requirements. Researchers wore positive air pressure respirators and protective suits in the BSL3+ animal facility at all times.
An initial experiment was conducted with the TLL vaccine using groups of young ducklings vaccinated with either high dose TLL antigen (HD group, 5 ducks), high dose with adjuvant (HDA group, 5 ducks), or low dose with adjuvant (LDA group, 6 ducks) compared with mock vaccinated control group (6 ducks). For logistical reasons, there were insufficient 1 week old ducklings available for experiment 1, so two age groups of ducklings (1 week and 3 weeks old) had to be used for this preliminary experiment with TLL vaccine. These ducklings were randomly allocated to the vaccine groups or mock vaccinated control group. They were immunised subcutaneously at the nape of neck with the vaccine (0.2 mL of HA antigen stock with equal volume of adjuvant or diluent) and a second dose of the respective vaccine was given 3 weeks later. Four weeks after the second vaccination H5N1 virus challenge was conducted in BSL3+ animal house as described below.
In the preliminary experiment with the TLL vaccine the challenge dose was based on a low dose challenge for H5N1 virus used in previous challenge studies in chickens but after completion of experiment 1 it was decided to use the higher H5N1 challenge dose that was more consistently used in other challenge studies with H5N1 in chickens and ducks using inactivated whole virus vaccines [
In the second experiment, groups of young ducklings were vaccinated with high dose TLL vaccine with adjuvant (TLL HDA group, 8 ducks), TT/H5N2 vaccine (7 ducks) or H5N2 vaccine (7 ducks) and compared with an unvaccinated control group (7 ducks). The ducklings were immunised subcutaneously at the nape of neck at 1 week of age, with TLL vaccine given as a 0.4 mL dose and the TT/H5N2 and H5N2 vaccines given as a 1.0 mL dose. A repeat dose of the respective vaccine was given 3 weeks later.
As postvaccination HI antibody response to the TLL vaccine was much weaker than for the whole virus vaccines and the challenge was to be at a higher virus dose, an additional higher dose of TLL vaccine (0.5 mL of HA antigen with equal volume of adjuvant) was given one week later (5 weeks of age) to this group. The H5N1 virus challenge was conducted in the BSL3+ animal house as described below when the birds were 9 weeks-old.
In both experiments, the ducks were individually identified by leg bands and blood samples were collected prevaccination, after vaccination 1, vaccination 2 (prechallenge for TT/H5N2 and H5N2 groups), vaccination 3 (prechallenge for TLL group) and postchallenge. Serum samples in both experiments were tested for H5 antibody by HI tests and in the second experiment for antibody levels to TT by ELISA assays as described below.
The challenge virus for both experiments was HPAI virus A/Vietnam/1203/04 (H5N1) (VN/1203/04), a clade 1 virus (WHO/OIE/FAO 2007), which was isolated from a human case early in the H5N1 epizootic, had been shown to be closely related genetically to H5N1 HPAI viruses from ducks and was shown to be highly pathogenic for ducks. Virus stocks were available at HKU. Preliminary titration (10-fold dilutions) of the virus stock was conducted in 7 week-old ducklings in the BSL3+ animal house to determine the challenge dose. The virus was administered by eye drop (100
In the BSL3+, animal house birds were anaesthetized by inhalation of isoflurane before being inoculated with 500
Ducks dying after challenge in the second experiment were subjected to postmortem examination to observe the gross and histopathology changes after challenge with a Vietnamese H5N1 HPAI virus. Pathological examination, including immunoperoxidase staining of tissues from affected ducks, was conducted as described [
Swab samples were collected into 1.0 mL aliquots of tissue culture medium 199 containing antibiotics (penicillin G (
Measurement of H5 antibody levels was conducted using the HI test procedure as described in the WHO Manual on Animal Influenza Diagnosis and Surveillance [
Antibody to TT was measured by both a competitive ELISA (C-ELISA) and an indirect ELISA as described previously [
In the indirect ELISA the antibody level was reported as a percentage of the positive =
The geometric mean titre (GMT) of the level of virus shed via cloaca and oropharynx from individual birds in each group was determined for consecutive days postchallenge and compared between groups by ANOVA. In groups where deaths occurred the GMT was determined for the remaining group members. The GMT of H5 HI antibody responses and the ELISA antibody responses to TT antigen were compared within and between groups pre- and postvaccination and pre- and postchallenge by ANOVA
In the preliminary study with the TLL vaccine at lower challenge dose, mild clinical signs of ocular discharge were noticed on day 2 and by day 3; mild ocular discharge or conjunctivitis was evident in 5/6 control and 5/5 HD ducks but only 1/6 LDA and 1/5 HDA ducks. Also by day 3 control ducks started to become depressed, unkempt with reduced grooming, and reduced interest in feed; one developed neurological signs (head tilt, head shake, tremors) and was found dead on day 4; another had neurological signs on day 5 and was euthanized; two others on day 7 and another on day 8 were moribund and were euthanized. The remaining control duck was observed with a mild head tilt on day 8 but continued to feed until the end of experiment on day 11. One HD duck was found dead on day 7. All HDA (5) and LDA (6) ducks and the other HD (4) ducks were healthy, active, alert and consuming feed until the completion of the experiment (Table
Efficacy of TLL H5 vaccine in ducks challenged with avian influenza virus [A/VN/1203/04 (H5N1)].
Vaccine group | Challenge dose | Morbidity | Mortality | HI serology |
Virus isolation-oropharyngeal |
Virus isolation-cloacal | |||
Prechallenge | Postchallenge | Day 2 | Day 4 | Day 2 | Day 4 | ||||
Control | 5/6 (6 days) | 0 | 1 (320) | 4/6 (2.27) | 6/6 (3.66) | 0/6 | 0/5 | ||
TLL H5 High-dose | 0/5 | 1/5 (day 7) | 0 | 4 (452) | 1/5 (0.42) | 3/5 (1.87) | 0/5 | 0/5 | |
TLL H5 Low-dose adjuvant | 0/6 | 0/6 | 0 | 5 (184) | 0/6 | 0/6 | 0/6 | 0/6 | |
TLL H5 High-dose adjuvant | 0/5 | 0/5 | 2 (14) | 3 (508) | 0/5 | 0/5 | 0/5 | 0/5 |
In the second phase of the study, with TLL HDA, H5N2/TT and H5N2 vaccinated birds and controls given the higher challenge dose, signs of ocular discharge were noticed on day 1 in a couple of control ducks and by day 2, 4/7 control birds had mild to copious eye discharge while only mild discharge was present in 1/8 TLL HDA vaccinated duck and in 1/7 H5N2 vaccinated birds. Two controls were found dead on day 2 and 3, respectively; one other showed neurological signs and another had a swollen, oedematous head on day 4 and both were found dead on day 5; two others developed mild ataxia and incoordination and another developed conjunctivitis on day 5, one of these became more severely ataxic although continuing to feed and was euthanized on day 9 and the remaining 2 control ducks had continued to feed but had neurological signs including incoordination and head tilt on day 10 when they were euthanized and the experiment was completed. In contrast, none of the TLL HDA vaccinated birds died or showed neurological disease signs, one H5N2 vaccinated duck was found dead on day 7 and one H5N2/TT vaccinated duck was found dead on day 8 (Table
Efficacy of H5 vaccines in ducks challenged with avian influenza virus [A/VN/1203/04 (H5N1)].
Vaccine group | Challenge dose | Morbidity | Mortality | HI serology |
Virus isolation-oropharyngeal |
Virus isolation-cloacal | |||
Prechallenge | Postchallenge | Day 2 | Day 4 | Day 2 | Day 4 | ||||
Control | 5/7 (4.8 days) | 0 | 1 (2560) | 7/7 (3.21) | 5/5 (3.42) | 1/7 (0.04) | 0/5 | ||
TLL H5 | 0/8 | 0/8 | 6 (16) | 8 (160) | 0/8 | 0/8 | 0/8 | 0/8 | |
Inactivated | 0/7 | 1/7 (day 7) | 7 (119) | 6 (160) | 0/7 | 0/7 | 0/7 | 0/7 | |
Inactivated | 0/7 | 1/7 (day 8) | 7 (49) | 6 (422) | 0/7 | 0/7 | 0/7 | 0/7 |
In the preliminary trial with the TLL vaccine, oropharyngeal virus shedding occurred in control ducks from the first day and continued for 5 days. On days 3 and 4, virus was detected from the oropharynx of all 6 ducks at titres between
In the comparative study with TLL HDA, H5N2 and H5N2/TT vaccines at the higher challenge dose, virus was detected in the oropharynx of control ducks between day 2 and day 5 at titres ranging from
In the preliminary TLL vaccine study, only 2/5 ducks from the HDA vaccine group showed weak H5 HI antibody responses (1 : 10 and 1 : 20) against VN/1203/04 antigen prior to challenge. None of the LDA, HD or control ducks had detectable antibody against VN/1203/04 prior to challenge. By day 11 postchallenge, most vaccinated ducks (12/15) showed a greater than four-fold increase in HI antibody titre against VN/1203/04. There were no significant difference in H5 HI GMT between HDA, LDA and HD groups postchallenge. The surviving control duck showed a postchallenge titre of 1 : 320.
In the comparative study with TLL HDA, H5N2 and H5N2/TT vaccinations, only 3/8 TLL HDA vaccinated ducks after the second vaccination showed low H5 HI antibody titres (1 : 10, 1 : 20 and 1 : 40) against VN/1203/04. After the third TLL vaccination (i.e., prechallenge) 6/8 ducks showed low (5 ducks at 1 : 10) or moderate (1 duck at 1 : 160) antibody titres to VN/1203/04. In contrast, all 7 H5N2 and all 7 H5N2/TT ducks had antibody to VN/1203/04 (HI titres from 1 : 20 to 1 : 640) after two vaccinations (i.e., prechallenge). The H5 HI GMT for the TLL HDA group prechallenge were significantly lower than H5N2 group (
None of the ducks in the TLL HDA, H5N2 and H5N2/TT vaccine groups or the control ducks had antibody to TT antigen in prevaccination serum samples, or in subsequent prechallenge or postchallenge samples from the TLL HDA or H5N2 vaccine groups, or controls with either the TT C-ELISA or the TT indirect ELISA.
In the H5N2/TT vaccine group after the first vaccination 3/7 ducks were positive in the TT C-ELISA and 1/7 ducks was positive by TT indirect ELISA; prechallenge all ducks were positive by TT C-ELISA and TT indirect ELISA; and postchallenge all were positive by TT indirect ELISA but 2 ducks were negative by TT C-ELISA. The mean % inhibition or % of mean positive controls OD, standard deviations for the C-ELISA and the indirect ELISA, respectively, and significance of differences at different time points are shown in Table
Antibody responses to tetanus toxoid antigens in ducks vaccinated with H5N2/TT marker vaccine and challenged with avian influenza virus [A/VN/1203/04 (H5N1)].
TT ELISA type | Post-vacc. 1 | Post-vacc. 2 | Postchallenge |
Mean (Std Dev.) | Mean (Std Dev.) | ||
TT C-ELISA | 30.7% | 62.4% | 35.7% |
TT Indirect ELISA | 7.7% | 65.4% | 67.7% |
Postmortem examination of dead control ducks from the second experiment revealed gross lesions of congestion and haemorrhage in multiple organs. These affected organs included the lungs, heart, pancreas, intestines, liver, spleen, kidney, and bursa. Increased pericardial fluid with fibrin clots, patches of pallor in the heart ventricle, areas of mottling and necrosis in pancreas, moist appearance of viscera, carcass emaciation and dehydration, and congested blood vessels in the brain were evident in some cases.
From the histopathology examination of tissue sections, the control ducks that died showed varying combinations of the following changes: congestion and oedema in the lungs; lymphohistiocytic tracheitis and necrosis of tracheal epithelial cells in some cases; congestion and multiple small foci of glial cell and neurone necrosis and/or multifocal lymphohistiocytic meningo-encephalitis, with some perivascular cuffing and gliosis in the brain; lymphohistiocytic myocarditis and multifocal myocardial necrosis in some cases; multifocal pancreatic necrosis; congestion and small foci of hepatic periportal lymphocyte necrosis; congestion and lymphoid depletion in the spleen and general viscera congestion.
Immunoperoxidase staining for H5 HA antigen in lung, brain, spleen, kidney, pancreas and heart sections for all control ducks and bursa, thymus and trachea in some control ducks confirmed the presence of H5N1 infection in these tissues.
However, gross and histopathological examination of the H5N2 duck that died on day 7 and the H5N2/TT duck that died on day 8 did not show lesions that were consistent with the above pathology that has been recorded previously for H5N1 HPAI in ducks [
The preliminary study showed that the TLL baculovirus-expressed H5 vaccine was able to protect ducks from severe disease and mortality following challenge from a dose of
The comparative vaccination study showed that the TLL HDA vaccine, given as a three dose regime and the TT/H5N2 marker vaccine (2 doses) were as effectively as commercial H5N2 avian influenza vaccine (2 doses) in protecting ducks from severe disease and mortality following challenge with a ten-fold higher dose (
The prechallenge serological responses to H5 HA for the ducks vaccinated with inactivated whole H5N2 virus and TT/H5N2 vaccines are similar to those reported previously with inactivated whole virus H5 vaccines [
In nature, replication of avian influenza viruses can occur in infected ducks without significant serum antibody response, but despite a poor antibody response, ducks were immune and could resist virus reinfection [
The more frequent reisolation of H5N1 virus in challenged ducks from the oropharynx and only sporadic isolation from cloacal swabs after challenge with A/VN/1203/04 H5N1 HPAI is consistent with previous studies where viral titres shed from the trachea of ducks were higher than from the cloaca for Eurasian H5N1 viruses since 2003 and is believed to be related to a shift in replication efficiency for the upper respiratory tract [
These challenge studies were conducted using a heterologous H5N1 virus strain and it could be expected that these vaccines would remain efficacious when used in geographical areas where different H5N1 virus clades exist. In contrast to human influenza vaccines, vaccines for poultry do not appear to require close antigenic homology with the haemagglutinin protein and remain able to offer broad cross-protection against diverse field viruses. For example, a single recombinant fowlpox-H5 vaccine was able to clinically protect chickens from challenge by nine different HP H5 strains that had between 87.3 and 100% HA protein sequence similarity with the vaccine strain [
Ducks vaccinated with the TT/H5N2 vaccine produced TT antibody responses similar to those in ducks vaccinated with TT/H6N2 vaccine in previous studies [
While vaccination remains an important disease control tool for avian influenza, experiences in Hong Kong, Italy, USA and elsewhere showed that vaccination should be part of a programme that incorporates use of quality and efficacious vaccines; quarantine, movement restriction, depopulation and disposal of affected flocks; application or enhancement of flock biosecurity in farms and markets; surveillance to monitor vaccine efficacy as well as field virus circulation; and public awareness on disease prevention and control [
The two vaccines evaluated in this study have shown equivalent efficacy to a commercial inactivated whole virus H5N2 vaccines in ducks, but also have some additional features that may enhance their effectiveness for field use in village poultry avian influenza control systems in developing countries. The recombinant H5 baculovirus vaccine has the major advantage that it does not require embryonated chicken eggs or a high biocontainment facility for production. It should also be relatively straightforward to alter the H5 insert in the baculovirus to match an evolving field H5N1 virus strain. This is a subunit vaccine and after 3 doses in this study produced only weak H5 antibody responses but protected ducks from disease and very significantly reduced virus shedding in H5N1 challenged ducks. The only qualifier with this vaccine is that the third vaccine dose at higher antigen load did produce increased H5 antibody response and this vaccine should be used at a higher antigen dose in ducks, which may induce stronger H5 HI antibody responses. Vaccinated and infected ducks developed an anamnestic antibody response to H5 HA after challenge but because it is a HA subunit vaccine, testing for antibody to influenza A proteins like NP, NS1 or M does confirm the presence of active infection and can be used as a DIVA strategy [
The TT/H5N2 marker vaccine is as efficacious as the commercial H5N2 vaccine and has the advantage of positively identifying vaccinated ducks. Serological surveillance of ducks for vaccination effectiveness and evidence of virus incursion is particularly difficult where accurate farm records or physical identification of vaccinated birds (leg or wing bands) are not available. Ducks vaccinated with this TT/H5N2 marker vaccine could be very effectively monitored for epidemiological purposes by simple ELISA and HI tests for TT antibody and H5 antibody, respectively. For example, if this vaccine was the approved vaccine for the district or region, simply testing a statistically appropriate sample by TT ELISA would determine the effectiveness of the coverage by the approved vaccine; concurrent testing of H5 antibody would monitor the potency of the vaccine in the field and whether the vaccine handling, storage and application was producing expected levels of flock immunity, and facilitate investigation of poor vaccine responses; testing for TT and H5 antibody during investigation of H5N1 outbreaks in vaccinated flocks would confirm that affected birds had, or had not been effectively vaccinated, and if the vaccine was not effective against the new circulating antigenic strains of the virus; and provide objective measures of risk for H5N1 cases in the district or region based on the level of flock immunity to allow prioritization of avian influenza control activities. Additionally, TT/H5N2 vaccinated ducks in this study, that were challenged with H5N1 viruses, showed significant rises in H5 antibody titre that could provide a signal of recent virus infection. Similar rises in H5 antibody titre have been observed in TT/H5N2 vaccinated chickens after challenge with H5N1 HPAI (Dr Deborah Middleton, CSIRO-AAHL, Geelong, personal communication). Field use of this vaccine could establish normal antibody response curves for TT and H5 antibodies in vaccinated, uninfected ducks. Higher than expected H5 antibody titres, indicating an anamnestic response, would then require further virological investigation.
Extensive vaccination programs for control of H5N1 HPAI have been or are currently being conducted in countries like Vietnam, Indonesia, China and Egypt but outbreaks are still occurring in village poultry systems with domestic ducks being implicated in virus persistence. The two novel vaccines evaluated in this study show equivalent efficacy to an existing commercial vaccine in ducks but they offer advantages for surveillance in village poultry systems in the above counties. The baculovirus recombinant H5 vaccine could be further developed and optimally manufactured in standard vaccine facilities in a developing country without the need for large scale chicken embryo culture facilities. The TT marker can be readily incorporated into any inactivated whole virus or subunit H5 vaccine and used in H5 vaccination programs to provide an effective positive marker of vaccination for surveillance of the small scale and village poultry industries. These tools are recommended for consideration as part of vaccination control programs in countries with ongoing problems of H5N1 HPAI control.
Avian influenza
Analysis of variance
Competitive enzyme linked immunosorbent assay
Differentiating infected from vaccinated animals
50% egg infectious dose
Food and Agricultural Organisation
Geometric mean titre
Haemagglutinin
High dose
High dose with adjuvant
Haemagglutination inhibition
Highly pathogenic avian influenza
Low dose with adjuvant
Lowly pathogenic avian influenza
Madin Darby canine kidney cells
Neuraminidase
Nuclear protein
Nonstructural protein
World Organisation for Animal Health
Phosphate-buffered saline/Tween 20
Temasek Life Sciences Laboratory
Tetanus toxoid
World Health Organisation.
The authors thank Temasek Life Sciences Laboratory, Singapore for providing the baculovirus recombinant H5 vaccine; Intervet International, Boxmeer, The Netherlands for formulating and providing the combined Nobilis Influenza H5 (H5N2) and TT vaccine; Pfizer, Melbourne, Australia for providing the prevaccine formulation TT antigen. The authors are grateful to the State Key Laboratory for Emerging Infectious Diseases, Department of Microbiology of the University of Hong Kong for provision of facilities and resources to conduct this study and particularly they would like to thank Issac Chow, Edward Ma, Iris Ng, Sia Sin Fun, Dr. Leo Poon, and other staff at the Department for their dedicated assistance. Funding for the project was jointly contributed by the Agri-Food and Veterinary Authority, Singapore with the support of the Director-General, Dr. Chua Sin Bin; the Australian Biosecurity Cooperative Research Centre; University Grants Committee of the Hong Kong Special Administrative Region, China (Project No. AoE/M12/06); Stone Ridge Ventures Pty Ltd and Murdoch University, Australia.