Edited by: Constantinos S. Kyriakis, Auburn University, United States
Reviewed by: Renukaradhya J. Gourapura, The Ohio State University, United States; Eric James, Sanaria, United States
This article was submitted to Veterinary Infectious Diseases, a section of the journal Frontiers in Veterinary Science
†These authors have contributed equally to this work and share last authorship
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
H9N2 viruses have become, over the last 20 years, one of the most diffused poultry pathogens and have reached a level of endemicity in several countries. Attempts to control the spread and reduce the circulation of H9N2 have relied mainly on vaccination in endemic countries. However, the high level of adaptation to poultry, testified by low minimum infectious doses, replication to high titers, and high transmissibility, has severely hampered the results of vaccination campaigns. Commercially available vaccines have demonstrated high efficacy in protecting against clinical disease, but variable results have also been observed in reducing the level of replication and viral shedding in domestic poultry species. Antigenic drift and increased chances of zoonotic infections are the results of incomplete protection offered by the currently available vaccines, of which the vast majority are based on formalin-inactivated whole virus antigens. In our work, we evaluated experimental vaccines based on an H9N2 virus, inactivated by irradiation treatment, in reducing viral shedding upon different challenge doses and compared their efficacy with formalin-inactivated vaccines. Moreover, we evaluated mucosal delivery of inactivated antigens as an alternative route to subcutaneous and intramuscular vaccination. The results showed complete protection and prevention of replication in subcutaneously vaccinated Specific Pathogen Free White Leghorn chickens at low-to-intermediate challenge doses but a limited reduction of shedding at a high challenge dose. Mucosally vaccinated chickens showed a more variable response to experimental infection at all tested challenge doses and the main effect of vaccination attained the reduction of infected birds in the early phase of infection. Concerning mucosal vaccination, the irradiated vaccine was the only one affording complete protection from infection at the lowest challenge dose. Vaccine formulations based on H9N2 inactivated by irradiation demonstrated a potential for better performances than vaccines based on the formalin-inactivated antigen in terms of reduction of shedding and prevention of infection.
Although wild waterfowl are the natural hosts of avian influenza (AI), H9N2 subtype viruses are relatively uncommon in wild birds (
There is no widespread consensus on the classification of H9N2 avian viruses; however, epidemiological and phylogenetic analyses of the hemagglutinin (HA) gene of H9N2 influenza viruses revealed that at least two major different lineages can be distinguished, the American and Eurasian lineage. The latter can be further divided into the BJ/94, the Y280/G9, the G1, and possibly a fourth lineage (unrelated to the previous three) found mainly in turkeys reared in Europe (
Of the different lineages of H9N2, the G1 lineage, first detected in Hong Kong in 1997 (
Vaccines have been used to reduce clinical disease and lower the burden on the poultry industry; however, insufficient attention has been focused on the effect of vaccines on the reduction of viral shedding (
The gamma-irradiation-mediated killing of viruses was explored with little success to develop vaccines since the 1950s. However, a renewed interest in this technology has risen due to: (a) the invention of newer and safer irradiators that can deliver high doses, (b) the introduction of radio-protective compounds that can preserve antigens during irradiation, and (c) a better understanding of the immune system (
In our work, we compared the immunogenicity and the efficacy of H9N2 experimental vaccines based on the antigens inactivated by chemical and irradiation methods, administered by mucosal or intramuscular routes in Specific Pathogen Free (SPF) White Leghorn chickens. The efficacy of the different vaccines and administration routes was measured upon challenge with different doses of an H9N2 isolate belonging to the G1 lineage, aiming to understand if the irradiation technology could improve current vaccination strategies and if the mucosal administration of the inactivated vaccines is able to elicit a protective level of immunity.
An AIV H9N2 isolate from the Middle East belonging to the G1 lineage (A/Chicken/Saudi Arabia/3622-31/13) was propagated and titrated in 10-day-old embryonated SPF chicken eggs (Charles Rivers) at 37°C for 72 h. Viral titrations were performed by inoculating 100 μL virus dilutions (10−3-10−9) in 10–11-day-old embryonated chicken eggs in the allantoic fluid. The inoculated eggs were incubated at 37°C and observed every 24 h to detect mortality for 7 days. Allantoic fluids harvested from the eggs were tested by the HA assay to detect viral replication, according to standard procedures (
Inactivation by irradiation was performed at the International Atomic Energy Agency (IAEA) Laboratories in Seibersdorf, Austria, by following established protocols. The virus stock was mixed with 1M trehalose (trehalose dihydrate; Sigma) 50% V/V, aliquoted into 5 mL volume, and immediately frozen. Frozen samples incubated with dry ice were irradiated using a Model 812 Co-60 irradiator at a dose rate of 66.532 Gy/min (Foss Therapy Services, Inc., California, USA). The irradiator was regularly calibrated using an ionization chamber that also mapped the delivered dose in the location where the samples are irradiated. Initial doses of 5, 10, 15, 20, 25, 30, and 40 kGy were applied to identify the D10 value, the dose required to reduce virus load by 90% or 1 log (
Aliquots of untreated, formalin-inactivated, and irradiated viruses were analyzed by negative staining TEM according to standard procedures for viral identification and examination. A formvar/carbon supported copper grid (Electron Microscopy Sciences Formvar/Carbon Copper Grid 200 Mesh) was placed flat on the bottom of the vial and 90 μL of samples were dispensed on the top of the grid. After high-speed centrifugation (28–30 psi or 100,000 ×
BID50 determination was based on the methods described by Swayne and Slemons (
A total of 40 one-day-old SPF White Leghorn chickens were equally divided into five groups and housed in BSL3 poultry isolators (HM 1900, Montair Andersen BV, Kronenberg, The Netherlands). Two groups were vaccinated oculo-nasally (ON) with either the irradiated-H9N2 (ON-Irr) or the formalin-inactivated H9N2 (ON-For) antigens without adjuvants, respectively. The other two groups were vaccinated subcutaneously (SC) with either irradiated-H9N2 (SC-Irr) or formalin-inactivated H9N2 (SC-For) antigens, respectively, in a water-in-oil (W/O) 7:3 (v/v) emulsion with a commercial adjuvant for poultry (ISA71VG, Seppic). A fifth group served as a negative non-vaccinated control. All groups were vaccinated twice at 14 and 28 days of age and blood samples were taken before each vaccination. The amount of H9N2 antigen given to each bird was standardized to 128 hemagglutinating units (HAU) for each immunization. Two weeks after the second dose (i.e., 42 days of age) blood samples were taken from all the chickens and a homologous challenge was performed by the oronasal route at a dose of 106 EID50/100 μL. Tracheal swabs were collected from all the birds on day 2, 4, and 7 p.i. to evaluate viral shedding. Fourteen days p.i. (dpi), a final blood sample was taken from all the birds to evaluate seroconversion.
A total of 150 one-day-old SPF White Leghorn chickens were divided into five different experimental groups. The first group was vaccinated ON with irradiated-H9N2 (ON-Irr-Adj) in a 1:1 suspension with the mucosal adjuvant IMS1313 (Seppic, France), the second group received by the ON route a formalin-inactivated H9N2 vaccine in a 1:1 suspension with IMS1313 (ON-For-Adj). The other two groups were vaccinated SC in the same way as in animal trial 1 (SC-Irr, SC-For). The amount of H9N2 antigen given to each bird was standardized to 128 HAU for each immunization. A fifth group served as negative non-vaccinated control. All the vaccinated birds received two doses of the experimental vaccines at 14 and 28 days of age and blood samples were taken before each vaccination. Two weeks after the second dose (i.e., 42 days of age) blood samples were taken, and within each group, birds were equally divided into subgroups of 15 birds each and challenged with either 103 or 104 EID50/100 μL of the homologous virus. Clinical signs were monitored daily and tracheal swabs for quantification of viral shedding were collected on day 1, 2, 3, 4, 5, 7, and 9 p.i. Fourteen days p.i., a final blood sample was taken from all the birds to evaluate seroconversion.
RRT-PCR targeting the M-gene was used to determine the BID50 and to compare viral shedding in the respiratory tract in each experimental group, in a qualitative and quantitative setup, respectively (
Ten-fold serial dilutions of strain-specific negative-sense
Viral replication was plotted as the mean viral load ± SD using the Prism 9.1.2 (GraphPad). For graphical and statistical purposes, samples testing negative or with a viral load below the LoQ were given a value of 100.7 copies/5 μL of total RNA.
To detect the humoral immune response of vaccination, hemagglutination inhibition (HI) assays and a commercial ELISA assay targeting the nucleoprotein (NP) of type A influenza viruses were performed on all serum samples collected during animal trials 1 and 2. HI assays were performed according to standard protocol using the homologous vaccine antigen (
The anti-NP ELISA (ID Screen® Influenza A Nucleoprotein Indirect, IDVet, France) was performed according to the manufacturer's recommendation using positive and negative controls provided with the commercial kit.
The shedding dynamics from 1 to 12 dpi of the control population and those administered with formalin-inactivated and the irradiated vaccines were modeled through General Additive Model (GAM). GAM was performed as implemented in the ‘mgcv' R package, which was also used to assess the concurvity and significance of base functions, model selection was performed through the Akaike information criterion (AIC), and the model assumptions were verified through the graphical assessment of the models' residuals using the R package ‘gratia' (
To assess the overall shedding difference significance among treatments for each dose/route combination, general linear mixed models (GLMM) of the log-transformed Shed01 were fitted using treatment as a fixed factor, DPI, and sample ID as random variables as implemented in the “lme4” R package (
The D10 dose was identified as 5.46 kGy (
Inactivation and safety of formalin-treated and irradiated H9N2 used in the experimental vaccines were confirmed by three blind passages in 10-day-old embryonated eggs. Additionally, no loss in HA titer was observed irrespective of the inactivation method used. Upon Transmission electron microscopy (TEM) examination (
Negative stain TEM images of virus (× 180,000).
To infer the BID50 of the challenge virus, we performed multiple infection experiments at different challenge doses (103−6 EID50/100 μL) in poultry isolators units. Following the challenge, tracheal swabs were collected daily and RRT-PCR tests were run to identify infected birds. The results of the challenge are shown in
BID50 determination in 6-weeks-old White Leghorn SPF chickens, each infection experiment was performed by oronasal installation of 100 μL of infectious allantoic fluid diluted in PBS to five (
2 | 3 | 40 (33.2) | |
3 | 2 | 60 (25.2) | |
5 | 0 | 100 (21.4) | |
5 | 0 | 100 (23.1) |
Formalin-inactivated vaccines represent the most common type of traditional vaccine available for AI. Despite the extensive knowledge of the protection offered by inactivated vaccines when administered SC, control of H9N2 infection is difficult under field conditions and most of the countries in which vaccination is applied are still endemic to H9N2. In our work, we aimed to compare the protective efficacy of formalin-inactivated and irradiation-inactivated H9N2 experimental vaccines against different challenge doses of a homologous virus to the vaccine antigen.
The serological analysis showed that both formulations, when administrated SC, were able to produce high-antibody titers in immunized birds before challenge (i.e., Geometric Mean Titer (GMT) >10 log2 in all of the immunized groups) according to both HI and NP-ELISA tests. Higher mean HI titers were observed in birds immunized in trial 2 compared to trial 1, possibly due to improved vaccine preparation methods (extended emulsion time, higher shearing speed, and preparation performed on ice), which were adjusted following discussion with the manufacturer. Nonetheless, in SC-For and SC-Irr groups challenged with the same dose, no significant difference was observed in the GMTs.
In animal trial 1, we challenged chickens with a 106 EID50 dose (i.e., 102.5 times greater than the BID50) of the H9N2 isolate. Quantitative RRT-PCR was performed on tracheal swabs collected on days 2, 4, and 7 p.i. showed only partial virological protection in birds, irrespective of the type of inactivation method. However, on day 2 p.i., viral shedding was significantly lower in SC-Irr (102.38 ± 102.70 copies/5 μL) and SC-For (103.54 ± 103.94 copies/5 μL) groups than in the control chickens (104.94 ± 104.83 copies/5 μL), but differences between the two vaccinated groups were not statistically significant. In both vaccinated groups, 75% (6/8) of birds resulted to be positive on day 2 p.i. (
Effects of subcutaneous (SC) vaccination upon infection at different H9N2 challenge doses.
Models of the shedding levels of irradiated and formalin-inactivated vaccines and a non-vaccinated control for 12 days post infection (DPI) for three challenge doses (103, 104, and 106 EID50). For each group, the dots represent the measured observations, the solid line represents the fitted model, and the shaded contour shape represents the 95% confidence intervals of the fitted model. For each plot, the inset compares the overall effect of the group on the shedding, with the solid circle representing the effect value and the vertical bars the 95% confidence interval. Panels
We then evaluated the protective efficacy at lower challenge doses (103 and 104 EID50), below and above the BID50, to better discriminate differences in the ability of these vaccines to prevent an infection. Shedding results showed complete prevention of infection in all challenged birds, resulting in 100% efficacy of both the experimental vaccines in preventing infection (
Mucosal vaccination of the upper respiratory tract in poultry is an attractive alternative to SC vaccination due to the potential advantages offered by mass administration and the capacity of mucosal vaccines to elicit mucosal immunity at the site of entry of respiratory viruses. To assess differences in the protective efficacy between irradiated and formalin-inactivated H9N2 antigens administered by the mucosal route, we performed three challenges, including doses of 103, 104, and 106 EID50.
Serological analyses showed variable HI titers before the challenge in all vaccinated birds and no significant differences were observed in terms of HI GMT between formalin inactivated and irradiated vaccinated groups (3.93 log2 and 4.71 log2, respectively). Interestingly, NP-ELISA results were clearly distinguishable. The NP-ELISA performed on sera collected before the challenge gave negative results in the irradiated vaccinated groups, while in the formalin inactivated groups few (3/38, 7.9%) chickens seroconverted. After the challenge, the NP-ELISA showed seroconversion in the infected chickens.
As shown in
Effects of mucosal (oculo-nasal, ON) vaccination upon infection at different H9N2 challenge doses.
At lower challenge doses, the effect of mucosal vaccination on the prevention of infection was evident, recording fewer positive birds compared to the control group during the first 2 days after the challenge. Upon inoculation with 104 EID50, the percentage of positive birds in the ON-Irr-Adj, the ON-For-Adj and the control groups ranged between 23.1% (3/13)−53.8% (7/13), 26.7% (4/15)−46.65 (7/15) and 66.7% (10/15)−86.6% (13/15), respectively (
Upon challenge with 103 EID50, the percentage of positive birds in the ON-Irr-Adj, the ON-For-Adj and the control groups ranged between 6.7% (1/15)−0.0% (0/13), 13.3% (2/15)−20.0% (3/15) and 6.7% (1/15)−20.0% (3/15), respectively (
Vaccination to prevent H9N2 AIV infection in poultry has been used extensively since the late 1990s first in China and then in regions that became endemic following the global spread of these poultry-adapted viruses (
In our study, we demonstrated that irradiation is a valid alternative to formalin for the inactivation viruses, as previously demonstrated for other human influenza strains (
When administered SC, no significant differences were recorded in terms of immunogenicity between the two vaccines. However, HI titers were higher than previously reported for similar antigen concentrations (
When birds were challenged with 106 EID50, SC vaccinated groups recorded a significant reduction of shedding (101.40-102.56 fold reduction) in the trachea and a reduced number of positive birds in the early phase of the study, on day 2 p.i. Nonetheless, such reduction did not affect the overall number of birds becoming infected in the vaccinated groups, while it delayed the peak of shedding to day 4 p.i. Interestingly, although mean loads on day 7 p.i. were higher in both vaccinated groups than in the control group, the irradiated and the formalinated vaccines afforded viral clearance in 3/8 and 1/8 animals, respectively, as opposed to the control group in which all birds were still actively shedding the virus. At lower challenges, both vaccines provided sterilizing immunity according to both virological and serological results, achieving a result rarely described in the literature for AI (
When ON vaccinated groups were challenged with 106 EID50, no significant differences were recorded in terms of cumulative shedding in comparison with the control group. However, a significant reduction in shedding was observed in the ON-Irr-adj group at 2 dpi leading to a substantial delay in the shedding dynamic, albeit failing to effectively reduce the circulation of the virus in the flock. This scenario closely replicated what was observed with the SC vaccination. To effectively immunize animals
For this reason, although we could not differentiate primary from secondary infections, we speculate that the number of positive birds identified during the first 2 days after the challenge, might be largely attributable to primarily infected birds. In light of this, the better performance of the irradiated vaccine during the early phases of the lowest challenge with 103 EID50 suggests the possibility that irradiated antigens administrated
This might depend on the higher avidity/affinity of secretory IgA (S-IgA) mounted against a better preserved antigenic structure inactivated through irradiation (
Altogether, our results indicate that vaccination with an antigen inactivated by gamma irradiation achieves excellent results in terms of prevention of infection against low-to-intermediate H9N2 challenges if the vaccine is administrated SC. Moreover, irradiation of the antigen resulted in a shorter duration of shedding when compared to the traditional formalin-inactivated antigen. Additional experimental evidence on the efficacy of irradiated antigens in protecting against H9N2 and other AI subtypes is necessary to confirm our observations and understand whether this method is either comparable or superior to others currently used in vaccine manufacturing. This inactivation method might represent an alternative to the traditional formalin-based approach, especially in light of the recent advancements replacing radioactive material with the safer and cheaper low-energy electron irradiation technology. Although ON vaccination with an inactivated antigen only partially reduced replication against a high challenge, the performance against a low-to-moderate challenge with a highly infectious strain of H9N2 proved the potential of this innovative delivery route, in particular when the antigen was inactivated by irradiation. Undoubtedly, vaccination by spray or mixing in drinking water can stimulate the mucosae of the upper respiratory and digestive tracts, and is also less expensive and more easily applicable in emergency situations than the SC vaccination and can be designed for periodic boosters. Interestingly, mucosal vaccination with the irradiated H9N2 antigen revealed a complete lack of seroconversion against the structural NP, offering a possible Differentiating Infected from Vaccinated (DIVA) approach that would simply rely on existing commercial anti-NP ELISA assays.
A significant reduction of environmental contamination is one of the secondary goals of vaccination campaigns in endemic countries where human exposure to zoonotic H9N2 viruses is of concern. Noticeable achievements in this sense have been recorded after the deployment of nationwide immunization against H5/H7 HPAI and LPAI viruses in China, with a dramatic drop in the number of H7N9 cases (
The original contributions presented in the study are included in the article/
The animal study was reviewed and approved by Istituto Zooprofilattico Sperimentale delle Venezie Ethics Committee and the Italian Ministry of Health.
CT, FB, and GC conceived and designed the study and supervised the study. AB, EM, VW, RK, and FB performed experiments. JB, SMan, EM, LP, and SMar performed laboratory analyses. MB performed statistical analyses of the generated datasets. AB and FB wrote the manuscript. VW, CT, VP, and FB assisted in the experimental design and preparation of the manuscript. All authors contributed to the article and approved the submitted version.
The present work was developed within the framework of the Coordinated Research Project (N° D32033), Irradiation of Transboundary Animal Disease (TAD) Pathogens as Vaccines and Immune Inducers, funded by IAEA.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
The authors are grateful to the Sebastien Deville of SEPPIC, the Air Liquide Healthcare Specialty Ingredients for kindly providing us with IMS1313 adjuvant, and to Francesca Ellero for proofreading and language editing of the manuscript.
The Supplementary Material for this article can be found online at: