Conceived and designed the experiments: YB JP. Performed the experiments: YB HK KH TH WX DO JP. Analyzed the data: YB JP. Contributed reagents/materials/analysis tools: YB JP. Wrote the paper: YB JP.
The authors have declared that no competing interests exist.
Triple reassortant (TR) H3N2 influenza viruses cause varying degrees of loss in egg production in breeder turkeys. In this study we characterized TR H3N2 viruses isolated from three breeder turkey farms diagnosed with a drop in egg production. The eight gene segments of the virus isolated from the first case submission (FAV-003) were all of TR H3N2 lineage. However, viruses from the two subsequent case submissions (FAV-009 and FAV-010) were unique reassortants with PB2, PA, nucleoprotein (NP) and matrix (M) gene segments from 2009 pandemic H1N1 and the remaining gene segments from TR H3N2. Phylogenetic analysis of the HA and NA genes placed the 3 virus isolates in 2 separate clades within cluster IV of TR H3N2 viruses. Birds from the latter two affected farms had been vaccinated with a H3N4 oil emulsion vaccine prior to the outbreak. The HAl subunit of the H3N4 vaccine strain had only a predicted amino acid identity of 79% with the isolate from FAV-003 and 80% for the isolates from FAV-009 and FAV-0010. By comparison, the predicted amino acid sequence identity between a prototype TR H3N2 cluster IV virus A/Sw/ON/33853/2005 and the three turkey isolates from this study was 95% while the identity between FAV-003 and FAV-009/10 isolates was 91%. When the previously identified antigenic sites A, B, C, D and E of HA1 were examined, isolates from FAV-003 and FAV-009/10 had a total of 19 and 16 amino acid substitutions respectively when compared with the H3N4 vaccine strain. These changes corresponded with the failure of the sera collected from turkeys that received this vaccine to neutralize any of the above three isolates
Type A influenza viruses belong to the family
Although the HA protein is an important target of the host immune response and subtype specific anti-HA antibodies usually provide protection against infection with viruses of same HA subtype
As a result of antigenic shift or drift, influenza viruses with novel combinations of gene segments or point mutations have been isolated from various animal species
Swine influenza viruses have received increased attention in recent years from both the veterinary and public health authorities because of pigs being viewed as potential “mixing vessels” for the generation of novel viruses. In North America, the viruses that have been responsible for outbreaks in swine since 1998 have changed dramatically from the viruses that were responsible for outbreaks in the previous 70 years
This situation changed in 1998 when a severe outbreak of swine influenza occurred in North Carolina followed by additional outbreaks in Minnesota, Iowa and Texas
Approximately 45% of the turkey production in Canada is located in Ontario, often in areas that are densely populated with both swine and turkey farms. Since 2005, drops in egg production in breeder turkeys attributed to and coinciding with the circulation of TR H3N2 viruses in adjacent pig farms
In a previous study
This study involved a field outbreak of influenza A virus in turkeys. The owners of the animals associated with this study have read and approved this manuscript but wish to remain anonymous. Virus isolation procedures involving embryonating chicken eggs were in compliance with Canadian Council for Animal Care guidelines. The chicken embryos were inoculated at 9 days of gestation and fluids harvested 5 days afterwards.
All three diagnostic submissions were comprised of tracheal and cloacal swabs. They originated from three geographically separate breeder turkey farms owned by the same company in Southern Ontario that had experienced significant drops in egg production. Two of three flocks were vaccinated with an inactivated A/Mallard Duck/MN/79/79 (H3N4) oil emulsion vaccine and had detectable antibody titers before exposure to field virus. The original samples concerning this study were collected by the company veterinarian and submitted to the provincial veterinary diagnostic laboratory (Animal Health Laboratory, University of Guelph). Diagnostic test results along with the samples were forwarded to the National Centre for Foreign Animal Disease (NCFAD) as part of a federally mandated disease surveillance program. Additional work was carried out at the company's request to determine the reason for vaccine failure. The owners of the birds provided NCFAD with the convalescent sera collected from sick turkeys and the sera collected from H3N4 vaccinated turkeys prior to field virus as part of a follow up study aimed at determining the reason for vaccine failure.
Reference antisera to A/Perth/16/2009 (H3N2) produced in ferrets was carried out under Animal Use Document H-07-13 “Preparation of reference antisera to various strains of live human influenza virus in ferrets” approved by The Canadian Science Centre for Human & Animal Health Animal Care Committee in compliance with Canadian Council for Animal Care guidelines. The remaining antisera were prepared under Animal Use Document C-08-002 “Production of antisera to avian influenza viruses and avian paramyxoviruses” approved by the same Animal Care Committee.
The chicken red blood cells are obtained on a weekly basis from the Canadian Food Inspection Agency's Ontario Laboratory Fallowfield specific-pathogen-free flock under animal use document ACC 11-03 “Blood collection from farm animals”, that was approved by the institutional Animal Care Committee in compliance with Canadian Council for Animal Care guidelines. The turkey red blood cells were purchased from LAMPIRE Biological Laboratories, Inc. P.O. Box 270 Pipersville, PA, USA.
On February 24, 2011 tracheal and cloacal swabs from a breeder turkey flock of 15,000 that was exhibiting a sudden drop in egg production with no other apparent clinical signs were submitted to the Animal Health Laboratory (AHL), University of Guelph. Turkey hens from this flock had not been vaccinated with the inactivated H3N4 vaccine mentioned above. A H3N2 subtype influenza A virus was identified by molecular means and on February 25 tracheal and cloacal swabs were forwarded to the National Centre for Foreign Animal Disease (NCFAD), Winnipeg for virus isolation and further characterization. Sixteen convalescent serum samples were submitted at a later time point as part of the follow up investigation.
On June 6, 2011, 34-week-old turkey hens from a different geographical location exhibited a 10% drop in egg production with no other apparent clinical signs. A H3 subtype influenza A virus was identified by molecular means at AHL, University of Guelph and tracheal and cloacal swabs were forwarded to NCFAD, Winnipeg for confirmation and further characterization. These turkeys had been previously immunized with an inactivated H3N4 vaccine. Follow up serum samples were not obtained from birds in this flock.
On June 17, 2011, a flock from a third geographical location exhibited a 20% drop in egg production with no other clinical signs. These turkeys had received the inactivated H3N4 vaccine. Tracheal and cloacal swabs were forwarded to NCFAD on June 20th for confirmation. Nineteen serum samples collected before the turkeys exhibited the drop in egg production were submitted as part of a follow up investigation.
Tracheal and cloacal swab specimens were clarified by centrifugation, 50 µl of sample was spiked with an exogenous internal control (for evaluating nucleic acid extraction efficiency and presence of PCR inhibitors during RT-PCR) and RNA was extracted with the MagMAX™-96 Total RNA Isolation Kit using the MagMax 96-well robotic system (Applied Biosystems/Ambion, Austin, Texas).
Total RNA extracted from the swab specimens were tested using the M1 gene specific real-time reverse transcription polymerase chain reaction (RRT-PCR) assay
Virus isolation was carried out by inoculating the allantoic cavity of 9-day-old specific-pathogen-free (SPF) embryonating chicken eggs with clarified and antibiotic treated swab samples. Embryos were monitored daily for mortality. Amnio-allantoic fluid (AAF) from dead embryos as well as from embryos at the end of 1st and 2nd passages were harvested and tested for the presence of hemagglutinating agents with chicken red blood cells (CRBC). The AAF were also tested for the presence of influenza A nucleic acids by a real-time RT-PCR assay as described above to exclude presence of influenza A viruses that did not hemagglutinate CRBC. All submissions underwent up to two passages before being considered negative.
Hemagglutination and hemagglutination inhibition (HI) tests were carried out using standard procedures. For the hemagglutination test, AAF was tested for the presence of hemagglutinating agents using CRBC or turkey red blood cells (TRBC). Antigenic characterization of the new isolates was performed by HI assay using a panel of reference antisera prepared against the 16 known HA subtypes of influenza A viruses. Two fold serial dilutions of each reference antiserum were mixed with 4 HA units of each virus, followed by the addition of 0.5% (v/v) suspensions of CRBC or TRBC. The reciprocal of the highest dilution of serum that completely inhibited hemagglutination was considered the HI titre. The reference antiserum that produced the highest HI titer indicated the HA subtype of the isolate.
Immune plaque reduction virus neutralization (IPRVN) assay was carried out using MDCK cells grown overnight to confluency in 96-well tissue culture plates (Corning, USA). Virus neutralization was carried out using a constant amount of H3N2 virus (100 plaque forming units) mixed with equal volumes of 2-fold serial dilutions (starting 1∶20) of convalescent sera collected from diseased turkeys, field sera collected from turkeys immunized with the H3N4 vaccine prior to exposure, as well as a panel of reference H3 antisera. After 1 hr of incubation at 37°C, the virus/antisera mixtures were applied to the MDCK cell monolayers and incubated for an additional 1 hr at 37°C. The virus/antiserum mixture was then replaced with DMEM containing 0.2% (w/v) bovine serum albumin and 1.5% carboxymethyl cellulose (Sigma). The cells were incubated at 37°C in a humidified atmosphere of 5% CO2 for 48 hrs after which they were fixed in 10% formalin solution in PBS. Cells were then permeablized with 20% acetone in PBS, washed with PBS-Tween and then primed with anti-influenza nucleoprotein monoclonal antibody
The following viruses were used in the IPRVN assay: A/Turkey/ON/FAV-003/2011 (H3N2), A/Turkey/ON/FAV-009/2011 (H3N2), A/Mallard/QC/2323-66/2006 (H3N2), A/Duck/BC/7846/2006 (H3N8) and A/Turkey/BC/1529-3/2005 (H3N2) a TR virus isolated from domestic turkeys. Reference antisera raised against A/Turkey/BC/1529-3/2005 (H3N2), A/Duck/BC/7846/2006 (H3N8) and A/Perth/16/2009 (H3N2) (donated by Dr Yan Li, National Microbiology Laboratory, Public Health Agency of Canada) along with field serum samples collected from turkeys that were vaccinated with A/Mallard Duck/MN/79/79 (H3N4) prior to exposure to wild type H3N2 virus and convalescent serum samples collected from turkeys exposed to A/Turkey/ON/FAV-003/2011 (H3N2), but that were not vaccinated with the H3N4 vaccine were assessed by IPRVN assay.
Viral RNA was extracted from infectious AAF collected from embryonating chicken eggs. Total RNA was extracted as described above using the MagMAX™-96 Total RNA Isolation Kit. Full-length influenza A gene segments were RT-PCR amplified using universal influenza A primers
Phylogenetic analysis was performed as described previously
The amino acid sequences of the HA1 subunit of A/Turkey/ON/FAV-003/2011 (H3N2) and A/Turkey/ON/FAV-009/2011 (H3N2) were aligned with those of A/Mallard Duck/MN/1979 (H3N4) and A/Swine/ON/33853/2005 (H3N2) to identify amino acid substitutions within the five antigenic sites (A, B, C, D and E). The HA crystal structure of the H3 subtype influenza virus, A/duck/Ukraine/1963 (PDB 1MQL)
All 5 swab specimens that were submitted to NCFAD on February 25, 2011 tested positive with the influenza A matrix real time RT-PCR assay developed by USDA
Submission Number | Number of swabs | Matrix real time RT-PCR | Isolation in embryonating chicken eggs | |
Spackman | Modified | |||
FAV-003 | 5 | 5/5 | ND | 1/5 |
FAV-009 | 6 | 1/6 | 6/6 | 1/6 |
FAV-0010 | 7 | ND | 5/6 | 3/6 |
ND = not determined.
Turkeys in this barn had not been immunized with the inactivated H3N4 vaccine. Convalescent serum samples (n = 16) submitted from this farm tested positive on HI assay using a panel of reference H3 viruses. The results, which are summarized in
Serum | A/Tk/BC/01529/2005(H3N2) | A/Mallard/QC/2323-6/2006(H3N2) | A/Tk/ON/FAV003/2011(H3N2) | A/Tk/ON/FAV009/2011(H3N2) | A/Tk/ON/FAV0010/2011(H3N2) |
|
|||||
1 | Neg | Neg | Neg | Neg | Neg |
2 | 16 | 256 | 8 | 16 | 32 |
3 | Neg | 32 | Neg | Neg | 8 |
4 | 16 | 256 | 8 | 8 | 16 |
5 | Neg | 32 | Neg | Neg | 32 |
6 | 32 | 64 | 8 | 16 | 32 |
7 | Neg | 256 | 8 | 8 | 32 |
8 | Neg | 256 | Neg | 8 | 32 |
9 | Neg | 32 | Neg | Neg | 4 |
10 | Neg | 64 | Neg | Neg | Neg |
11 | 8 | 256 | 8 | 8 | 32 |
12 | Neg | 256 | 8 | 8 | 32 |
13 | Neg | 128 | Neg | Neg | 32 |
14 | Neg | 512 | Neg | 8 | 32 |
15 | Neg | 1024 | 8 | 16 | 64 |
16 | 8 | 16 | Neg | 4 | 32 |
17 | Neg | 32 | Neg | Neg | 8 |
18 | Neg | 32 | Neg | Neg | 16 |
19 | Neg | 64 | Neg | Neg | 16 |
|
|||||
20 | >8192 | 64 | >8192 | 256 | 1024 |
21 | >8192 | 512 | >8192 | 512 | 1024 |
22 | >8192 | 512 | >8192 | 512 | 4096 |
23 | 4096 | 128 | >8192 | 256 | 2048 |
24 | 2048 | 64 | 4096 | 512 | 4096 |
25 | >8192 | 4096 | 4096 | 512 | 2048 |
26 | >8192 | 4096 | 4096 | 1024 | 4096 |
27 | 4096 | 64 | >8192 | 256 | 1024 |
28 | >8192 | 2048 | 4096 | 1024 | 2048 |
29 | >8192 | 512 | 1024 | 512 | 4096 |
30 | 4096 | 128 | >8192 | 512 | 2048 |
31 | >8196 | 256 | >8196 | 512 | 1024 |
32 | 4096 | 128 | 4096 | 512 | 2048 |
33 | >8192 | 64 | >8192 | 512 | 2048 |
34 | >8192 | 2048 | >8192 | 1024 | >8192 |
35 | >8192 | 128 | >8192 | 512 | >8192 |
Sera from submission FAV-010 tested negative for antibodies to pH1N1 by hemagglutination-inhibition assay.
Three tracheal and 3 cloacal swab specimens from this farm tested negative using the influenza A matrix real time RT-PCR assay originally described by USDA
The swab specimens from this submission also tested negative when the real time RT-PCR assay targeting the matrix gene as originally described previously
For the single isolate from submission FAV-003, virus HA subtyping was done using molecular techniques as this virus did not hemagglutinate CRBC. For this purpose, the full HA gene segment was amplified, cloned and sequenced. NA subtyping was also determined by molecular means using the universal primers described by Hoffman et al.
The 8 gene segments for the three H3N2 isolates were amplified, cloned and sequenced. The sequence data for the H3N2 viruses isolated in this study were deposited in GenBank (FAV-003 acc # JN683626-33; FAV-009 acc # JN683634-41 and FAV-0010 acc # JN706697-704). Genetic relatedness of each gene segment from each of the isolates from the three farms was compared with other published influenza A sequences using the basic alignment search tool (BLAST) from the GenBank database. Based on the BLAST search, we were able to identify two genetically distinct H3N2 viruses. The single isolate from submission FAV-003 was identified as a triple reassortant virus containing gene segments of avian (PB2, PA), human (PB1, HA, NA) and swine (NP, M, NS) influenza virus origin. The two virus isolates from submissions FAV-009 and FAV-0010 contained a constellation of genes that has not been previously described. Gene segments PB2, PA, NP and M were from pandemic H1N1 2009 while the remaining gene segments (PB1, HA, NA and NS) originated from the TR H3N2 viruses that were isolated from pigs in the USA beginning in 1998 and in Canada beginning in 2005. The percentage of genetic relatedness of the three H3N2 isolates to other published influenza viruses in the NCBI database are summarized in
Segment | A/Tk/ON/FAV003/2011(H3N2) | A/Tk/ON/FAV009/2011(H3N2) | |
PB2 | NA | 99% A/swine/QC/1840-2/2009(H3N2) | 99% A/Ontario/315637/2009(H1N1) |
AA | 99% A/swine/QC/1840-2/2009(H3N2) | 99% A/Australia/24/2009(H1N1) | |
PB1 | NA | 99% A/swine/QC/1698-1/2009(H3N2) | 99% A/swine/Minnesota/66853/2006(H3N2) |
AA | 99% A/swine/QC/1840-2/2009(H3N2) | 99% A/turkey/BC/1529-3/2005(H3N2) | |
PA | NA | 99% A/swine/QC/1698-1/2009(H3N2) | 99% A/Ontaio/9739/2009(H1N1) |
AA | 99% A/swine/QC/1840-2/2009(H3N2) | 99% A/Canada-MB/RV2023/2009(H1N1) | |
HA | NA | 99% A/swine/QC/1698-2/2009(H3N2) | 99% A/swine/QC/1268883/2010(H3N2) |
AA | 98% A/swine/QC/1840-2/2009(H3N2) | 99% A/swine/QC/1268883/2010(H3N2) | |
NP | NA | 99% A/swine/QC/1698-1/2009(H3N2) | 99% A/swine/Taiwan/CH-1204/2004(H1N1) |
AA | 99% A/swine/QC/1697-1/2009(H3N2) | 98% A/Regensburg/D6/2009(H1N1) | |
NA | NA | 99% A/swine/QC/1698-2/2009(H3N2) | 98% A/Tk/BC/1529-3/2005(H3N2) |
AA | 99% A/swine/QC/1698-5/2009(H3N2) | 98% A/Ontario/RV1273/2005(H3N2) | |
M | NA | 99% A/swine/QC/1698-5/2009(H3N2) | 99% A/Taiwan/126/2009(H1N1) |
AA | 99% A/swine/QC/1840-2/2009(H3N2) | 100% A/Ontario/RV1527/2009(H1N1) | |
NS | NA | 98% A/swine/QC/1698-1/2009(H3N2) | 98% A/Tk/OH/313053/2004(H3N2) |
AA | 97% A/swine/QC/1698-1/2009(H3N2) | 97% A/Sw/N.Carolina/02023/2008(H1N1) |
Phylogenetic analysis of the HA and NA genes of the virus isolated from submission FAV-003 showed that they clustered with other TR H3N2 viruses that were isolated from pigs in Quebec in 2009 (
The HA (a) and NA (b) gene segments of the unique TR H3N2 viruses isolated from turkeys were compared with other TR H3N2 viruses from turkey (open circle), quail (open diamond) and pig (open triangles) that were previously sequenced by our laboratory
We compared the ability of the three newly isolated TR H3N2 viruses to react with different reference H3 antisera as well as field sera that had been collected from turkeys that were vaccinated with the inactivated H3N4 vaccine (submission FAV-0010) and convalescent serum collected from turkeys following H3N2 exposure (submission FAV-003). Polyclonal antisera raised against the TR H3N2 virus A/Turkey/BC/1529-3/2005 and the human seasonal H3N2 virus A/Perth/16/2009 were able to neutralize all three of the newly isolated viruses. In contrast, polyclonal antiserum raised against A/Duck/BC/7846/2006 (H3N8) poorly neutralized the TR H3N2 viruses that were isolated from the turkeys in this study. The sera collected from H3N4 vaccinated turkeys (submission FAV-0010) showed strong cross-neutralizing activity against A/Mallard/QC/2323-66/2006 (H3N2), but weak activity (titer = 1/40) against the three newly isolated TR H3N2 viruses as well as an earlier TR H3N2 virus from turkeys (A/ Turkey/BC/1529-3/2005). The convalescent sera collected from turkeys from submission FAV-003 cross-neutralized all three of the newly isolated TR H3N2 viruses as well as A/Mallard/QC/2323-66/2006 (H3N2) and A/Turkey/BC/1529-3/2005 (H3N2). This indicated that the flock might have been previously exposed to H3 viruses of avian and swine TR H3N2 origin. The virus neutralization titer was higher (>2560) against an isolate from the same farm (FAV-003). Results for HI assays (
Polyclonal Antisera | Viruses | ||||||
Dk/BC/7846/06 | Tk/BC/1529/05 | Dk/ON/05/00 | Perth/16/09 | Tk/ON/FAV003/11 | Tk/ON/FAV009/11 | Tk/ON/FAV010/11 | |
Rabbit 1anti- A/Dk/BC/7846/06(H3N8) | 2048 | 32 | ND | ND | 32 | 16 | 16 |
Rabbit 2anti- A/Tk/BC/1529/05(TR H3N2) | ND | 4096 | 32 | ND | 2048 | 128 | 256 |
Goat 3anti- A/Dk/ON/05/00(TR H3N2) | ND | 32 | 2048 | ND | 32 | 16 | 16 |
Ferret 4anti- A/Perth/16/09(H3N2) | ND | 128 | ND | 640 | 512 | 256 | 128 |
Negative Rabbit Serum | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
PolyclonalAntisera | Viruses | ||||
A/Tk/BC/15293/05 (TR H3N2) | A/Dk/BC/7846/06 (H3N8) | A/Tk/ON/FAV003/11 (H3N2) | Tk/ON/FAV009/11(H3N2) | A/Mal/QC/2323-66/06 (H3N2) | |
A/Tk/BC/1529-3/05(TR H3N2) | 2560 | ND | 640 | 640 | 40 |
A/Dk/BC/7846/06(H3N8) | 80 | 2560 | 40 | 40 | 2560 |
A/Perth/16/09(H3N2) | 1280 | ND | 640 | 1280 | 40 |
FAV-010 (#1)H3N4 vaccinated | 40 | ND | 40 | 40 | 1280 |
FAV-010 (#2)H3N4 vaccinated | 40 | ND | 40 | 40 | 2560 |
Convalescent serum FAV-003 (#1) | 2560 | ND | 2560 | 2560 | 1280 |
Convalescent serum FAV-003 (#2) | 1280 | ND | 1280 | 1280 | 2560 |
ND – Not Done.
To determine the level of genetic relatedness between the three H3N2 turkey isolates, the full hemagglutinin protein (HA0) and the 328 residues of the HA1 subunit were subjected to pairwise amino acid identity comparisons with the A/Mallard duck/79/79 (H3N4) vaccine strain
A/Tk/ON/FAV003/2011 | A/Tk/ON/FAV010/2011 | A/Sw/ON/33853/2005 | ||||
Isolate | HA0 | HA1 | HA0 | HA1 | HA0 | HA1 |
A/Tk/ON/FAV009/2011 | 94% | 91% | - | - | - | - |
A/Sw/ON/33853/2005 | 97% | 95% | 96% | 95% | - | - |
A/Tk/ON/FAV010/2011 | 94% | 91% | 99% | 99% | 96% | 95% |
A/M.duck/MN/79/1979 | ND | 79% | ND | 80% | ND | 80% |
ND = Not done [full HA sequence of A/Mallard duck/MN/79/1979 (H3N4) was not available].
The predicted amino acid identities of the HAl subunit (excluding the signal peptide) between the H3N4 vaccine strain and FAV-003 was 79% and between the H3N4 vaccine strain and FAV-009 or FAV-0010 was 80%. By comparison, the predicted amino acid identity between a prototype H3N2 (A/Sw/ON/33853/2005) virus from cluster IV and the three turkey isolates was 95%. When the turkey isolates were compared with each other, the predicted amino acid identity decreased to 91%.
To determine whether the amino acid changes occurred in any of the previously identified
(I) Side view of a HA monomer in cartoon format with major antigenic sites A, B, C and D shown in spheres (II, III, IV, and V??amino acid changes identified at the major antigenic sites. The locations of changed amino acids are indicated and colored in red. (VI, VII, VIII, IX, and X) Back view of panels I, II, III, IV, and V, respectively. The view in panels I to V is rotated 180° along Y-axis.TK/ON/FAV-003/2011 vs. DK/MN/1979 (II and VII); TK/ON/FAV-009/2011 vs. DK/MN/1979 (III and VIII); TK/ON/FAV-003/2011 vs. SW/ON/33853/2005 (IV and IX); TK/ON/FAV-009/2011 vs. SW/ON/33853/2005 (V and X).
Amino acids of the HA1 subunit of the three unique turkey isolates, the duck H3N4 vaccine strain, a prototype cluster IV TR H3N2 virus (A/SW/ON/33853/2005) and phylogenetically related isolates A/Sw/QC/1265553/2010 (H3N2) and A/SW/QC/1698-1/2009 (H3N2) were aligned. Residues shown in red, green, blue and purple represent previously identified antigenic sites A, B, C and D respectively. Potential glycosylation sites are underlined.
To predict N-linked glycosylation sites (Asn-X-Ser/Thr, where X is any amino acid except Pro), we used the NetNGlyc 1.0 Server. Based on this, the H3N4 vaccine strain had 5 glycosylation sites at positions 8, 22, 38, 165 and 285 as described previously
In a previous report, we isolated pandemic H1N1 2009 virus from a breeder turkey flock that exhibited respiratory illness and a drop in egg production
Phylogenetic analysis of the HA and NA genes from FAV-009 and FAV-0010 showed that they are closely related to A/Swine/QC/126553/2010 (H3N2) and A/Swine/QC/1268883/2010 (H3N2) which were isolated from pigs with respiratory illness in Quebec. The latter H3N2 isolate from swine also has pandemic H1N1 internal genes
In contrast to the chicken reproductive tract that contains only α2-3-linked sialic acids receptors, turkeys may contain both α2-3-linked and α2-6-linked sialic acids receptors in their reproductive tract which could be involved in the attachment and replication of human and swine-like influenza viruses
Despite the fact that pigs are normally viewed as “mixing vessels” for the generation of reassortant influenza A viruses, we cannot exclude the possibility that the reassortment between TR H3N2 and pH1N1 took place in turkeys, even though sera from submission FAV-0010 gave negative HI results when using pandemic H1N1 antigen (data not shown). Nevertheless, the results described here should be of concern, considering the reassortment capacity of this virus and the susceptibility of turkeys to influenza viruses of H1 to H16 subtypes. According to Nobusawa et al.
The hemagglutinin protein of H3 influenza viruses has accumulated a number of glycosylation sites during its evolution over the past 40 years
In Canada, the only approved vaccine for turkeys against H3N2 viruses is the inactivated H3N4 vaccine made from a strain isolated from a mallard duck in 1979. Although, two of the breeder turkey flocks in this study were immunized with this vaccine, the field strains described here were able to infect the turkeys and cause a loss in egg production. The fact that the sera collected from H3N4 vaccinated turkeys were not able to neutralize any of the newly isolated H3N2 viruses described in this study confirms the poor efficacy of this vaccine. We believe the vaccine failure was associated with amino acid substitutions in the globular head domain of the HA1 subunit which contains the immunodominant antigenic sites. The percent amino acid identity of the HA1 between the H3N4 vaccine strain and FAV-003 was 79% and with FAV-009 or FAV-0010 – 80%.
Previous studies have also shown that mutations in the 5 immunodominant antigenic sites located on the globular head of the HA1 subunit play a key role in virus escape from host immune pressure as a result of accumulated conformational changes
In both humans and domestic animals influenza variants frequently emerge as a result of point mutations in the HA gene, resulting in new variants that escape the host immune response. The vaccine failure described in this study is not surprising; selection of viruses for animal influenza vaccines should be based on the results of recent epizootologic, virologic and immunologic surveillance results. Two of the H3N2 isolates characterized in this study contained a unique combination of genes derived from pandemic H1N1 (2009) which underscores the need for continuous surveillance and monitoring of the genetic changes of influenza A viruses circulating in domestic animals to not only aid in the selection of the most appropriate vaccine strains but to track the evolution of viruses that might pose new threats to human and animal health.
We would like to thank Dr Soren Alexandersen for critical review of the manuscript. We also would like to thank Dr Chris Kranendonk for help with specimen receiving, laboratory staff of the Animal Health Laboratory, University of Guelph and the Canadian Food Inspection Agency's Guelph District Office for sending the samples used in this report.