Equine influenza is one of the most economically important contagious respiratory diseases of horses (1,2). This disease is characterized by rapidly spreading signs that include pyrexia, depression, anorexia, a harsh dry cough, a nasal discharge, and secondary bacterial respiratory infections (3). Chronic sequelae that include bronchopneumonia and exercise-induced hemorrhage, have been identified (4). Outbreaks of equine influenza have been reported worldwide. Influenza is endemic in the horse populations of Europe and North America, and several outbreaks have been reported throughout Canada (4–7). In the most recent epidemiological study done in Ontario, equine influenza virus was found to be the causal agent in 56% of 23 outbreaks of acute viral respiratory disease investigated over a 2-year period (7). In this study, the vaccination rate was quite low, as only 36% of the horses had been given an equine influenza vaccine at least once during the previous year. This is far below the 70% vaccination rate that has been suggested to prevent epidemics of influenza in horses (1).
Over the years, strain variants of equine influenza virus have emerged due to antigenic shift and antigenic drift (8,9). In an attempt to monitor the evolution of equine influenza virus in Canada, the Diagnostic Veterinary Virology Laboratory (DVVL) of the Faculté de médecine vétérinaire (FMV) of the Université de Montréal has isolated 7 equine influenza virus strains from clinically affected animals between 2003 and 2006. These strains were isolated from embryonated eggs inoculated with nasal swabs. Five of these strains were recovered from horses in Quebec and 2 were recovered from horses in Alberta (Table 1). The subtype (H3N8 or equine-2; and H7N7 or equine-1) of each influenza virus isolate was determined with an hemagglutination-inhibition test, using specific antibodies for each subtype. All isolated strains were subtype H3N8, which was not a surprising result, since H7N7 virus has not been isolated for over 35 y in Canada.
Table 1.
Canadian equine H3N8 influenza virus isolates and H3N8 vaccine strains description
| Strain name | Geographical origin | Year of isolation | Related vaccine trademark name |
|---|---|---|---|
| Saskatoon/90 | Canada/Saskatoon | 1990 | n.a. |
| Ontario/97 | Canada/Ontario | 1997 | n.a. |
| Ontario/01 | Canada/Ontario | 2001 | n.a. |
| Ontario/02 | Canada/Ontario | 2002 | n.a. |
| Manitoba/43433/01 | Canada/Manitoba | 2001 | n.a. |
| Canada/2447/03 | Canada/Quebec | 2003 | n.a. |
| Canada/3112/04 | Canada/Quebec | 2004 | n.a. |
| Canada/2/04 | Canada/Quebec | 2004 | n.a. |
| Quebec/05 | Canada/Quebec | 2005 | n.a. |
| Quebec/06 | Canada/Quebec | 2006 | n.a. |
| Alberta/1/06 | Canada/Alberta | 2006 | n.a. |
| Alberta/2/06 | Canada/Alberta | 2006 | n.a. |
| Kentucky/91 | United States/Kentucky | 1991 | FluAvert® |
| Kentucky/93 | United States/Kentucky | 1993 | Prestige®1 |
| Kentucky/95 | United States/Kentucky | 1995 | Calvenza®2 |
| Newmarket/2/93 | United Kingdom/Newmarket | 1993 | Calvenza®2 |
| Kentucky/97 | United States/Kentucky | 1997 | Fluvac Innovator®3 |
| Kentucky/98 | United States/Kentucky | 1998 | Flumune® |
n.a. — None applicable
The Prestige® vaccine contains moreover one H7N7 strain, Pennsylvania/63
The Calvenza® vaccine contains moreover one H7N7 strain, Newmarket/77
The Fluvac Innovator® vaccine contains moreover one H7N7 strain, Prague/56
After isolation, either the viruses or the hemagglutinin (HA) and the neuraminidase (NA) RNA virus sequences were sent for genomic characterization at the Animal Health Trust (AHT) laboratory in UK, a World Organization for Animal Health (OIE) reference laboratory for equine influenza and equine rhinopneumonitis. A phylogenetic tree was built by the AHT to determine the relatedness of Canadian isolates to worldwide reference strains. Moreover, equine influenza virus vaccine strains currently used in Canada and Canadian equine influenza virus isolates recovered prior to 2003 were included in the phylogenetic analysis. The HA gene was targeted in the phylogenetic analysis, because its mutation rate has been demonstrated to be greater than that of other viral genes and because it is one of the main targets of the humoral protective immune response (10).
As shown in Figure 1, the 7 Canadian equine influenza viruses isolated since 2003 by the DVVL are variant-American (Florida) lineage strains. The HA1 amino acids (aa) identity between the recent 7 Canadian isolates was 98.21% to 100% (data not shown). Interestingly, the HA1 aa identity between recent and older (2002 to 1997) Canadian isolates was 95.24% to 99.4% (data not shown), confirming the existence of a constant evolution of the genetic code of equine influenza, as previously reported (8). Several vaccines are currently used in Canada, or will be in the near future. Consequently, we were interested in evaluating the genetic relationship between recent Canadian equine influenza isolates and the equine influenza vaccine strains that are listed in Table 1.
Figure 1.
Phylogenetic analysis of the HA1 gene nucleotide sequence of recent Canadian equine influenza virus isolates. Original names of Canadian isolates and vaccine strains, and their origin and date of isolation are given in Table 1. Canadian isolates are identified with an arrow and vaccine strains used in Canada are identified with an asterix. The numbers at the main partitions of the phylogenetic tree are maximum likelihood (ML) bootstrap values. Horizontal branch lengths are drawn to scale with the bar indicating 0.004 nucleotide substitutions per site on HA1.
As illustrated in Figure 1, the Kentucky/97 vaccine strain (Genbank accession number AF197249) is the virus most genetically related to recent Canadian equine influenza field isolates and the only commercially available vaccinal strain that belongs to the same lineage, the variant-American (Florida). Indeed, the HA1 aa identity between the Kentucky/97 vaccine strain and recent Canadian isolates was established to be between 97.02% to 98.21% (Figure 1, data not shown). The 2nd most related vaccine strain to recent Canadian isolates was Kentucky/93, a vaccinal strain belonging to the mainstream American (Kentucky) lineage that shares an HA1 aa identity with recent Canadian isolates of 96.13% to 97.32% (Figure 1, data not shown). The HA1 aa identities of other vaccine strains with recent Canadian strains were established to be 93.45% to 96.73% (Figure 1, data not shown).
None of the recent Canadian strains belongs to the Eurasian lineage. To our knowledge, Saskatoon/90 was the only strain isolated in North America that belongs to this lineage. This suggests that this strain did not spread out in the North American horse population and probably became extinct.
It is interesting to point out that the Kentucky/97 strain contained in the Fluvac Innovator® commercial vaccine is the most closely genetically related vaccine strain to recent Canadian influenza virus isolates, based on HA1 nucleotide and protein comparison analysis (Figure 1, data not shown). How important this antigenic closeness is with regard to field vaccine efficacy is difficult to assess. It was previously reported that the antigenic disparity between the circulating virus and the vaccine strain is a significant factor for vaccine failure (1,11). Antigenicity and sequencing studies suggested that significant antigenic drift contributed to reduce vaccine efficacy; therefore, it was recommended to update vaccines with the most representative recent virus strain approximately every 4 y (12). However, what constitutes a significant antigenic drift is not clearly defined. Yates and Mumford (13) reported that the ability of a vaccine to reduce virus excretion correlates directly to the antigenic relatedness of the vaccine strain with the challenge strain. Lai et al (11) came to the conclusion that the 2 phylogenetic lineages, the mainstream American (Kentucky) and the variant-American (Florida), are antigenically distinct. In their experiment, horse sera reacted differentially to viruses of 1 lineage, with a higher reactivity towards the homologous virus and viruses in the homologous lineage (11). In an experimental challenge study, the less efficient vaccine was the vaccine made with the most distant strain from the challenge strain (13). Field outbreaks were reported in the UK and in Italy in horses vaccinated with a mainstream American (Kentucky) strain and subsequently found to be infected with a variant-American (Florida) strain (3,14). In the Italy outbreak, the field infecting strain displayed 9 aa changes in the HA1 subunit protein with respect to the vaccine strain.
In the present report, the most closely related vaccine strain (Kentucky/97) displayed 6 to 10 aa changes in the HA1 subunit protein with respect to recently isolated Canadian influenza strains. On the other hand, the 2nd most related vaccine strain (Kentucky/93) and the least related vaccine strain (Newmarket/2/93) displayed 9 to 13 and 19 to 23 aa changes, respectively, in the HA1 subunit protein with respect to recent Canadian strains. As previously reported (14), use of an equine influenza vaccine strain closely related to field isolates may not guarantee the efficacy of protection induced in the vaccinated animals, but, the probability of inducing a protective immune response and reducing clinical disease is higher when the vaccine strain used is closely related to field isolate.
Footnotes
This work was subsidized by Wyeth Animal Health and the Horserace Betting Levy Board (HBLB).
References
- 1.van Maanen C, Cullinane A. Equine influenza virus infections: An update. Vet Q. 2002;24:79–94. doi: 10.1080/01652176.2002.9695127. [DOI] [PubMed] [Google Scholar]
- 2.Timoney PJ. Equine influenza. Comp Immunol Microbiol Infect Dis. 1996;19:205–211. doi: 10.1016/0147-9571(96)00006-9. [DOI] [PubMed] [Google Scholar]
- 3.Newton JR, Daly JM, Spencer L, Mumford JA. Description of the outbreak of equine influenza (H3N8) in the United Kingdom in 2003, during which recently vaccinated horses in Newmarket developed respiratory disease. Vet Rec. 2006;158:185–192. doi: 10.1136/vr.158.6.185. [DOI] [PubMed] [Google Scholar]
- 4.Carman S, Rosendal S, Huber L, et al. Infectious agents in acute respiratory disease in horses in Ontario. J Vet Diagn Invest. 1997;9:17–23. doi: 10.1177/104063879700900104. [DOI] [PubMed] [Google Scholar]
- 5.Laporte V, Lavoie JP. Influenza équin au Québec: l’épidémie de 2001. Le Méd Vét du Québec. 2004;34:227, 242, 245. [Google Scholar]
- 6.Sherman J, Mitchell WR, Martin SW, Thorsen J, Ingram DG. Control of equine respiratory disease on Ontario Standardbred racetracks. Can Vet J. 1977;18:154–158. [PMC free article] [PubMed] [Google Scholar]
- 7.Mendes AD. Master thesis. University of Guelph, Guelph; Ontario, Canada: 2006. Surveillance of equine influenza virus and cause specific morbidity rate in young horses in Ontario. [Google Scholar]
- 8.Oxburgh L, Berg M, Klingeborn B, Emmoth E, Linné T. Evolution of H3N8 equine influenza virus from 1963 to 1991. Virus Res. 1994;34:153–165. doi: 10.1016/0168-1702(94)90097-3. [DOI] [PubMed] [Google Scholar]
- 9.Lai AC, Chambers TM, Holland RE, Jr, et al. Diverged evolution of recent equine-2 influenza (H3N8) viruses in the Western Hemisphere. Arch Virol. 2001;146:1063–1074. doi: 10.1007/s007050170106. [DOI] [PubMed] [Google Scholar]
- 10.Paillot R, Hannant D, Kydd JH, Daly JM. Vaccination against equine influenza: Quid novi? Vaccine. 2006;24:4047–4061. doi: 10.1016/j.vaccine.2006.02.030. [DOI] [PubMed] [Google Scholar]
- 11.Lai AC, Rogers KM, Glaser A, Tudor L, Chambers T. Alternate circulation of recent equine-2 influenza viruses (H3N8) from two distinct lineages in the United States. Virus Res. 2004;100:159–164. doi: 10.1016/j.virusres.2003.11.019. [DOI] [PubMed] [Google Scholar]
- 12.Daly JM, Lai AC, Binns MM, Chambers TM, Barrandeguy M, Mumford JA. Antigenic and genetic evolution of equine H3N8 influenza A viruses. J Gen Virol. 1996;77:661–671. doi: 10.1099/0022-1317-77-4-661. [DOI] [PubMed] [Google Scholar]
- 13.Yates P, Mumford JA. Equine influenza vaccine efficacy: The significance of antigenic variation. Vet Microbiol. 2000;74:173–177. doi: 10.1016/s0378-1135(00)00177-2. [DOI] [PubMed] [Google Scholar]
- 14.Martella V, Elia G, Decaro N, et al. An outbreak of equine influenza virus in vaccinated horses in Italy is due to an H3N8 strain closely related to recent North American representatives of the Florida sub-lineage. Vet Microbiol. 2007;121:56–63. doi: 10.1016/j.vetmic.2006.11.017. [DOI] [PubMed] [Google Scholar]