VACCINATION AGAINST EQUINE INFLUENZA: QUID NOVI?
R. Paillot *, D. Hannant, J.H. Kydd, J.M. Daly
Animal Health Trust, Centre for Preventive Medicine, Lanwades Park, Newmarket, Suffolk
CB8 7UU, UK
Corresponding author. Tel.: +44 8700 502460x1247; fax: +44 8700 502461.
E-mail address: firstname.lastname@example.org (R. Paillot).
R. Paillot et al. Vaccine 24 (2006) 4047–4061
Equine influenza virus is a leading cause of respiratory disease in the horse. Equine influenza vaccines containing inactivated virus were first developed in the 1960s. Despite their intensive use, equine influenza outbreaks still continue to occur and therefore new strategies of vaccination are necessary to improve vaccine efficacy. Numerous methods of vaccination have been evaluated and commercialised in the horse, the most recent being the cold-adapted influenza virus and poxvirus-based vaccines. As a large animal model, the horse is also a useful species in which to evaluate the potential of new generations of influenza vaccine such as live-attenuated influenza virus engineered by reverse genetics. This report details the equine immune responses conferring protection against influenza. It then undertakes a selective review of different strategies of vaccination against equine influenza that have been developed over the last two decades and discusses factors that may influence the efficacy of vaccination. Finally it outlines progress in the development of a novel vaccination strategy against equine influenza using reverse genetics.
© 2006 Elsevier Ltd. All rights reserved.
Keywords: Influenza; Vaccine; Horse species
1. Introduction....................................................................................................... 4048
2. Equine influenza virus............................................................................................. 4048
2.1. Viral structure............................................................................................... 4048
2.2. Disease .................................................................................................... 4048
2.3. Epidemiology in horses...................................................................................... 4049
2.3.1. EIV subtypes ....................................................................................... 4049
2.3.2. Interspecies transmission............................................................................. 4049
2.3.3. Strain selection for vaccination....................................................................... 4049
2.4. Protective immune responses................................................................................. 4049
2.4.1. Humoral immune esponse........................................................................... 4050
2.4.2. Cellular immune response............................................................................ 4051
3. Vaccines against equine influenza virus..............................................................4051
3.1. ‘Dead’ vaccines............................................................................................. 4051
3.1.1. Inactivated whole influenza virus vaccines...................................................4051
3.1.2. Subunit vaccines.................................................................................... 4053
3.1.3. DNA vaccines...................................................................................... 4053
3.2. Live virus/vector vaccines.................................................................................... 4055
3.2.1. Live attenuated influenza virus vaccine................................................................ 4055
3.2.2. Virus-based vector vaccines.......................................................................... 4056
3.2.3. Live attenuated influenza virus designed by reverse genetics.............................4057
3.3. Effectof maternal antibodies and age on equine influenza vaccination................ 4057
4. Conclusion ....................................................................................................... 4058
Equine influenza A viruses (EIV; H7N7 and H3N8 subtypes) are a leading cause of respiratory disease in the horse. Although the H7N7 influenza virus subtype has not been detected recently, the H3N8 subtype of equine influenza virus has not been controlled successfully by vaccination and remains today a serious threat to horse welfare and an economic problem for the horse industry. Historically, equine influenza vaccines were mainly composed of whole inactivated viruses, which provide protection against influenza through the induction of a short-lived humoral immunity. This is in contrast to immunity stimulated by natural infection, which is more robust and longer lived due to the stimulation of both humoral and cellular immune responses. The development of new strategies of vaccination that mimic more closely the stimulation of the immune system induced by EIV infection, has been the focus of EIV vaccine development in the last two decades. Thus, modern vaccines composed of either
live attenuated influenza virus, DNA plasmids or poxvirusvectors coding for influenza virus proteins were developed and some have been commercialised. Today, a new approach to EIV vaccination using live-attenuated influenza virus engineered by reverse genetics, is also under development. This report will describe the equine influenza virus, its pathogenesis, epidemiology and immunology in the horse and then review different vaccination approaches.
2. Equine influenza virus
2.1. Viral structure
Influenza viruses belong to the family orthomyxoviridae and are classified as A, B and C based on antigenic differences in their nucleoprotein (NP) and matrix (M1) protein. EIV is a type A influenza virus similar to type A human influenza virus. Type A influenza viruses contain segmented RNA (about 13.6 kb) consisting of eight linear, single stranded genomic fragments of negative polarity. Six segments code for single proteins haemagglutinin (HA), neuraminidase (NA), nucleoprotein (NP), three subunits of viral polymerase (PA, PB1 and PB2), one segment codes for matrix 1 and 2 (M1 and M2) and one segment codes for nonstructural protein 1(NS1) and nuclear export protein (NEP) [1,2]. In some strains, the +1 reading frame of PB1 codes for the PB1-F2 protein .
Equine influenza viruses may be spherical (with a diameter of 80–120 nm) or filamentous. The virus is composed of a bilayer lipidic envelope containing HA (involved in binding of the virus to host cells) and NA glycoproteins (controls the release of virus from infected cells). The envelope is supported by adherent matrix proteins (M1 and M2) and surrounds the nucleocapsid, which consists of the genomic RNA segments, enclosed within a helicallyarranged NP. PB1, PB2 and PA, which form the RNA-dependent RNA polymerase complex, are associated with the nucleocapsid. In human influenza virus, NS1 is involved in RNA transport, splicing and translation. NEP participates in the export of virion RNAs out of the nucleus .
Like human influenza virus, EIV is highly contagious and is contracted by inhalation. EIV infects and replicates in the ciliated epithelial cells of the upper and lower respiratory tract, causing de-ciliation of large areas within 4–6 days. After an incubation period of 1–3 days, the first clinical sign of equine influenza is an elevation of body temperature (up to 41 .C), which can last 4–5 days. A harsh dry cough, which releases large quantities of virus, contributes to the spread of infection. Coughing is commonly accompanied by serous and/or mucopurulent nasal discharges. Other clinical signs of disease are myalgia, inappetance and enlarged submandibular lymph nodes (reviewed in  and ). However, the severity of the disease depends principally on the strain of EIV and the immune status of the individual. Infected horses can develop bronchitis, bronchiolitis, and interstitial pneumonia associated with congestion and alveolar oedema.
Cases of laryngitis, tracheitis and myocarditis have also been reported. Clinical signs wane 2 weeks post-infection, unless there are complications due to secondary bacterial infections
[5,6]. In some rare cases (e.g. old or stressed horses), EIV may penetrate the basal membrane of infected respiratory epithelial cells, a phenomenon that could explain exceptional cases of myositis, myocarditis, oedema and encephalitis observed after influenza infection . As for human influenza, there is still some debate as to whether systemic signs result from systemic infection or immuno-pathogenesis induced by viral products or inflammatory cytokines. The severity of clinical signs of disease induced by EIV infection has been correlated to the local synthesis of interleukin-6 (IL-6) and interferon (IFN) . Equine influenza induces a high morbidity and mortality in foals, horses in poor health and donkeys. In adults, mortality is generally associated with secondary bacterial infections leading to pleuritis, pneumonia or purpura haemorrhagica.
2.3. Epidemiology in horses
2.3.1. EIV subtypes
Influenza A viruses are subtyped according to the antigenicity of HA and NA. Currently, 16 HA and 9 NA subtypes have been identified among influenza A viruses . All these subtypes are found in avian species. However, influenza viruses infecting mammals are limited to a few subtypes (reviewed in ). Two equine influenza virus subtypes have been recognised, the H7N7 subtype (equi-1) and the H3N8 subtype (equi-2). Due to genome segmentation, genetic reassortment can occur in cells simultaneously infected with two different subtypes of influenza A virus. Reassortment of the HA or NA genes is the basis for antigenic shift but this has not been observed in horses .
The H7N7 subtype was first isolated from horses in Czechoslovakia in 1956 (prototype strain: A/eq/Prague/1/56) . The last confirmed outbreak occurred in 1979 in Italy although this subtype was isolated in India in 1987  and Egypt in 1989 . Nevertheless, serological evidence of continued circulation of this virus subtype was reported in Central Europe and Asia as late as 1991 . The H7N7 equine influenza virus may still circulate in a subclinical form and persist at a very low level in some parts of the world.
The H3N8 subtype of EIV was first isolated in Miami in 1963 (prototype strain: A/eq/Miami/1/63) , and has caused outbreaks of disease (e.g. USA 1963; UK 1976; Europe and North America 1978–1981; South-Africa 1986; India 1987; China 1993–1994; UK 2003).
2.3.2. Interspecies transmission
Influenza A viruses are known to have evolved from an ancestral avian precursor into different lineages infecting specific species, including horses . In 1989, an equine influenza epidemic was reported in north-eastern China and resulted in up to 20% mortality in some herds. The virus causing this outbreak (A/eq/Jilin/89) was shown to be more closely related to avian H3N8 influenza virus than to contemporary equine influenza virus . Despite a successful cross-species transmission, this strain did not spread beyond or persist in China. However, this outbreak illustrates the role of birds, as a natural reservoir of influenza A virus, in the potential emergence of new strains of influenza virus able to infect mammals. Mammal-to mammal transmission of influenza A virus also occurs (e.g. swine to human). In recent years, influenza viruses isolated from dogs (e.g. greyhounds) suffering severe respiratory disease with a high percentage of mortality, were identified as closely related to equine influenza virus (H3N8) and could indicate a new case of inter-species transmission .
2.3.3. Strain selection for vaccination
For decades, commercial vaccines contained representative strains of equine H7N7 and H3N8 influenza virus. Although vaccination against H7N7 influenza virus was successful, H3N8 influenza virus-targeted vaccines have repeatedly failed to protect horses. This discrepancy has been attributed to differences in virulence between the two types of equine influenza virus. Antigenic drift of the H3N8 influenza virus also contributes to the reduced efficacy of vaccines [17, 18]. Phylogenetic analysis revealed that since the late 1980s, H3N8 equine influenza viruses have diverged into two different lineages (i.e. ‘American’ and ‘European’).
Antigenic differences between ‘European’ and ‘American’ H3N8 lineages are sufficient to compromise cross-protection [18,19] and as a result, current guidelines recommend that vaccines against equine influenza contain a representative of each (H3N8) lineage [20,21]. The current recommendation of the Expert Surveillance Panel on equine influenza is that vaccines contain an A/eq/South Africa/4/2003 (H3N8) like virus (American lineage) and an A/eq/Newmarket/2/93 (H3N8)-like virus (European lineage). The presence of H7N7 equine influenza virus strains (i.e. A/eq/Prague/56 or A/eq/Newmarket/77) in vaccines is not recommended any more.
Despite widespread compliance with OIE recommendations, outbreaks of equine influenza have continued to occur in Europe, America and other parts of the world since 1989 (e.g. Hong Kong in 1992; Dubai in 1995 and Philippines in 1997). In areas previously free of the disease, the appearance or introduction of EIV induced devastating outbreaks (e.g. South Africa 1986–1987, India 1987). Today, only Australia, New Zealand and Iceland are known to remain free from equine influenza.
2.4. Protective immune responses
The infection of epithelial cells with EIV is likely to stimulate an innate immune response in horses, as described in humans after influenza virus infection . However, this response has not been extensively studied in the horse. EIV infection has been associated with the local synthesis of IL6 and IFN, which is likely to be type I IFN (i.e. IFN/) (Fig. 1A) . Both IL-6 and IFN/. are pro-inflammatory cytokines. IL-6 is a cytokine showing a wide range of activities on B and T cells and is involved in the development of mucosal IgA responses . IFN/ presents a variety of immunomodulatory activities affecting both innate (e.g. enhancement of natural killer (NK) activity) and adaptive immunity (e.g. promotion of Th1 response). The activation of NK cells, another component of the innate response to influenza virus, is probably involved since the derivation in vitro of cells exerting a genetically non-restricted cytotoxic activity to EIV infected target cells has been reported .
Natural infection with EIV confers a long-term immunity to re-infection. Experimentally, ponies primed to EIV by a previous infection are protected against clinical signs and by reverse genetics).
Fig. 1. Immune response induced by infection with live EIV at the mucosal surface.
(A) Primary EIV infection (wild type, reassortant, attenuated or produced virus shedding when re-challenged. Complete clinical and virological immunity against infection with an homologous strain was shown to persist at least 32 weeks after the primary infection, and partial protection was still achieved after 1 year
Epithelial cells become infected inducing the local synthesis of IFN and IL-6 . The presentation of EIV antigen in lymphoid tissues induces the synthesis of serum or mucosal antibodies able to neutralise excreted virus, and stimulates the development of virus-specific CTL that lyse infected cells.
(B) Secondary EIV infection. Virus-specific antibodies act as a first line of defence in the case of re-infection with EIV . However, if the level of antibody is insufficient, epithelial cell infection could occur and induce the activation of memory responses (25)..
(Ab: antibody; APC: antigen presenting cells; B: B cell; m: memory; NALT/MALT: nasal/mucosal associated lymphoid tissue; Th: T helper lymphocyte; CTL: cytotoxic T lymphocytes; CTLp: precursor CTL.)
Both humoral and cellular immune responses have been shown to protect against EIV infection . These responses are likely to be initiated in the nasal associated lymphoid tissues (NALT) of the upper respiratory tract, which possess a specialised epithelium known as follicle-associated epithelium (FAE). FAE is characterised by membranous microvillus (M) cells that deliver virus antigens to T and B cells of the underlying lymphoid tissue .
2.4.1. Humoral immune response
HA and NA molecules are the main targets of the humoral immune response against influenza virus. In na¨ive ponies, experimental infection with EIV induces high levels of serum IgM (which generally decline within 50 days after infection) , and virus-specific IgA, IgGa and IgGb antibody in both serum and nasal secretions .
In mammals, nasal IgA may be very important for neutralising intracellular influenza virus, for preventing virus shedding following infection, and for cross-protection against antigenically different influenza viruses [29,30]. Therefore, a mucosal immune response is important to protect horses against EIV. In foals, IgA can be induced locally in mucosa or transported from the serum to nasal secretions in the case of dimeric IgA antibody . After experimental infection of ponies, serum and mucosal (i.e. nasopharynx and trachea) IgA rose during the second week following infection with EIV and declined quickly (only 20% of serum IgA remaining 2 months after the infection). The serum IgA anamnestic response is more rapid and persistent than the primary response [25, 28].
Complement fixing antibody levels measured by single radial hemolysis (SRH) also increased significantly in serum 7 days after a primary infection of ponies. However, the SRH antibody response was shown to decline rapidly (only 25% of the peak value 3 months after infection), and the majority of horses were negative for SRH antibodies 62 weeks after exposure to EIV . IgGa and IgGb isotypes are effective in both complement fixation and antibody-dependant cellular cytotoxicity assays in vitro [32, 33], but their effector function in vivo has not been determined. Serum virus-specific IgGa/b antibody also increased by 7 days post-infection with EIV but was shown to decline to pre-challenge levels over a period of 15 months .
Small antigenic changes in HA and NA epitopes (antigenic drift) allow the virus to evade the protective humoral immune system of the host . There is no cross-protection between antibodies to the H7N7 and H3N8 subtypes of equine influenza .
2.4.2. Cellular immune response
EIV infection induces a long-term immunity independent of circulating antibody. Ponies with low or undetectable titres of HA-specific antibodies were clinically and virologically protected from challenge infection more than 1 year after infection with EIV [25, 34]. In the absence of a detectable antibody response, it was likely that a cell-mediated immunity (CMI) was responsible for this protection. However, there are only a few reports on CMI after equine influenza infection.
A major histocompatibility complex (MHC) restricted cytotoxic T lymphocyte (CTL) activity specific for equine influenza was demonstrated 14 days after experimental infection  and a significant virus-specific CTL response was still detectable 6 months after infection. This activity was increased by a second influenza infection (Fig. 1B) .It therefore appears that, in the absence of any detectable antibody, a CTL response is coincident with protection against infection, although this hypothesis has not been tested in horses. More recently, an up-regulation of mRNA encoding IFN, IL-4 and IL-2 was shown in peripheral blood mononuclear cells (PBMC) and lymph nodes 14 days after experimental infection , but the significance of this in relation to protection is unknown. An assay measuring virus specific IFN-. synthesis has been published recently  and it may be adaptable to EIV in the near future, making measurement of cellular immunity against this virus easier.
3. Vaccines against equine influenza virus
The principal aim of influenza vaccination is to reduce clinical signs of the disease, with subsequent improved animal welfare leading to a shortened convalescent period and reducing secondary infections. Reduced shedding of virus has important implications for the spread of infection and is certainly the other major target that should be achieved by vaccination. Vaccination should also provide long-term immunity, an efficient memory response and cross-protection against influenza viruses of different strains. It has been estimated that 70% of a given population of horses needs to be fully vaccinated to prevent epidemics of influenza .
In assessing vaccine efficacy in horses, clinical protection against equine influenza is defined by the absence of pyrexia, and other clinical signs induced by infection, such as nasal discharge and cough. Virological protection is defined by the absence of virus in mucosal secretions as detected by titration of virus in nasal swab extracts. ‘Seroconversion’ is defined as a significant increase of antibody.
Vaccination schedules can differ according to a country’s regulations, the type of vaccine and the vaccine manufacturer’s recommendations. However, a standard schedule for equine influenza vaccination requires that primarily vaccinated horses have to receive a second vaccination within 3 months. Generally, booster immunisations must be administrated within 6 months of the second vaccination and at least annually thereafter, but in some cases, boosters are given more frequently [39, 40].
Current vaccination strategies can be divided into the administration of either ‘dead’ or ‘live’ vaccines. ‘Dead’ vaccines include killed whole virus, subunit proteins and DNA vaccination. ‘Live’ vaccines include attenuated virus or living virus-based vector vaccines.
3.1. ‘Dead’ vaccines
Since the introduction of EIV vaccines in the 1960s, the majority of equine influenza vaccines commercially available contained inactivated whole virus or subunits. The main advantages of these vaccines are the absence of pathogenicity, virus replication and subsequent spread between hosts. For the preparation of these vaccines, EIV has been traditionally grown in embryonated hens’ eggs. To reduce the subsequent reactogenicity that may occur to repeated immunisation with egg protein, methods of tissue culture have been developed.
3.1.1. Inactivated whole influenza virus vaccines
Protection from influenza disease conferred by conventional inactivated vaccines is strongly associated with the levels of circulating antibodies against HA, provided that the vaccine strain and the challenge infection strain are genetically and antigenically similar. For example, in an early study, ponies with a single radial haemolysis (SRH) antibody level >154 mm2 were resistant to infection with EIV, and resistant to clinical signs of disease with SRH antibody levels >85 mm2 . In a second study, ponies were protected from infection with an SRH antibody level >120 mm2, and from clinical signs of disease with SRH antibody >90 mm2 . It should be noted that the severity of disease could vary depending on the method of infection (e.g. intranasal instillation versus nebulised aerosol) and the titre of virus used for challenge infection, and these factors could influence the level of SRH antibody required for protection. However, it is clear that high levels of SRH antibody immediately prior to exposure to influenza virus play an important role in protection (Fig. 2A).
Fig. 2. Immune response induced by inactivated whole EIV or subunit vaccines. (A) Vaccination by intramuscular injection induces a humoral immune response , producing serum antibodies able to neutralise incoming EIV . However, this antibody response is short lived and infection could occur in the case of low level of antibody or a significantly mistmach between the vaccine strain of EIV and the infectious EIV . (B) Inactivated whole virus vaccine administrated intranasally and adjuvanted with CTB induces a local immune response producing serum and mucosal antibodies with VN activity. Ab: antibody; APC: antigen presenting cells; B: B cell; CTB: cholera toxin B; NALT/MALT: nasal/mucosal associated lymphoid tissue; Th: T helper lymphocyte; Tc: cytotoxic T lymphocytes.
The antibody isotype induced by a conventional inactivated influenza virus vaccine (A/eq/Kentucky/1/81) was analysed and shown to be composed only of short-lived (<100 activity="" adjuvanted="" against="" aluminum="" and="" antibodies="" are="" bacterial="" by="" can="" complement="" ctl="" days="" durability="" efficacy="" eiv.="" eiv="" enhance="" failed="" fix="" fixation="" for="" generally="" hydroxide="" igg="" immune="" in="" inactivated="" induce="" influenza="" inhibit="" intestinal="" involved="" isotype="" known="" level="" neutralising="" o:p="" of="" or="" other="" parasites.="" phosphate="" poorly="" post-vaccination="" protective="" response.="" responses="" specific="" systemic="" the="" their="" therefore="" to="" toxins="" unable="" vaccines="" virus="" whole="" with="">100>
Amorphous aluminum hydroxyphosphate gel (alhydrogel or adjus-phos) are common adjuvants, which, in mice, stimulate IL-4 synthesis and activate Th2 cells, with enhanced IgG1 and IgE production . Other adjuvants include polymers of acrylic acid cross-linked with polyallylsucrose, or a lipidic, non-aluminum, dual phase adjuvant (MetaStimTM). Carbomer, an emulsifying agent that is a polyacrylic acid with an extremely high molecular weight, is also used as an adjuvant . The intranasal inoculation of the inactivated influenza vaccine with cholera toxin B (CTB) as adjuvant has been shown to induce a local immune response, composed of virus-specific VN and IgA antibodies, which protects ponies from infection with EIV (Fig. 2B) . Conventional inactivated vaccines administered intramuscularly can cause very occasional adverse reactions (e.g. local pain and swelling, pyrexia) .
3.1.2. Subunit vaccines
Current subunit vaccines contains purified HA and NA proteins. These membrane proteins (antigens) are generally adjuvanted with Quillaia saponin (Quil A) or integrated into immuno-stimulating complexes (ISCOMTM1) to improve their antigenicity. ISCOMsTM are formed from a matrix (i.e. a combination of Quil A, phospholipids and cholesterol), which incorporates membrane proteins to give stable, self adjuvanting particles, held together by hydrophobic interactions. In other species, ISCOMTM vaccines have shown successful protection against numerous pathogens .In the late 1980s, an ISCOM vaccine containing HA antigen from A/eq/Solvalla/79 (H3N8) was shown to induce IgGab antibody lasting about 25 weeks in trachea and nasopharynx of horses .
In the mid 1990s, an ISCOMTM vaccine (EquipTM; Pitman-Moore) containing HA protein from both A/eq/ Newmarket/77 (H7N7) and A/eq/Brentwood/79 (H3N8) was evaluated in ponies before a challenge infection with A/eq/Sussex/89 (H3N8), 15 months after three immunisations (two intramuscular vaccinations 6 weeks apart, and a booster immunisation 5 months later). Vaccinated ponies showed an SRH antibody response against both virus subtypes after immunisation. Only 3/7 vaccinates developed a mild and transient pyrexia, but vaccinated ponies were nearly completely protected from virus shedding after challenge
infection (six out of seven ponies) compared with the control group (zero out of five ponies). Two vaccinates did not even demonstrate a significant increase in antibody after infection
. With an equivalent amount of HA antigen, an ISCOM vaccine was shown to induce SRH antibodies more efficiently than an inactivated whole virus vaccine . More recently,
an ISCOM-based vaccine containing HA antigen from Newmarket/77 (H7N7), Borl¨ange/91 (H3N8) and Kentucky/98 (H3N8) (EQUIP F, Schering-Plough Animal Health) provided
strong protective immunity against EIV challenge.
Ponies received two intra-muscular immunisations, 6 weeks apart, and were challenge infected with Newmarket/1/93 influenza virus 4 weeks after the second vaccination. Serum
SRH antibodies increased after the first immunisation and were boosted after the second vaccination. Virus-specific antibodies were predominantly of the IgGa and IgGb isotypes
(the best mediators of complement fixing (CF) antibody and antibody-dependent cell cytotoxicity). Virus-specific IgGc and IgG(T) increased after the second immunisation. In nasal
secretions, only virus-specific IgG antibodies were detected. Vaccinated ponies were completely protected against clinical signs of the disease but 43% showed virus shedding after
challenge infection . In another study, the mucosal administration of this ISCOM-based vaccine in systemically primed ponies increased the level of nasal IgA after challenge infection,
induced a full protection against clinical signs of disease and reduced the duration of virus excretion compared with unvaccinated controls . However, immunity induced by
1 ISCOM is a trademark of CSL limited. sub-unit vaccines is limited and dependent on the role of the particular antigens in protection. Ponies vaccinated with a subunit vaccine containing Quil A as an adjuvant failed to induce an influenza-virus specific CTL activity .
In humans, an experimental ISCOM based vaccine for influenza containing HA molecules has been shown to increase the humoral immune response and could have some effects on CMI after vaccination compared with a conventional inactivated vaccine .
In summary, the efficacy of inactivated whole virus or subunit vaccines is based on their ability to stimulate high levels of SRH antibody that are required for protection. However, this antibody response is short-lived, could be highly specific, and subsequent protection could be sensitive to antigenic variation of influenza viruses. In horses, the stimulation of an influenza virus-specific CMI by these vaccines has not been demonstrated. Comparison of the differences between immune responses induced by immunisation with killed virus or subunit vaccines and viral infection suggest that vaccine design can be improved.
3.1.3. DNA vaccines
Administration of DNA plasmids offers a different form of vaccination and efficient protection against influenza infection has been demonstrated in several species (e.g. mice, ferrets, chicken [51,52]). DNA vaccination results in the in vivo expression of antigenic proteins, leading to the stimulation of both humoral and cellular immune responses. This approach is therefore advantageous compared with inactivated whole virus or sub-unit vaccines. Technically, plasmid DNA is inexpensive to produce, has good stability and can be lyophilised for long-term storage.
In mice, Olsen et al.  demonstrated that gene gun delivery of plasmid DNA encoding the A/eq/Kentucky/1/81 (H3N8) HA protein provided full protection against challenge infection, with induction of VN antibody, serum virusspecific IgG antibodies and serum IgA . The impact of skin and mucosal immunisation with a plasmid DNA vaccine coding for the HA protein of A/eq/Kentucky/1/81 influenza virus was therefore studied in the horse [36,54]. Ponies were immunised by gene gun injection at skin and mucosal sites (60 sites of injection per immunisation, 3 immunisations in total, around 60 days between each one). Vaccinated horses were partially to completely protected against clinical signs of the disease and virus shedding induced by challenge infection with a homologous strain of influenza virus, 30 days after the last vaccination. DNA vaccination induced serum IgGa and IgGb antibody responses (i.e. an isotype response similar to EIV infection) but no mucosal IgA responses were
detected. Virus-specific lymphoproliferative responses and up-regulation of IFN-. mRNA were also observed after three immunisations with the DNA vaccine but differed markedly from responses induced by equine influenza vrius infection in terms of regional distribution (i.e. peripheral blood, draining or hilar lymph nodes) . Protection resulting from DNA vaccination was associated with the presence of both or the intramuscular injection of live vector-based vaccines will induce the expression of EIV antigens and stimulate both humoral and cellular immune responses . Serum antibodies will neutralise infectious EIV
and EIV-specific CTL will lyse infected epithelial cells . (B) The administration of DNA vaccine via the nasal route with CTB as adjuvant, has been shown to induce the synthesis of a mucosal antibody response . Ab: antibody; APC: antigen presenting cells; B: B cell; CTB: cholera toxin B; NALT/MALT: nasal/mucosal associated lymphoid tissue; Th: T helper lymphocyte; Tc: cytotoxic T lymphoctes. serum IgGa/IgGb antibodies and mucosal IgG, supporting the idea that mucosal IgA is not essential for protection (Fig. 3A) [36,54]. Absence of virus shedding was correlated with the presence of mucosal IgGb at the time of challenge infection . However, as IgA has been shown to be an important part of the immune response induced by natural infection, DNA vaccines were modified in an attempt to achieve such a response. Co-administration of DNA coding for IL-6 failed to improve either IgA synthesis or protection from infection but may have caused a shift to a Th-2 like antibody response . Another recent study showed nasal IgA synthesis after two intranasal DNA-HA vaccinations (33 days apart, derived from strain A/eq/Kentucky/1/81) followed by two skin/mucosal vaccinations (36 days apart). Co-administration of cholera toxin (CT) and CTB, which stimulate both humoral and cellular immune responses, as adjuvant increased nasal IgA titre after vaccination compared with the DNA-HA vaccination alone (Fig. 3B). IgA and IgGb antibody responses were also increased in CT + DNA-HA vaccinated ponies after challenge infection with EIV (81 days after the last immunisation, strain A/eq/Kentucky/1/81) compared with unvaccinated ponies. However, vaccinated ponies were only partially protected against clinical signs of disease and virus shedding . Interestingly, some horses were clinically protected, despite very low levels of serum antibody at the time of challenge infection.
In summary, DNA vaccination was shown to mount a protective influenza-specific immune response in horses with good longevity compared with conventional inactivated virus vaccines. However, studies have shown some variability in the level of protection from clinical signs of disease and prevention of virus shedding achieved. The method of administration of DNA vaccines (multiple sites of inoculation and multiple injections for each immunisation) is obviously impractical for veterinary practice. A DNA vaccine against West Nile virus has been recently licensed in North America and will allow the evaluation of DNA vaccination feasibility
in the field.
Fig. 3. Immune response induced by DNA or live vector-based vaccines. (A) A skin DNA immunisation
3.2. Live virus/vector vaccines
Hannant et al.  demonstrated that the immune response induced by experimental infection with equine influenza virus protects horses against re-infection for up to 1 year
. These results suggest that an equine influenza virus vaccine mimicking natural infection should be more efficient in affording protection than conventional inactivated vaccines.
The use of a modified/attenuated live influenza virus or a live virus-vector coding for EIV proteins are the main approaches investigated in the last 20 years to produce a live equine
3.2.1. Live attenuated influenza virus vaccine
Immunisation with a live attenuated influenza virus closely mimics natural infection. Antigens are presented to the immune system via both exogenous and endogenous pathways, and vaccines are therefore expected to stimulate an immune response similar to those induced by infection (Fig. 1A).
188.8.131.52. Reassortant influenza virus.
The genome of influenza virus consists of eight single strand RNA segments. Genetic reassortment can occur when a cell is infected simultaneously with two different subtypes of influenza A virus. An influenza virus with a limited growth in mammals (e.g. avian influenza virus) can be reassorted with a mammalian influenza virus to obtain a live attenuated virus able to induce a protective immune response. A reassortant expressing internal proteins of the avian influenza virus A/Duck/ New York/6750/78 (H2N2) and surface glycoproteins ofthe equine influenza virus A/eq/Georgia/1/81 (H3N8) was evaluated for efficacy and safety in hamsters and ponies. Exposure of ponies to this reassortant induced mild clinical signs compared with the parental equine influenza virus. The reassortant induced an antibody response and was shed after infection. Five-and-a half months after the exposure to the reassortant, ponies were partially protected from a challenge infection with the parental equine influenza virus, the duration of virus shedding was also slightly reduced . The limited protection observed in this study may have originated from the immune response raised against equine influenza surface glycoproteins born by the reassortant. The response raised against avian internal proteins probably played at best a limited role in protection
against the equine influenza virus. Vaccines based on avian/equine reassortants certainly would be too risky to use. Reassortants could be safe in horses but induced disease in birds . Furthermore, avian influenza virus can transmit directly to horses and has caused a severe outbreak in the past (i.e. equine influenza in China during 1989 cause by an avian influenza (H3N8) virus ). Also taking into consideration recent infections of humans
with avian influenza virus in Asia, the use of such reassortants for horse vaccination would raise a lot of safety concerns.
184.108.40.206. Temperature-sensitive (Ts) or cold-adapted influenza virus.
A temperature-sensitive (Ts) influenza virus is expected to show a significant decrease (up to 100-fold) in titre at a restrictive temperature (e.g. 39 .C) compared with the titre obtained at a permissive temperature (e.g. 34 .C). In vivo, Ts influenza virus multiplies efficiently in the cooler environment of the upper respiratory tract where they induce local and systemic immune responses. However, replication is inefficient in the warmer environment of the lower respiratory tract in horses (38–39 .C), where replication of wild type virus may induce bronchitis, bronchiolitis, interstitial pneumonia, congestion and oedema . It has also been
speculated that the high body temperature of horses might repress virus reversion .
In the late 1980s, the Ts clone 8B1 (bearing a Ts lesion on the PA gene) was derived from the re-assortment of a human influenza A Ts donor virus (bearing Ts lesions on the polymerase
and NP genes) with a wild type A/eq/Cornell/16/74 (H7N7) equine influenza virus. In the horse, intranasal instillation of clone 8B1 did not cause detectable clinical signs or febrile responses. Vaccine virus that was shed retained the Ts phenotype. When challenge infected (28 days after the last vaccine exposure) with the parental equine H7N7 influenza virus, ponies were partially to fully protected from clinical signs of the disease and virus shedding (21 of 40
vaccinated ponies did not shed virus after challenge infection) . In the mid 1990s, clone 8B1 virus was used as the donor virus with the wild type A/eq/Kentucky/1/81 (H3N8)
influenza virus to produce an H3N8 Ts reassortant virus (Clone 255). Ponies vaccinated by a nebulised suspension containing the reassortant virus showed no clinical signs after
immunisation with Clone 255 but did seroconvert for haemagglutination inhibiting (HI) antibody after exposure to the reassortant, indicating antigenicity of the reassortant. When
they were challenge infected with the A/eq/Kentucky/1/81 parental influenza virus 1 or 2 months after immunisation, vaccinated ponies were fully protected against clinical signs
of disease but virus shedding was observed in 20–33% of vaccinated ponies. Ponies remained protected against clinical signs of disease when they were challenged 7 and 10 months
after immunisation, but the percentage of ponies shedding virus increased to 55% .
Cold adapted, Ts live equine influenza viruses were also produced by serial passage in embryonated hens’ eggs at temperatures gradually reduced to 26 .C . More recently,
a cold-adapted, Ts, modified-live equine influenza virus vaccine (Flu Avert®2 I.N. Vaccine; Intervet) derived from the wild-type A/eq/Kentucky/1/91 (H3N8) influenza virus, has been evaluated and is available commercially in north America. With one nasal mucosal immunisation, vaccinated ponies were fully protected from clinical signs of the disease
when challenge infected 5 weeks later with the parental A/eq/Kentucky/1/91 influenza virus. Vaccinates showed significant clinical protection when re-challenged 6 months
(2 Avert is a registered trademark of Heska corporation. ) post-vaccination and lower rectal temperatures when challenged 12 months after vaccination. Virus shedding was also strongly reduced after vaccination. Only 20% of vaccinates shed virus when challenge infected 5 weeks post-vaccination. Six and 12 months post-vaccination, the duration of virus shedding had decreased significantly compared with unvaccinated control ponies. Although no correlation between SRH antibody response and protection was observed, vaccinated ponies developed an anamnestic SRH response after challenge . This vaccine has also been shown to protect ponies against infection with an heterologous influenza virus . Vaccination with these live attenuated vaccines induced longterm immunity that clearly limits the duration and severity of
clinical signs and the shedding of virus after challenge infection with equine influenza virus. However these vaccines did not provide a sterile immunity.
The use of modified/attenuated-live influenza virus vaccines has always raised concerns due to the high mutation rate and potential reassortment of influenza virus. The segmented nature of the influenza virus genome could allow for reassortment with a cocirculating wild-type virus and subsequent loss of attenuation or/and emergence of a new highly pathogenic influenza virus. In humans however, a live attenuated influenza virus vaccine (mixture of two type A viruses (H1N1 and H3N2) and a type B virus) has been approved and is currently in use (FluMist®3). Moreover, Russia and Eastern European countries have been using live human influenza vaccines for decades in widespread vaccination programmes, without the emergence of new influenza virus reassortants. In horses, studies have revealed the phenotypic stability of
the cold adapted, Ts modified live equine influenza virus used as vaccine, its low spontaneous transmission to unvaccinated ponies , and the absence of clinical signs of disease induced by vaccination in ponies with exercise-induced immunosuppression .
3.2.2. Virus-based vector vaccines
Live recombinant vector vaccines are constructed by inserting selected genes from the pathogen of interest into live, infectious, but non-disease-causing viruses. With virusbased
vector vaccines, viral antigens are expressed and synthesised de novo within the infected cell. Thus, because they are presented via MHC class I (endogenous) and class II (exogenous) antigen-processing routes, the selected viral antigens can stimulate both humoral and cellular immune responses (Fig. 3A). This strategy of vaccination has been the focus of intense scientific investigation in many species, including the horse.
Recombinant poxviruses have been widely used for vaccination [64,65]. Poxviruses are genetically stable and allow the insertion of a large segment of foreign DNA coding for
selected antigens. Recombinant poxviruses derived from vaccinia or avipoxvirus (i.e. canarypox virus) are commercially (3 FluMist is a registered trademark of MedImmuneVaccines Inc.)
available and several recombinant canarypox-based vaccines have been developed for the horse [66,67].
In the late 1980s, immunisation of horses with recombinant vaccinia virus expressing HA and NA proteins of equine influenza virus induced antibody responses successfully (VN, HI and SRH) . More recently, Breathnach et al.  investigated the efficiency of a priming immunisation with a DNA-vaccine coding for HA and NP proteins (Kentucky/1/81) and booster vaccinations, 6 and 10 weeks later with a modified vaccinia Ankara (MVA) virus encoding the
same proteins. MVA virus is a highly attenuated strain of vaccinia virus showing a defect in replication, which is frequently used in prime-boost vaccination protocols. Following extensive in vitro passage, the parental vaccinia virus became replication-deficient in virtually all mammalian cells . Primary DNA vaccination did not induce detectable virus specific humoral or cellular immune responses, but booster administration of recombinant MVA strongly induced virusspecific serum IgGa and IgGb as well as a virus-specific lymphoproliferative response and up-regulation of IFN-. mRNA. Interestingly, both HA and NP antigens stimulated humoral and cellular immune responses. The administration of a second booster vaccination had little effect on virus specific antibodies or lymphoproliferative responses but up regulated IFN-. mRNA. The efficiency of such constructs against EIV infection was investigated in a recent study. Ponies were vaccinated with MVA-HA alone (three immunisations) or as part of a prime boost regimen including a DNA priming and two subsequent boosts with MVA-HA or MVA-NP (e.g. DNA-NP plus MVA-NP). Ponies vaccinated
with MVA-HA alone or associated with a DNA prime were significantly protected against clinical signs of disease and virus shedding after infection with EIV when compared with unvaccinated ponies. Vaccination stimulated both humoral and cellular immune responses, including an IgA antibody response in nasal secretion. Protection induced by vaccination
with MVA-NP was lower but could be explained by the absence of a VN antibody response, as NP is an internal protein of EIV . These studies provided evidence that recombinant MVA vectors coding for EIV proteins efficiently stimulated the immune system and afford a good level of protection.
In 2003, a modified live canarypox virus-vectored influenza vaccine (ProteqFlu®4; Merial Ltd., UK) was licensed in the European Union for use in horses. This live, vectored vaccine is safe, as canarypox virus undergoes an abortive infection in mammalian cells . This vaccine contains
two canarypox recombinant viruses coding for the HA proteins from Newmarket/2/93 and Kentucky/94 EIV strains (European and American representative strains, respectively) adjuvanted with Carbormer 974P. The efficacy of this new commercialised vaccine has been recently studied [72,73]. Na¨ive ponies received one or two intramuscular immunisations
with ProteqFlu® and were challenged 14 days after (4 ProteqFlu is a registered trademark of Merial Ltd. )the last immunisation with a virulent strain of EIV (Newmarket/5/03; isolated recently in the UK that induced severe disease after field infection). After the first immunisation, all vaccinated ponies developed detectable levels of SRH antibodies and a few ponies mounted an anamnestic response after the second immunisation. Singly vaccinated ponies did not shed virus and were protected or developed only mild signs of the disease after challenge infection with EIV and similar results were obtained with ponies vaccinated twice.
Virus excretion was almost completely suppressed in vaccinates . An increased IFN-. protein synthesis was also measured in vaccinated ponies after challenge infection with
EIV . ProteqFlu® is therefore successful in preventing infection with related strains of EIV. The immunity induced by this vaccine is directed against HA proteins that are subject to antigenic drift. Therefore regular updating of the vaccine will be required to ensure continued efficacy. However, if there is a significant mismatch between vaccine HA and circulating virus, the first line of defence (i.e. antibody) may fail and a virus-specific T cell response would become important in promoting recovery and virus clearance. Further work is required to evaluate the benefit of including additional recombinant live vectors coding for the more conserved NP or M proteins, which are postulated to contain immunodominant CTL epitopes (and shown to induce a good level of protection in the case of NP ), to reinforce the virus-specific T cell response provided by vaccines such as ProteqFlu®.
Systemic inoculation is the most frequent route of poxvirus-based vaccination. Few studies have evaluated the efficacy of recombinant poxviruses delivered by mucosal routes. In mice, oral immunisation with a MVA-based influenza vaccine afforded complete protection against a
lethal influenza challenge . In humans, canarypox virus is not considered as a strong mucosal immunogen .However, intranasal immunisation with canarypox virus-based
vectors coding for the canine distemper virus (CDV) haemagglutinin and fusion genes protected juvenile ferrets against an intranasal challenge with virulent CDV . In horses,
mucosal influenza vaccination with poxvirus-based vectors remains to be evaluated.
One concern with poxvirus-based vaccines is the presence of pre-existing immunity specific to poxvirus and its impact on subsequent vaccination with the same recombinant. Suppression of the antibody response to recombinant antigen encoded by a vaccinia-based vector has been
reported [77,78]. However, at least in horses, vaccination with canarypox-based vaccines had no discernable effect on the efficiency of subsequent immunisation using the same vector coding for homologous or heterologous antigens (Dr. J.H. Minke, personal communication).
3.2.3. Live attenuated influenza virus designed by reverse genetics
Reverse genetics allows the generation of entirely artificial recombinant influenza viruses from cloned DNA plasmids . In the mid 1990s, application of the reverse genetics method to equine influenza virus was discussed after the generation of an attenuated influenza virus by the insertion of a mutation in the cytoplasmic tail of NA protein . Ten years elapsed before the generation by reverse genetics of attenuated equine influenza viruses encoding a carboxy-terminus truncated NS1 protein . Briefly, an infectious recombinant virus was recovered after the transfection of a dog kidney cell line (MDCK) with plasmids coding for each of the eight influenza RNAs (PB1, PB2, PA, HA, NP, NA, M and NS) simultaneously with protein expression plasmids (PB1, PB2, PA, NP and NS1). Derived from the A/eq/Kentucky/5/02
(H3N8) influenza virus, the 3 recombinant viruses obtained showed a low replication capacity both in vitro in MDCK cells, and also in vivo in embryonated hens’ eggs (9 to 11-dayold) or in mice after an intranasal infection. The recombinant viruses ability to inhibit the production of IFN/. by infected cells (previously associated with the NS1 protein in wild type virus ), was also decreased . However, recombinant virus replicated well in 7-day-old embryonated hens’ eggs, which lack a competent IFN system compared with 9 to 11-day-old eggs. An unaltered IFN/. synthesis induced by cellular infection with recombinant viruses is interesting
because these cytokines possess a powerful adjuvant activity when co-administered with a human influenza vaccine, inducing a Th1 immune response and protection against challenge
infection in mice . An equine influenza virus with an altered NS1 protein could be a good master strain for live attenuated viral vaccines, reverse genetics allowing the insertion of HA and NA genes of epidemic equine influenza virus strains. The same approach has been suggested in the context of human vaccination by Schickli et al. ,
Reverse genetics also allows the generation of ‘cocktail’ virus vaccines as exemplified by the recent live bivalent vaccine for parainfluenza and influenza virus. This recombinant
influenza virus (A/WSN/33, H1N1) possesses the coding regions for the hemagglutinin/neuraminidase ectodomain of a murine parainfluenza virus type 1 (Sendai virus) instead of the influenza virus neuraminidase. Vaccinated mice were protected from a lethal challenge infection with both parental viruses . However, such combination needs attention in the gene selection. In this case, the sialidase activity of the hemagglutin/neuraminidase ectodomain of the parainfluenza virus compensated for the lack of influenza virus neuraminidase necessary for the bivalent virus to replicate.
Reverse genetics is an efficient method to generate recombinant live attenuated influenza virus. However, such recombinants for equine influenza virus remain to be tested in vivo
in horses to assess their antigenicity and efficacy in protection against challenge infection.
3.3. Effect of maternal antibodies and age on equine influenza vaccination
Foals are highly sensitive to disease. Their only defence against pathogens are maternal antibodies acquired through the ingestion of colostrum during the first hours of life. This
passive immunity can protect foals against influenza. However, the duration of maternal antibodies in foals is controversial [85–87], but antibodies specific for equine influenza
generally decrease to a low but detectable level by 6 months [88,89]. Protection of foals can be improved by increasing the titre of protective antibodies in the colostrum, mainly by influenza vaccination of the mare a few weeks prior to foaling. Thoroughbred foals are generally vaccinated early to avoid any windows of susceptibility (when protection against
influenza is not anymore assumed by maternal antibodies and not yet by vaccination). However, residual maternally derived antibodies can reduce vaccine efficiency, at least in
the case of inactivated, sub-unit or attenuated virus vaccines [87,89–91]. The development of tolerance in 3 months old foals to influenza subunit vaccination has been reported 
but remains controversial . Primary vaccination should therefore occur after the complete disappearance of maternal antibodies (i.e. after 6 months of age). The use of virus-based
vectors for equine influenza in foals could circumvent the neutralising effect of maternal antibodies on vaccine efficacy. A second factor that influences vaccine efficacy is the
foal’s weak immune response to vaccination, which results in poor protection, and have been associated with an immature activity of antigen presenting cells compared with adult horses . Thus, the use of vaccines specifically designed to induce a strong CMI could be less effective than expected. Mechanisms driving the maturation of the neonate’s immune
system are not yet fully understood, and as a result, effective vaccination schedules in foals still need investigation.
Another factor that influences the efficacy of vaccination is old age. The immune senescence observed in human and equine populations presents a few similarities . B and T cell activity is altered [94,95] and titres of EIV-specific antibody are lower in old animals when compared with younger animals [94,96]. The use of live modified/attenuated equine influenza virus as vaccine could be a problem in this context of immunosuppression. Therefore, annual vaccination with a conventional inactivated or sub-unit vaccines is likely to be safer in this particular population. Further studies will be necessary to evaluate the effect of age on other approaches to vaccination (e.g. DNA, poxvirus-based vectors).
Today, the main types of equine influenza vaccines in use contain whole inactivated virus or subunits. Protection afforded by this first generation of vaccines is based on high levels of protective antibodies. Second generation vaccines (i.e. live attenuated and poxvirus-based vaccines) are now available. These stimulate both humoral and cellular immune responses and so mimic more closely the protective immunity induced by natural infection with influenza virus. These vaccines are not yet widely used and time is necessary to evaluate their performance in the field. The delivery routes of DNA vaccines remain a major issue, despite their ability to protect horses against influenza. It is anticipated that live attenuated
influenza virus engineered by reverse genetics will almost certainly be the next generation of vaccines against equine influenza. A large battery of weapons is therefore becoming
available to the equine clinician. Several laboratory animal models (e.g. rodents, ferrets and monkeys) are used to evaluate vaccine efficacy against influenza infection. However the horse model allows field studies of a range of vaccine techniques in the natural host population, and integration of natural external factors (e.g. population dynamics, genetic diversity) that can influence the outcome of vaccination. Therefore, the horse is an interesting model of vaccination in the battle against the evolutionary strategy of influenza viruses.
 Wilson WD. Equine influenza. Vet Clin North Am Equine Pract 1993; 9(2):257–82.
 Studdert MJ. Orthomyxoviridae. In: Studdert MJ, Horzinek MC,
editors. Virus infections of vertebrates, Virus Infections of equines. Amsterdam: Elsevier Science Publishers; 1996. p. 281–4.
 Gibbs JS, Malide D, Hornung F, Bennink JR, Yewdell JW. The influenza A virus PB1-F2 protein targets the inner mitochondrial membrane via a predicted basic amphipathic helix that disrupts mitochondrial function. J Virol 2003;77(13):7214–24.
 O’Neill RE, Talon J, Palese P. The influenza virus NEP (NS2 protein) mediates the nuclear export of viral ribonucleoproteins. EMBO J 1998;17(1):288–96.
 Hannant D, Mumford JA. Equine influenza. In: Studdert MJ, Horzinek MC, editors. Virus infections of vertebrates, Virus infections of equines. Amsterdam: Elsevier Science Publishers; 1996. p. 285–93.
 Daly JM, Mumford JA. Influenza infections. In: Leukeux P, editor. Equine respiratory diseases. Ithaca, NY: International Veterinary Information Service (IVIS); 2001.
 Wattrang E, Jessett DM, Yates P, Fuxler L, Hannant D. Experimental infection of ponies with equine influenza A2 (H3N8) virus strains of different pathogenicity elicits varying interferon and interleukin-6 responses. Viral Immunol 2003;16(1):57–67.
 Fouchier RA, Munster V, Wallensten A, Bestebroer TM, Herfst S, Smith D, et al. Characterization of a novel influenza A virus hemagglutinin subtype (H16) obtained from black-headed gulls. J Virol 2005;79(5):2814–22.
 Horimoto T, Kawaoka Y. Pandemic threat posed by avian influenza A viruses. Clin Microbiol Rev 2001;14(1):129–49.
 Sovinova O, Tumova B, Pouska F, Nemec J. Isolation of a virus causing respiratory disease in horses. Acta Virol 1958;2(1):52–61.
 Singh G. Characterization of A/eq-1 virus isolated during the equine influenza epidemic in India. Acta Virol 1994;38(1):25–6.
 Ismail TM, Sami AM, Youssef HM, Abou Zaid AA. An outbreak of equine influenza type 1 in Egypt in 1989. Vet Med J Giza 1990;38(2):195–206.
 Madic J, Martinovic S, Naglic T, Hajsig D, Cvetnic S. Serological evidence for the presence of A/equine-1 influenza virus in unvaccinated horses in Croatia. Vet Rec 1996;138(3):68.
GH, Teigland MB, Sigel MM. A new influenza virus associated with equine respiratory disease. J Am Vet Med Assoc 1963;143:587–90.
 Guo Y, Wang M, Kawaoka Y, Gorman O, Ito T, Saito T, et al. Characterization of a new avian-like influenza A virus from horses in China. Virology 1992;188(1):245–55.
 Crawford PC, Dubovi EJ, Castleman WL, Stephenson I, Gibbs EP, Chen L, et al. Transmission of equine influenza virus to dogs. Science 2005;310(5747):482–5.
 Park AW, Wood JL, Daly JM, Newton JR, Glass K, Henley W, et al. The effects of strain heterology on the epidemiology of equine influenza in a vaccinated population. Proc Biol Sci 2004;271(1548): 1547–55.
 Daly JM, Yates RJ, Browse G, Swann Z, Newton JR, Jessett D, et al. Comparison of hamster and pony challenge models for evaluation of effect of antigenic drift on cross protection afforded by equine influenza vaccines. Equine Vet J 2003;35(5):458–62.
 Yates P, Mumford JA. Equine influenza vaccine efficacy: the significance of antigenic variation. Vet Microbiol 2000;74:173–7.
 OIE. Conclusions and recommendations from the consultation of OIE and WHO experts on equine influenza, Newmarket, United Kingdom, September 18–19, 1995. OIE Bull 1996;108:482–4.
 Daly JM, Yates PJ, Newton JR, Park A, Henley W, Wood JL, et al. Evidence supporting the inclusion of strains from each of the two co-circulating lineages of H3N8 equine influenza virus in vaccines. Vaccine 2004;22(29/30):4101–9.
 Tamura S, Kurata T. Defense mechanisms against influenza virus infection in the respiratory tract mucosa. Jpn J Infect Dis 2004;57(6): 236–47.
 Ramsay AJ, Husband AJ, Ramshaw IA, Bao S, Matthaei KI, Koehler G, et al. The role of interleukin-6 in mucosal IgA antibody responses in vivo. Science 1994;264(5158):561–3.
 Hannant D, Mumford JA. Cell mediated immune responses in ponies following infection with equine influenza virus (H3N8): the influence of induction culture conditions on the properties of cytotoxic effector cells. Vet Immunol Immunopathol 1989;21(3/4):327–37.
 Hannant D, Jessett DM, O’Neill T, Sundquist B, Mumford JA. Nasopharyngeal, tracheobronchial, and sytemic immune responses to vaccination and aerosol infection with equine-2 influenza A virus (H3N8). In: Powell DG, editor. Equine infectious diseases. V. Proceedings of the Fifth International Conference. Lexington, KY: University Press of Kentucky; 1987. p. 66–73.
 Slater J, Hannant D. Equine immunity to viruses. Vet Clin North Am Equine Pract 2000;16(1):49–68.
 Kumar P, Timoney JF, Sheoran AS. M cells and associated lymphoid tissue of the equine nasopharyngeal tonsil. Equine Vet J 2001;33(3):224–30.
 Nelson KM, Schram BR, McGregor MW, Sheoran AS, Olsen CW, Lunn DP. Local and systemic isotype-specific antibody responses to equine influenza virus infection versus conventional vaccination. Vaccine 1998;16(13):1306–13.
 Mazanec MB, Coudret CL, Fletcher DR. Intracellular neutralization of influenza virus by immunoglobulin A anti-hemagglutinin monoclonal antibodies. J Virol 1995;69(2):1339–43.
 Tamura S, Funato H, Hirabayashi Y, Suzuki Y, Nagamine T, Aizawa C, et al. Cross-protection against influenza A virus infection by passively transferred respiratory tract IgA antibodies to different hemagglutinin molecules. Eur J Immunol 1991;21(6):1337–44.
 Galan JE, Timoney JF, Lengemann FW. Passive transfer of mucosal antibody to Streptococcus equi in the foal. Infect Immun 1986;54(1):202–6.
 McGuire TC, Crawford TB. Passive immunity in the foal: measurement of immunoglobulin classes and specific antibody. Am J Vet Res 1973;34(10):1299–303.
 Fujimiya Y, Perryman LE, Crawford TB. Leukocyte cytotoxicity in a persistent virus infection: presence of direct cytotoxicity but absence of antibody-dependent cellular cytotoxicity in horses infected with equine infectious anemia virus. Infect Immun 1979;24(3):628–36.
 Hannant D, Mumford JA, Jessett DM. Duration of circulating antibody and immunity following infection with equine influenza virus. Vet Rec 1988;122:125–8.
 Hannant D, Jessett DM, O’Neill T, Livesay GJ, Mumford JA. Cellular immune responses stimulated by inactivated virus vaccines and infection with equine influenza virus (H3N8). In: Nakajima H, Plowright W, editors. Equine infectious diseases. VII: Proceedings of Seventh International Conference on Equine Infectious Diseases. Newmarket, UK:R&W Publications Ltd.; 1994. p. 169–74.
 Soboll G, Horohov DW, Aldridge BM, Olsen CW, McGregor MW, Drape RJ, et al. Regional antibody and cellular immune responses to equine influenza virus infection, and particle mediated DNA vaccination. Vet Immunol Immunopathol 2003;94(1/2):47–62.
 Paillot R, Daly J, Juillard V, Minke JM, Hannant D, Kydd JH. Equine interferon gamma synthesis in lymphocytes after in vivo infection and in vitro stimulation with EHV-1. Vaccine 2005;23(36):4541–51.
 Baker DJ. Rationale for the use of influenza vaccines in horses and the importance of antigenic drift. Equine Vet J 1986;18(2):93–6.
 EquiFluNet. Global surveillance network for equine influenza.
 Newton JR, Texier MJ, Shepherd MC. Modifying likely protection from equine influenza vaccination by varying dosage intervals within the Jockey Club Rules of Racing. Equine Vet Educ
 Mumford JA, Wood J. Establishing an acceptability threshold for equine influenza vaccines. Dev Biol Stand 1992;79:137–46.
 Mumford JA, Jessett DM, Dunleavy U, Wood JLN, Hannant D, Sundquist B, et al. Antigenicity and immunogenicity of experimental equine influenza ISCOM vaccines. Vaccine 1994;12(9):857–63.
 Vogel FR, Powell MF, Alving CR. A compendium of vaccine adjuvants and excipients. 2nd ed. Bethesda, MD: National Institute of Allergy & Infectious Diseases; 1998.
 ChemIDplus. Polyacrylic acid, RN 9003-01-4.
 Hannant D, Easeman R, Mumford JA. Equine mucosal immune system: intranasal vaccination with inactivated equine influenza virus protects from infection. In: Wernery U, Wade JF, Mumford JA, Kaaden OR, editors. Equine infectious diseases. VIII. Proceedings
of the Eigth International Conference on Equine Infectious Diseases. Newmarket, UK: R&W Publications (Newmarket); 1999. p. 118–9.
 Sanders MT, Brown LE, Deliyannis G, Pearse MJ. ISCOM-based vaccines: the second decade. Immunol Cell Biol 2005;83(2):119–28.
 Mumford JA, Jessett DM, Rollinson EA, Hannant D, Draper ME. Duration of protective efficacy of equine influenza immunostimulating complex/tetanus vaccines. Vet Rec 1994;134:158–62.
 Crouch CF, Daly J, Hannant D, Wilkins J, Francis MJ. Immune responses and protective efficacy in ponies immunised with an equine influenza ISCOM vaccine containing an ‘American lineage’ H3N8 virus. Vaccine 2004;23(3):418–25.
 Crouch CF, Daly J, Henley W, Hannant D, Wilkins J, Francis MJ. The use of a systemic prime/mucosal boost strategy with an equine influenza ISCOM vaccine to induce protective immunity in horses. Vet Immunol Immunopathol 2005.
 Rimmelzwaan GF, Nieuwkoop N, Brandenburg A, Sutter G, Beyer WE, Maher D, et al. A randomized, double blind study in young healthy adults comparing cell mediated and humoral immune responses induced by influenza ISCOM vaccines and conventional vaccines. Vaccine 2000;19(9/10):1180–7.
 Fynan EF, Webster RG, Fuller DH, Haynes JR, Santoro JC, Robinson HL. DNA vaccines: protective immunizations by parenteral, mucosal, and gene-gun inoculations. Proc Natl Acad Sci USA 1993;90(24):11478–82.
 Webster RG, Fynan EF, Santoro JC, Robinson H. Protection of ferrets against influenza challenge with a DNA vaccine to the haemagglutinin. Vaccine 1994;12(16):1495–8.
 Olsen CW, McGregor MW, Dybdahl-Sissoko N, Schram BR, Nelson KM, Lunn DP, et al. Immunogenicity and efficacy of baculovirusexpressed and DNA-based equine influenza virus hemagglutinin vaccines in mice. Vaccine 1997;15(10):1149–56.
 Lunn DP, Soboll G, Schram BR, Quass J, McGregor MW, Drape RJ, et al. Antibody responses to DNA vaccination of horses using the influenza virus hemagglutinin gene. Vaccine 1999;17(18):2245–58.
 Soboll G, Nelson KM, Leuthner ES, Clark RJ, Drape R, Macklin MD, et al. Mucosal co-administration of cholera toxin and influenza virus hemagglutinin-DNA in ponies generates a local IgA response. Vaccine 2003;21(21/22):3081–92.
 Mumford JA, Jessett DM, Hannant D, Daly J, Uppal P, Yadav MP, et al. A model system for investigating immunity stimulated by live equine influenza vaccines. In: Wernery U, Wade JF, Mumford JA, Kaaden OR, editors. Equine infectious diseases. VIII. Proceedings of the Eigth International Conference on Equine Infectious Diseases. Newmarket, UK: R&W Publications (Newmarket); 1998. p. 44–9.
 Banbura MW, Kawaoka Y, Thomas TL, Webster RG. Reassortants with equine 1 (H7N7) influenza virus hemagglutinin in an avian influenza virus genetic background are pathogenic in chickens. Virology 1991;184(1):469–71.
 Holmes DF, Lamb LM, Coggins L, Murphy BR, Gillespie JH. Live temperature senseitive equine-2 influenza A virus vaccine : production and efficacy in experimental ponies. In: Plowright W, Rossdale PD, Wade JF, editors. Equine infectious diseases. VI. Proceedings of
the Sixth International Conference. Newmarket: R&W Publications Ltd.; 1991. p. 253–8.
 Holmes DF, Lamb LM, Anguish LM, Coggins L, Murphy BR, Gillespie JH. Live temperature-sensitive equine-1 influenza A virus: efficacy in experimental ponies. In: Powell DG, editor. Equine infectious diseases. V. Proceedings of the Fifth International Conference. Lexington, KY: University Press of Kentucky; 1988. p. 88–93.
 Youngner JS, Whitaker-Dowling P, Chambers TM, Rushlow KE, Sebring R. Derivation and characterization of a live attenuated equine influenza vaccine virus. Am J Vet Res 2001;62(8):1290–4.
 Townsend HG, Penner SJ, Watts TC, Cook A, Bogdan J, Haines DM, et al. Efficacy of a cold-adapted, intranasal, equine influenza vaccine: challenge trials. Equine Vet J 2001;33(7):637–43.
 Chambers TM, Holland RE, Tudor LR, Townsend HG, Cook A, Bogdan J, et al. A new modified live equine influenza virus vaccine: phenotypic stability, restricted spread and efficacy against heterologous virus challenge. Equine Vet J 2001;33(7):630–6.
 Lunn DP, Hussey S, Sebing R, Rushlow KE, Radecki SV, Whitaker-Dowling P, et al. Safety, efficacy, and immunogenicity of a modified-live equine influenza virus vaccine in ponies after induction of exercise-induced immunosuppression. J Am Vet Med Assoc 2001;218(6):900–6.
 Paoletti E. Applications of pox virus vectors to vaccination: an update. Proc Natl Acad Sci USA 1996;93(21):11349–53.
 Moss B. Genetically engineered poxviruses for recombinant gene expression, vaccination, and safety. Proc Natl Acad Sci USA 1996;93(21):11341–8.
 Minke JM, Siger L, Karaca K, Austgen L, Gordy P, Bowen R, et al. Recombinant canarypoxvirus vaccine carrying the prM/E genes of West Nile virus protects horses against a West Nile virus-mosquito challenge. Arch Virol Suppl 2004;(18):221–30.
 Minke JM, Fischer L, Baudu P, Guigal PM, Sindle T, Mumford JA, et al. Use of DNA and recombinant canarypox (ALVAC) vectors for equine herpes vaccination. Vet Immunol Immunopathol 2005, in press.
 Dale B, Brown R, Kloss JM, Cordell B, O’Moore B, Yilma T. Generation of vaccinia virus-equine influenza A virus recombinants and their use as immunogens in horses. In: Powell DG, editor. Equine infectious diseases. V. Proceedings of the Fifth International Conference.
Lexington, KY: University Press of Kentucky; 1988. p. 80–7.
 Breathnach CC, Rudersdorf R, Lunn DP. Use of recombinant modified vaccinia Ankara viral vectors for equine influenza vaccination. Vet Immunol Immunopathol 2004;98(3/4):127–36.
 Breathnach CC, Clark HJ, Clark RC, Olsen CW, Townsend HG, Lunn DP. Immunization with recombinant modified vaccinia Ankara (rMVA) constructs encoding the HA or NP gene protects ponies from equine influenza virus challenge. Vaccine 2006;24(8):1180–90.
 Plotkin SA, Cadoz M, Meignier B, Meric C, Leroy O, Excler JL, et al. The safety and use of canarypox vectored vaccines. Dev Biol Stand 1995;84:165–70.
 Edlund Toulemonde C, Daly J, Sindle T, Guigal PM, Audonnet JC, Minke JM. Efficacy of a recombinant equine influenza vaccine against challenge with an American lineage H3N8 influenza virus responsible for the 2003 outbreak in the United Kingdom. Vet Rec 2005;156(12):367–71.
 Paillot R, Kydd J, Sindle T, Hannant D, Edlund-Toulemonde C, Audonnet JC, et al. Antibody and IFN-. responses induced by a recombinant canarypox vaccine and challenge infection with equine influenza virus. Vet Immunol Immunopathol 2006, in press.
 Bender BS, Rowe CA, Taylor SF, Wyatt LS, Moss B, Small Jr PA. Oral immunization with a replication-deficient recombinant vaccinia virus protects mice against influenza. J Virol 1996;70(9):6418–24.
 Wright PF, Mestecky J, McElrath MJ, Keefer MC, Gorse GJ, Goepfert PA, et al. Comparison of systemic and mucosal delivery of 2 canarypox virus vaccines expressing either HIV-1 genes or the gene for rabies virus G protein. J Infect Dis 2004;189(7):1221–31.
 Welter J, Taylor J, Tartaglia J, Paoletti E, Stephensen CB. Mucosal vaccination with recombinant poxvirus vaccines protects ferrets against symptomatic CDV infection. Vaccine 1999;17(4):308–18.
 Rooney JF, Wohlenberg C, Cremer KJ, Moss B, Notkins AL. Immunization with a vaccinia virus recombinant expressing herpes simplex virus type 1 glycoprotein D: long-term protection and effect of revaccination. J Virol 1988;62(5):1530–4.
 Kanesa-thasan N, Smucny JJ, Hoke CH, Marks DH, Konishi
E, Kurane I, et al. Safety and immunogenicity of NYVACJEV and ALVAC-JEV attenuated recombinant Japanese encephalitis virus—poxvirus vaccines in vaccinia-nonimmune and vacciniaimmune humans. Vaccine 2000;19(4/5):483–91.
 Schickli JH, Flandorfer A, Nakaya T, Martinez-Sobrido L, Garcia-Sastre A, Palese P. Plasmid-only rescue of influenza A virus vaccine candidates. Philos Trans R Soc Lond B Biol Sci 2001;356(1416): 1965–73.
 Bilsel P, Kawaoka Y. Generation of a stably attenuated influenza virus by reverse genetics. In: Nakajima H, Plowright W, editors. Equine infectious diseases. VII. Proceedings of Seventh International Conference on Equine Infectious Diseases. Newmarket, UK: R&W Publications Ltd.; 1994. p. 165–8.
 Quinlivan M, Zamarin D, Garcia-Sastre A, Cullinane A, Chambers T, Palese P. Attenuation of equine influenza viruses through truncations of the NS1 protein. J Virol 2005;79(13):8431–9.
 Kittel C, Sereinig S, Ferko B, Stasakova J, Romanova J, Wolkerstorfer A, et al. Rescue of influenza virus expressing GFP from the NS1 reading frame. Virology 2004;324(1):67–73.
 Proietti E, Bracci L, Puzelli S, Di Pucchio T, Sestili P, De Vincenzi E, et al. Type I IFN as a natural adjuvant for a protective immune response: lessons from the influenza vaccine model. J Immunol 2002;169(1):375–83.
 Maeda Y, Hatta M, Takada A, Watanabe T, Goto H, Neumann G, et al. Live bivalent vaccine for parainfluenza and influenza virus infections. J Virol 2005;79(11):6674–9.
 Liu IK, Pascoe DR, Chang LW, Zee YC. Duration of maternally derived antibodies against equine influenza in newborn foals. Am J Vet Res 1985; 46(10):2078–80.
 Van Oirschot JT, Bruin G, de Boer-Luytze E, Smolders G. Maternal antibodies against equine influenza virus in foals and their interference with vaccination. Zentralbl Veterinarmed B 1991;38(5):391–6.
 van Maanen C, Bruin G, de Boer-Luijtze E, Smolders G, de Boer GF. Interference of maternal antibodies with the immune response of foals after vaccination against equine influenza. Vet Q 1992;14(1):13–7.
 Cullinane A, Weld J, Osborne M, Nelly M, McBride C, Walsh C. Field studies on equine influenza vaccination regimes in thoroughbred foals and yearlings. Vet J 2001;161(2):174–85.
 Wilson WD, Mihalyi JE, Hussey S, Lunn DP. Passive transfer of maternal immunoglobulinisotype antibodies against tetanus and influenza and their effect on the response of foals to vaccination. Equine Vet J 2001;33(7):644–50.
 Conboy HS, Berry DB, Fallon EH, Holland RE, Powell DG, Chambers T. Failure of foal seroconversion following equine influenza vaccination. AAEP Proc 1997;43(22/23).
 Holland RE, Conboy HS, Berry DB, Fallon EH, Powell DG, Tudor LR, et al. Age dependent of foal vaccination for equine influenza: new evidence from the USA. In: Wernery U, Wade JF, Mumford JA, Kaaden OR, editors. Equine infectious diseases. VIII. Proceedings of the Eighth International Conference on Equine Infectious Diseases. Newmarket, UK: R&W Publications (Newmarket); 1999. p. 547–8.
 Lopez AM, Hines MT, Palmer GH, Knowles DP, Alperin DC, Hines SA. Analysis of anamnestic immune responses in adult horses and priming in neonates induced by a DNA vaccine expressing the vapA gene of Rhodococcus equi. Vaccine 2003;21(25–26):3815–25.
 Fermaglich DH, Horohov DW. The effect of aging on immune responses. Vet Clin North Am Equine Pract 2002;18(3):621–30, ix.
 Goto H, Yamamoto Y, Ohta C, Shirahata T, Higuchi T, Ohishi H. Antibody responses of Japanese horses to influenza viruses in the past few years. J Vet Med Sci 1993;55(1):33–7.
 Horohov DW, Kydd JH, Hannant D. The effect of aging on T cell responses in the horse. Dev Comp Immunol 2002;26(1):121–8.
 Horohov DW, Dimock A, Guirnalda P, Folsom RW, McKeever KH, Malinowski K. Effect of exercise on the immune response of young and old horses. Am J Vet Res 1999;60(5):643–7.