Antibodies to influenza A virus (H1 and H3) in companion animals in Iowa, USA
- B. M. Seiler, BSc1,
- K-J. Yoon, DVM, PhD, DiplACVM2,
- C. B. Andreasen, DVM, PhD, DiplACVP3,
- S. M. Block, BSc2,
- S. Marsden, BSc3 and
- B. J. Blitvich, PhD1
- Department of Veterinary Microbiology and Preventive Medicine
- Veterinary and Diagnostic Animal Production Medicine
- Department of Veterinary Pathology, College of Veterinary Medicine, Iowa State University, Ames, IA 50011, USA
- Correspondence to Dr Blitvich, e-mail:
INFLUENZA A virus (IAV) is the aetiological agent of an economically important respiratory disease of human beings, horses, pigs and poultry throughout the world (Lipatov and others 2004). Historically, IAV has not been regarded as a major pathogen of dogs or cats. However, in 2004, the equine subtype H3N8 was responsible for an outbreak of respiratory disease in racing greyhounds in Florida, USA (Crawford and others 2005). From 2004 to 2006, outbreaks of H3N8 in racing greyhounds occurred in several other states of the USA, including Iowa (Crawford and others 2005, Yoon and others 2005). A retrospective study performed using archived sera provided evidence that H3N8 had been circulating in the greyhound population in Florida since as long ago as 1999. Another retrospective study revealed that H3N8 was the cause of an outbreak of severe respiratory disease in English foxhounds in the UK in 2002 (Daly and others 2008). The avian subtypes H5N1 and H3N2 recently caused fatal respiratory disease in dogs in Thailand and Korea, respectively (Songserm and others 2006b, Song and others 2008, Lee and others 2009). The first reported case of fatal influenza in a domestic cat with a naturally acquired infection occurred in Thailand in 2004 (Songserm and others 2006a); H5N1 AIV was the aetiological agent. Additionally, H5N1 RNA was detected in dead cats in Iraq in 2006 (Yingst and others 2006), and pandemic H1N1 was isolated from a cat with respiratory signs in Iowa in 2009 (Sponseller and others 2010). In view of the continued burden that IAV places on human and animal health worldwide, a serological investigation was performed to estimate the seroprevalence of IAV in dogs and cats in Iowa.
A total of 953 serum samples were collected from a convenience population consisting of 731 dogs and 222 cats. The animals were sampled between January and June 2009 at the Department of Veterinary Pathology, College of Veterinary Medicine, Iowa State University. Most of the animals (84 per cent) were clinically ill at the time of presentation; the reasons for presentation were as follows: non-respiratory systemic illness (45 per cent of the total study population), surgery (19 per cent), neoplastic disease (8 per cent), skin disease (5 per cent), ocular disease (4 per cent), cardiac disease (2 per cent) and respiratory disease (1 per cent). The remainder (16 per cent) of the animals were healthy and were presented for dental, vaccination or breeding examinations. All the animals resided in Iowa.
All the sera were tested for antibodies to IAV using a species-independent, epitope-blocking ELISA (bELISA) as previously described by Sullivan and others (2009). Briefly, the assay utilises the IAV nucleoprotein-specific monoclonal antibody (mAb) clone A1 (Millipore), and recombinant IAV nucleoprotein (Imgenex). The IAV nucleoprotein is well conserved (Gorman and others 1990), and thus the bELISA can detect antibodies to all IAV subtypes. The ability of each serum sample to block the binding of the mAb to IAV antigen was compared with the blocking ability of serum without antibody to IAV (Equitech-Bio). The data were expressed as relative percentages, and inhibition values of 32 per cent or more were considered to indicate the presence of antibodies to IAV. Recent studies have shown that this assay is a rapid and reliable serological technique for the detection of IAV-specific antibodies in taxonomically diverse mammalian and avian species (Sullivan and others 2009). All sera with bELISA antibodies to IAV were further tested using a competitive ELISA (FlockChek Avian Influenza MultiS-Screen Antibody Test Kit; IDEXX Laboratories), and by haemagglutination inhibition (HI) (Pedersen 2008a) and neuraminidase inhibition (NI) tests (Pedersen 2008b). HI tests were performed using avian H1 to H16, canine H3, equine H3 and H7, human H1 (pandemic strain) and swine H1 and H3 strains (Table 1). Sera were tested at a starting dilution of 1:10, and HI titres of 20 or more were considered to indicate the presence of antibodies to IAV. NI tests were performed using avian N1 to N7 and N9, and equine N8 strains (Table 1). Before the HI and NI tests, the sera were pretreated with receptor-destroying enzyme to remove non-specific inhibitors, using published protocols (Dowdle and others 1979).
Antibodies to IAV were detected by bELISA in serum samples from 15 (2.1 per cent) dogs and two (0.9 per cent) cats (Table 2). None of the seropositive animals had a history of respiratory illness. Twelve dogs and one cat that were positive by bELISA also had antibodies to IAV by the HI and/or NI test (Tables 2, 3). One dog had been infected with more than one IAV subtype, as indicated by the detection of antibodies to both H1 and H3 antigens. Eight other dogs also had antibodies to H1 antigen: one was seropositive for H1N1 and seven had no detectable antibodies to any of the N antigens (the subtype in these dogs has been denoted as H1N? [Table 3]). Another dog was seropositive for H3N8, and two others had antibodies to H3 but no detectable antibodies to any of the N antigens (denoted H3N?). One cat had antibodies to H1 antigen and was negative in all of the NI tests (denoted H1N?). HI titres were usually highest when A/Swine/Iowa/13281/2007 (H1N2) or A/Canine/Iowa/13628/2005 (H3N8) was used as the reagent, suggesting that the majority of seropositive animals had been infected with an IAV of mammalian origin. Three dogs and one cat with bELISA antibodies to IAV were negative by HI, NI and competitive ELISA, indicating that the bELISA results could represent false positives.
The seroprevalence of IAV in the dogs in this study was low. Low rates of seropositivity have also been reported in domestic dogs in Canada (0.4 per cent) (Kruth and others 2008) and Korea (0.5 per cent) (Lee and others 2009). Other studies have reported much higher (19 to 58 per cent) rates of seropositivity, although most dogs in those studies had signs of respiratory disease or were temporally and spatially associated with sick dogs (Crawford and others 2005, Lee and others 2009).
The seroprevalence of IAV in the cats in the present study was also low. In Italy, all of 196 cats sampled between 1999 and 2005 were negative for all subtypes of IAV (Paltrinieri and others 2007). Furthermore, all of 171 cats sampled in Germany and Austria in 2006 were negative for antibodies to H5N1 despite residing in areas where H5N1-infected birds had been found (Marschall and others 2008). In contrast, eight of 111 (7.2 per cent) cats in Thailand sampled in 2005 had antibodies to H5N1 (Butler 2006), as did three of 40 (7.5 per cent) cats in an animal shelter in Austria in 2006 that had housed a H5N1-infected swan (Leschnik and others 2007).
In summary, antibodies to IAV were detected in a sample of domestic dogs and cats in Iowa. In view of the growing number of cases of IAV infection in domestic mammalian species that are traditionally regarded as not being at risk of disease associated with IAV, the continual burden that this virus places on human and animal health worldwide, and reports of interspecies transmission of the virus to companion animals from human beings (Sponseller and others 2010), continued surveillance for IAV in domestic animal populations is clearly warranted.
This study was supported by a cooperative agreement with the United States Department of Agriculture and by grants from Pfizer Animal Health and the Healthy Livestock Initiative. The authors thank the staff of the Veterinary Diagnostic Laboratory at Iowa State University for their support.
- British Veterinary Association