Evidence for Seasonal Patterns in the Relative Abundance of Avian Influenza Virus Subtypes in Blue-Winged Teal (Anas discors)

Since the recognition of wild waterbirds as the natural reservoir of influenza A viruses (IAVs; Slemons et al. 1974), significant progress has been made in elucidating seasonal dynamics of infection. The annual congregation during spring of shorebirds and gulls (order Charadriiformes) at Delaware Bay to feed on horseshoe crab (Limulus polyphemus) eggs leads to amplification of IAVs and a peak in virus prevalence at this location along the Atlantic coast of North America (Krauss et al. 2010). In late summer and autumn, large numbers of hatch year, immunologically naïve waterfowl are infected by IAVs, leading to an annual peak in virus prevalence among ducks (order Anseriformes) throughout the northern hemisphere (Hinshaw et al. 1980). Conversely, lower rates of IAV infection have generally been detected in Charadriiformes sampled at locations other than Delaware Bay (Munster et al. 2007) and in ducks sampled in other seasons (Stallknecht et al. 1990). Although seasonal patterns of prevalence of IAVs in reservoir hosts have been described, trends in subtype diversity have received less attention.

Seasonal dynamics of IAV prevalence in wild birds are influenced by susceptibility, density, and immune status of host populations (Hinshaw et al. 1980; Krauss et al. 2010). Homo- and heterospecific immunity may also drive IAV subtype diversity in these populations (Latorre-Margalef et al. 2013). Such immunity could temporally affect the prevalence of specific subtypes. To investigate seasonal trends in IAV subtype diversity, we focused on Blue-winged Teal (Anas discors), an abundant and highly migratory waterfowl species in North America and the Neotropics (Rohwer et al. 2002). We selected this species based on its broad geographic distribution and the extent of interseasonal sampling efforts targeting this taxon relative to other waterfowl species.

Paired cloacal and oropharyngeal swabs were collected from live-captured or hunter-harvested Blue-winged Teal at locations in Canada (New Brunswick, Nova Scotia, and Prince Edward Island), Guatemala, and the US (Louisiana, Minnesota, North Dakota, and Texas) during 2007–13. Pooled swab samples were placed in viral transport media and stored at −80 C until testing. Samples were either first screened using real-time reverse transcriptase PCR (rRT-PCR; Canada and Guatemala samples; Spackman et al. 2002) with all rRT-PCR–positive samples inoculated into embryonating specific-pathogen-free (SPF) eggs for virus isolation (Woolcock 2008) or inoculated directly into SPF eggs without prior screening (US samples; Stallknecht et al. 1990). Harvested allantoic fluid was screened for IAV by rRT-PCR (Spackman et al. 2002) or Flu Detect (Synbiotics, San Diego, California, USA) and positive samples subtyped using serologic (Pedersen 2008a, b) or molecular methods (e.g., Ramey et al. 2010; complete reaction conditions and primer sequences available upon request). Virus isolation and hemagglutinin (HA) and neuraminidase (NA) subtypes were summarized by season: summer/autumn (July–September), winter (November–January), and spring (February–March). Seasons approximate the molt/southward migration, wintering, and northward migration periods, respectively, for Blue-winged Teal (Rohwer et al. 2002).

We compared our results to published accounts of subtype diversity of IAVs isolated from Blue-winged Teal from which season-specific information could be extracted (Hinshaw et al. 1980; Stallknecht et al. 1990; Hanson et al. 2005; Ferro et al. 2010; González-Reiche et al. 2012). Only sample collections in which two or more IAVs were isolated from Blue-winged Teal were used for comparative purposes. Nonparametric Friedman tests, incorporating data from previous studies presenting IAV subtype data from Blue-winged Teal, were used to determine whether the relative frequency of isolation differed across subtypes for each season, with significance assessed after adjusting for ties in rank. Separate tests were conducted for each surface glycoprotein per season with subtype serving as the independent variable and frequency serving as the dependent variable for blocked sampling events.

A total of 568 IAVs were isolated from 6,901 paired swab samples collected from Blue-winged Teal during 22 sampling events in eight states or provinces in 2007–13 (Table 1). The virus isolation rate, calculated as the percentage of paired swab samples from which IAVs were isolated, ranged from 29.9% in summer/autumn in Nova Scotia in 2010 to 0.9% in spring in Texas in 2013 (Table 1). The overall virus isolation rate was 10.3% in summer/autumn, 6.1% in winter, and 2.9% in spring (Table 1).

From the 568 IAVs isolated from 2007–13, 11 HA subtypes and all nine NA subtypes were identified (Fig. 1). The relative frequency of isolation for HA subtypes differed significantly in summer/autumn (chi-square [χ²] = 147.97, df = 11, P<0.001) and in spring (χ² = 35.40, df = 11, P<0.001), but not in winter (χ² = 15.99, df = 11, P = 0.142). The H3 or H4 HA subtypes were predominant or codominant during summer/autumn in 21 of 24 sampling events (Fig. 1); H7 was the predominant HA subtype isolated from paired Blue-winged Teal swab samples collected during spring in five of six sampling events (Fig. 1). The relative frequency of isolation for NA subtypes differed significantly during summer/autumn (χ² = 91.29, df = 8, P<0.001) but not during winter (χ² = 9.39, df = 8, P = 0.311) or spring seasons (χ² = 11.64, df = 8, P = 0.168). The N6 or N8 NA subtype (or both) were the predominant or codominant subtype during summer/autumn in 21 of 24 sampling events (Fig. 2), similar to results for H3 and H4 subtypes for this season.

Seasonal rates of virus isolation from paired Blue-winged Teal swab samples collected during 22 sampling events as part of this study are consistent with previous reports for this species and for other wild ducks sampled in North America (Hinshaw et al. 1980; Stallknecht et al. 1990; Hanson et al. 2003; Krauss et al. 2004; Hanson et al. 2005; Ferro et al. 2010; González-Reiche et al. 2012). The repeated isolation of H3 and H4 subtype IAVs in summer/autumn, regardless of location/year, is also consistent with prior reports for North American ducks sampled during this season (Hanson et al. 2003; Krauss et al. 2004; Ramey et al. 2011). Fewer reports on IAV subtypes exist for wintering waterfowl, and our data support the circulation of subtypes less commonly observed during summer/autumn, including the H14 HA subtype. The H14 HA subtype was only recently detected in North America (Nolting et al. 2012), also having been isolated from wild birds sampled during winter. Even fewer data exist for IAV isolation from wild ducks sampled during spring. The consistent isolation of H7 HA subtype IAVs during this season provides evidence that H7 viruses may be relatively abundant in Blue-winged Teal and perhaps in other species of waterfowl in spring. This finding may be important to both public and domestic animal health, as H7 viruses have recently caused human disease as well as poultry outbreaks of highly pathogenic IAV in Mexico and low pathogenic IAV in the US. Through the identification of seasonal trends in H7 subtype IAVs in North American waterfowl, surveillance efforts can be implemented or expanded in spring to more effectively monitor for viruses of this subtype.

Immune response of avian hosts to NA subtypes may be weaker than for HA (Latorre-Margalef et al. 2013). Thus, the finding of significantly different relative frequencies of isolation for NA subtypes in summer/autumn but not winter or spring may be explained by the affinity for surface glycoproteins to form stable genomic constellations rather than by the antigenicity of this surface glycoprotein (Wagner et al. 2002). The N6 and N8 NA subtypes have frequently been found to be associated with the H4 and H3 HA subtypes, respectively, in IAVs isolated from North American waterfowl (Stallknecht et al. 1990; Krauss et al. 2004; Ferro et al. 2010). Thus, population immunity to HA subtypes may influence seasonal patterns in NA relative abundance. Additional interseasonal sampling of waterfowl in North America and the Neotropics would be useful for better understanding patterns of virus circulation at the subtype level.

We are grateful to many biologists and volunteers for assistance with captures including P. Pauling, P. Oesterle, W. Broussard, A. Fojtik, J. Slagter, N. Davis-Fields, J. Starr, J. LaCour, J. Gray, and K. DeMarco. We acknowledge the Canadian Cooperative Wildlife Health Centre and Environment Canada for providing samples for this study. We thank the City of Beaumont, Texas for access to Cattail Marsh for captures conducted in 2012 and 2013. We appreciate the cooperation of sport hunters for access to harvested birds for sample collection. M. Whalen provided assistance formatting figures for submission. Financial and administrative support was provided by J. Pearce, T. DeGange, P. Bright, K. Briggs, and S. Gross. We appreciate reviews provided by J. Pearce, C. Ely, and two anonymous reviewers. This study was funded, in part, with federal funds provided by the National Institute of Allergy and Infectious Diseases, National Institute of Health, Department of Health and Human Services under contracts HHSN266200700007C and HHSN266200700010C. None of the authors have any financial interests or conflict of interest with this article. Any use of trade names is for descriptive purposes only and does not imply endorsement by the US government.