Influenza A Viruses in Whistling Ducks (Subfamily Dendrocygninae)

As compared to other Anseriformes, data related to influenza A virus (IAV) detection and isolation, and IAV antibody detection in whistling ducks (Dendrocygna spp. and Thalassornis leuconotus; subfamily Dendrocygninae) are limited. To better evaluate the potential role of whistling ducks in the epidemiology of IAV, we (1) conducted surveillance for IAV from black-bellied whistling ducks (BBWD, Dendrocygna autumnalis) sampled in coastal Louisiana, USA, during February 2018 and 2019, and (2) reviewed the published literature and Influenza Resource Database (IRD) that reported results of IAV surveillance of whistling ducks. In the prospective study, from 166 BBWD sampled, one H10N7 IAV was isolated (0.6% prevalence), and overall blocking enzyme-linked immunosorbent assay (bELISA) antibody seroprevalence was 10%. The literature review included publications and data in the IRD from 1984 to 2020 that reported results from nearly 5000 collected samples. For any given collection, the IAV isolation rate never exceeded 5.5%, and seroprevalence estimates ranged from 0 to 42%. Results from our prospective study in Louisiana are consistent with this historic literature; however, although all data consistently demonstrated a low prevalence of infection, the potential role of this species in the epidemiology of IAV should not be totally discounted. In sum, whistling ducks can be infected with IAV, they represent important species on many areas where waterfowl winter, and their distribution across the globe appears to be changing.


Introduction
Wild waterfowl in Order Anseriformes are important natural reservoirs for influenza A virus (IAV), but there have been limited reports of IAV in species of whistling ducks (WD; Family Anatidae, Subfamily Dendrocygninae). One of the eight currently recognized species of WD, the black-bellied whistling duck (BBWD; Dendrocygna autumnalis) is common in the Americas. BBWD breed throughout Louisiana, Texas, Mexico, and Central and South-central America, and form large flocks when not breeding, often in association with fulvous whistling ducks (FUWD; Dendrocygna bicolor) [1]. BBWD are largely residential apart from local movement, and tropical populations of BBWD are migratory with movement in the winter months [1]. The distribution of the species is expanding, and BBWD are now abundant along the southeastern coast of the United States [2]. Nocturnal feeders, BBWD forage on seeds, aquatic invertebrates, and vegetation in shallow freshwater [3].
In addition to the BBWD and FUWD, six other WD species are recognized worldwide in genus Dendrocygna, including Dendrocygna arborea (West Indian WD); Dendrocygna arcuata (wandering WD), Dendrocygna eytoni (plumed WD), Dendrocygna guttata (spotted WD), In addition to the BBWD and FUWD, six other WD species are recognized worldwide in genus Dendrocygna, including Dendrocygna arborea (West Indian WD); Dendrocygna arcuata (wandering WD), Dendrocygna eytoni (plumed WD), Dendrocygna guttata (spotted WD), Dendrocygna javanica (lesser WD; LEWD), and Dendrocygna viduata (white-faced WD; WFWD); the white-backed duck (Thalassornis leuconatus) is also classified in the Dendrocygninae subfamily. Broadly, WD primarily exhibit only localized movements within their breeding ranges, with some exceptions. Small populations of LEWD in Asia have been noted to leave their northernmost breeding areas in the winter [4], and the globally widespread FUWD are migratory only in the most northern portions of their Mexican range [5].
Though infrequently sampled as compared to some other species in Order Anseriformes, exposure to and infection of IAV has been detected in resident and migrating Dendrocygninae, mostly outside of the United States. To better understand the role of WD in the ecology of IAV, we (1) conducted IAV surveillance of wintering BBWD along the Gulf coast of the United States over two years, and (2) reviewed published literature and the Influenza Resource Database (IRD) [6] for results from worldwide surveillance for IAV in this avian subfamily.

Virus Isolation and Antibody Detection
Cloacal and oropharyngeal (COP) swabs and serum were opportunistically collected from live-captured BBWD in the Louisiana coastal parishes of Vermilion and Jefferson in February 2018 and 2019, respectively ( Figure 1). In each year, approximately 40% of the sampled birds were after hatch year (AHY). Collection sites included rice fields (2018) and a cargo ship terminal situated on the Mississippi River, where thousands of BBWD and other waterfowl were observed (personal observation, Paul Link, 2019). Virus isolation (VI) was performed on COP swabs as previously described [7]. Viral RNA was extracted (QIAamp Viral RNA Mini Kit, Qiagen Inc., Valencia, CA, USA) from all VI-positive samples following the manufacturer's recommendation, and tested by IAV matrix real-time reverse transcriptase PCR (RRT-PCR) [8]. Whole-genome sequencing of IAV-positive isolates was carried out at J. Craig Venter Institute. Phylogenetic analyses were conducted using MEGA version 6 [9]; the top 2 NCBI basic local alignment search tool (BLAST) results, based on percent nucleotide (nt) similarity, were chosen on the basis of the criteria that they were from viruses identified Collection sites included rice fields (2018) and a cargo ship terminal situated on the Mississippi River, where thousands of BBWD and other waterfowl were observed (personal observation, Paul Link, 2019). Virus isolation (VI) was performed on COP swabs as previously described [7]. Viral RNA was extracted (QIAamp Viral RNA Mini Kit, Qiagen Inc., Valencia, CA, USA) from all VI-positive samples following the manufacturer's recommendation, and tested by IAV matrix real-time reverse transcriptase PCR (RRT-PCR) [8]. Whole-genome sequencing of IAV-positive isolates was carried out at J. Craig Venter Institute. Phylogenetic analyses were conducted using MEGA version 6 [9]; the top 2 NCBI basic local alignment search tool (BLAST) results, based on percent nucleotide (nt) similarity, were chosen on the basis of the criteria that they were from viruses identified from avian species at times predating the query sequence. Sterile blood samples were collected at a total volume less than or equal to 1% of the bird body mass via the brachial wing vein. Serum fractions were tested by nucleoprotein (NP) blocking enzyme-linked immunosorbent assay (bELISA; IDEXX, Westbrooke, ME, USA); samples with S/N ratios Viruses 2021, 13,192 3 of 10 <0.7 were considered seropositive [10,11]. Animal work was approved by UGA Institutional Animal Care and Use Committee approval A2016 05-020-Y3-A6.

Literature Review
We searched over 3 decades (1984-2020) worth of publications on IAV surveillance in wild birds in subfamily Dendrocygninae: D. arcuata, D. autumnalis, D. bicolor, D. eytoni, D. javanica, D. viduata, and T. leuconatus. Prevalence data and metadata, including sampling country, species, sampling season, and year, and testing method(s) were extracted from each publication as available. Google Scholar (http://scholar.google.com) and PubMed (www.ncbi.nlm.nih.gov/pubmed, accessed on 1 September 2020) were searched from 1 September 2020 to 8 September 2020. Data were also obtained from the National Institute of Allergy and Infectious Diseases (NIAID) Influenza Research Database (IRD) [6] through the website of http://www.fludb.org, accessed on 15 October 2020; every attempt was made to assure that results reported from this outlet are not in publication elsewhere.

Prospective-Virus Isolation and Antibody Detection
In total, 92 (2018) and 74 (2019) BBWD COP swab and serum samples were tested by both VI and NP bELISA. One H10N7 IAV was isolated in 2018 (A/black-bellied whistling duck/Louisiana/UGAI18-484/2018); no viruses were isolated in 2019. The gene sequences of the H10 hemagglutinin and N7 neuraminidase surface glycoproteins were most closely related to those of A/mallard/Ohio/16OS0620/2016 (H10N7) ( Table 1). All internal genes were closely related to recently identified North American wild waterfowl sequences (Table 1). Three serum samples were NP bELISA-positive in 2018 (3.3% seroprevalence), and 14 in 2019 (18.9% seroprevalence), suggesting previous exposure to avian IAV. Seroprevalence in AHY birds across both years was 12.5% compared to 9.7% in hatch year (HY) birds.

Retrospective-Literature Review
As compared to some anatids, especially several Anas and Anser species, IAV surveillance has been fairly limited in Dendrocygninae, and often is the result of opportunistic sampling. Results of virus isolation (4942 samples) and/or PCR (4837 samples) testing were reported in 18 publications and include collection efforts of WD in Africa, Asia, Australia, and North and South America ( Table 2). The overall PCR positivity rate was 1.7% (0-100% range), and the overall mean prevalence based on virus isolation was 0.1% (0-5.6%, range). In western Africa, 38 of 1269 (3.0%) whistling ducks tested IAV matrix-positive by PCR [12]. Cumming et al. [13] surveyed 165 avian species in southern Africa; IAV prevalence was 5.2% for birds in subfamily Dendrocygninae, twice as high as that reported from other ducks in the same study. In west Africa, highly pathogenic H5 genomes were identified in apparently healthy WFWD [14]. In the Americas, only three species of whistling ducks are known: BBWD, FUWD, and WFWD. Although infrequently sampled, IAV has been detected in each of these species. Influenza A viral RNA was detected in 14/489 (2.9%) BBWD and 5/71 (7.0%) WFWD in Colombia [15], and low pathogenicity H5N2 viruses were isolated from each of these hosts. In Texas, IAV was detected in 2/18 overwintering FUWD, resulting in the isolation of an H6N1 subtype IAV [16]. Five publications reported serological results from WD. Primarily on the basis of ELISA testing, we found that seroprevalence ranged from 0 to 42% (Table 3).  In addition to the published literature, the IRD included 4248 IAV results for WD sampled from 2007 to 2019 (Table 4). The majority of WD surveillance submissions to IRD (3655, 86%) were from LEWD sampled on the Asian continent. Of these, 223 (6.1%) were reported as IAV-positive.

Discussion
Results from efforts reported here indicate that BBWD are exposed to (seroprevalence range 3.3%-18.9%) and can be infected by (isolation of H10N7) avian IAV; the observed prevalence of IAV infection as determined by virus isolation (0.6%) was found to be low but consistent with prevalence estimates for other wintering duck species elsewhere in the Americas [7,[33][34][35]. However, the NP antibody prevalence described in this study was substantially lower than that observed in mallard and teal species collected at sites in Louisiana at the same time of year (average 65%, 2018-2019; unpublished data). Low apparent seroprevalence in these resident WD may represent limited lifetime exposure to IAV due to this species' limited geographic range within waterfowl wintering habitats in North America; this is supported by the low overall seroprevalence in adult (AHY) BBWD in this study (12.5%, average). Unlike mallard and teal species, WD are not present on traditional waterfowl breeding and staging areas during fall migration where IAV prevalence is consistently at its highest [34]. Disparate behavior and foraging strategies, largely localized movement, single species flocks, and differential susceptibility to IAV may also limit IAV exposure and play a role in the limited detection of antibodies we report in this resident species. Given that the ability of avian hosts to respond to exposure to IAV varies by species [36], differences in the ability of BBWD to mount a measurable immune response, as compared to other anatids, also cannot be discounted.
Travelling in large flocks and globally abundant on many waterfowl wintering areas, species of WD have been shown to be infected with similar IAV as migratory waterfowl. Further, the detection of highly pathogenic avian influenza (HPAI) virus in healthy WFWD in Africa [14] emphasizes the role that WD may have in the distribution and possibly the maintenance of HPAI and other pathogens of concern. Taken together, these results indicate that Dendrocygninae may play a role in the ecology and epidemiology of IAV, but given that they have been understudied in comparison to some anatid species, their potential importance in the natural history of IAV remains unknown. Results here, which are consistent with historic reports, demonstrate a low prevalence of infection in all WD species surveyed. Given that WD can be infected, the fact that they represent important species on many areas where waterfowl winter, and the fact that they have changing distribution in some parts of the world, the potential role of these species in the epidemiology of IAV should not be totally discounted.  Table S1: Antibody detection (bELISA) and virus isolation results for serum and swab samples collected in this study.
Funding: Funding was provided by the CEIRS, National Institute of Allergy and Infectious Diseases (NIAID), National Institutes of Health (NIH), Department of Health and Human Services (HHS), under contract HHSN272201400006C. The Influenza Research Database has been wholly funded with federal funds from NIAID, NIH, and HHS under contract no. 75N93019C00076. This project has also been funded in part with federal funds from the NIAID, NIH, and HHS under award number U19AI110819. The funding agencies were not involved in the design, implementation, or publishing of this study and the research presented herein represents the opinions of the authors, but not necessarily the opinions of the funding agencies.

Institutional Review Board Statement:
The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board (or Ethics Committee) of the University of Georgia Institutional Animal Care and Use Committee approval A2016 05-020-Y3-A6.

Informed Consent Statement: Not applicable.
Data Availability Statement: The data presented in this study are available in Supplementary  Table S1.