Influenza A virus prevalence in duck populations

Based on IA virus isolation, the prevalence of infected waterfowl was 13.1±5.8% and 9.9±1.2% in 2006 and 2007, respectively. The H4N6 subtype represented 5.9±11.2% of the total viruses isolated from ducks in 2006 (1 of 17 isolated viruses, unpublished data), but was the predominant subtype in 2007 (20.7±5.3%; 46 of 222 isolated viruses [23]). Similarly, the H3N8 subtype represented 15.3±4.7% of the total viruses isolated in 2007 (34 of 222 isolated viruses [23]) and was the predominant subtype in 2008 (26±4.1%; 114 of 438 isolated viruses [23]).

Genetic characterization

The genetic comparisons of the two water isolates showed variable percentages of similarity for the internal segments: PB2: 91%; PB1: 91%; PA: 87%; NP: 95%; M: 98%; NS: 71%. Results from the phylogenetic analysis indicate that, except for the Matrix gene, these genes belonged to different genetic lineages (Figures S1, S2, S3, S4, S5, S6). When considering the internal segments, both viruses showed high genetic relatedness with a large diversity of IA virus subtypes circulating in wild waterbirds in Minnesota during the same season. In particular, A/Surface water/Minnesota/W07-2241/2007 (H3N8) was closely related to different virus subtypes (H1N3, H3N8, H4N6, H6N6, H10N1, H10N3, H10N6, H11N2, H11N9) for PB1, PA and M genes and A/Surface water/Minnesota/NW1-T/2006 (H4N6) was genetically related to A/Mallard/MN/464334/2006 (H5N2) for PB2 and NP genes (99% and 96% similarity, respectively).

Analysis of the HA and NA segments provided additional information concerning the circulation of these two subtypes in wild waterbirds in North America (Figures 1 and S7, S8, S9, S10). A/Surface water/Minnesota/NW1-T/2006 (H4N6) was closely related to A/Blue-winged Teal/TX/Sg-00157/2007 (H4N6) for both genes. For A/Surface water/Minnesota/W07-2241/2007 (H3N8), the HA was closely related to H3N6 and H3N8 viruses isolated in Alaska and Alberta the previous years (2005 and 2006) and, for the NA, to a virus isolated in 2008 in North-Dakota: A/Blue-winged Teal/ND/Sg-00739/2008 (H3N8).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Genetic sub-lineages including the two water-isolated viruses, for the HA and NA genes.

Posterior probability values are indicated when higher than 0.95. Complete maximum clade credibility trees are available in Figures S7, S8, S9 and S10.

https://doi.org/10.1371/journal.pone.0026566.g001

Persistence under laboratory conditions

Results of viral persistence trials are presented in Table 1 (adjusted R² and P values are detailed in Table S1). Overall, increasing temperature decreased virus persistence. A slight decrease of viral titer was observed at 4°C; both viruses were detected 365 days post-inoculation. Infectivity was quickly reduced when exposed to repeated freeze-thaw cycles (−20°C/4°C); however, viruses remained viable when the temperature was changed between 17°C and 23°C, consistent with the results of maintenance at constant temperatures (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Estimated time (in days) required for a decrease of the viral titer by 1 log₁₀ TCID₅₀/mL.

https://doi.org/10.1371/journal.pone.0026566.t001

Replication in Mallards

None of the birds were shedding virus at the time of inoculation and all tested negative for IA virus antibodies. Both viruses were isolated from all ducks from day 1 to day 5 post-inoculation and intermittently from day 7 to 14 post-inoculation (Table 2). Viruses were isolated from both oropharyngeal and cloacal swabs. Real-time RT-PCR results were consistent with viral isolation results for the first five days post-inoculations, except for day 1 for A/Surface water/Minnesota/W07-2241/2007 (H3N8) and day 5 for one oropharyngeal swab for A/Surface water/Minnesota/NW1-T/2006 (H4N6).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Virus shedding pattern based on oropharyngeal (OP) and cloacal (CL) swab samples.

https://doi.org/10.1371/journal.pone.0026566.t002

For A/Surface water/Minnesota/W07-2241/2007 (H3N8), four of the five ducks were ELISA positive at day 4 post-inoculation and all ducks were positive at days 7, 10 and 14. All ducks inoculated with A/Surface water/Minnesota/NW1-T/2006 (H4N6) tested positive for IA antibody from day 4 until the end of the experiment (day 14).

Viruses

Both viruses were isolated from surface lake water in Minnesota, USA: A/Surface water/Minnesota/NW1-T/2006 H4N6 was isolated from a water sample collected in Thief Lake Wildlife Management Area (48°30′40″N, 95°55′12″W), in September 18, 2006; and A/Surface water/Minnesota/W07-2241/2007 H3N8 from a sample collected in Agassiz National Wildlife Refuge (48°18′58″N, 96°00′23″W), in October 28, 2007. Information related to water sample collection and virus isolation have been previously described [18]. Second (H4N6) or third (H3N8) passages of stock viruses were propagated in 9 to 11 days old specific pathogen free (SPF) embryonating chicken eggs. Amnio-allantoic fluids were harvested 5 days post infection and were titrated in both 9 to 11 days old SPF embryonating chicken eggs and Madin Darby canine kidney (MDCK) cells [3]; titers were determined according to the method described by Reed and Muench [26].

Influenza A virus prevalence in duck populations

Prevalence estimates and subtype diversity were available from ongoing long-term population studies of wild duck populations in Northwestern Minnesota [23]. In total, 130 and 2246 cloacal samples were collected from wild ducks in 2006 and 2007, respectively. Sampling was performed in September in 2006 and from July to October in 2007. Detailed method for sample collection, virus isolation and subtyping were previously described [23]. In brief, cloacal swabs were obtained using sterile cotton-tipped applicators (Puritan, Medical Products Company LLC, Guilford, ME) and placed in 2 ml of Brain Heart Infusion media (Becton Dickinson and Co., Sparks, MD) supplemented with penicillin G (1,000 units/ml), streptomycin (1 mg/ml), kanamycin (0.5 mg/ml), gentamicin (0.25 mg/ml), and amphotericin B (0.025 mg/ml) (Sigma Chemical Company, St. Louis, MO). Samples were stored at 4°C (24–72 hrs) and then frozen at −80°C until processed. Each sample was inoculated into four 9–11 day old SPF embryonated chicken eggs via the allantoic route (0.25 ml/egg). Amnio allantoic fluid were tested by hemagglutination assay and positive samples were confirmed by RT-PCR [27]. Subtyping was done at the National Veterinary Services Laboratories (Ames, Iowa) using hemagglutination inhibition and neuraminidase inhibition tests and via genotyping of the HA and NA genes, at the University of Minnesota, St. Paul, Minnesota and the University of California Davis, using standard Sanger sequencing for HA and NA segments.

For virus and subtype prevalences, 95% confidence intervals (CI) were calculated with the equation for the normal approximation for the binomial CI (CI = 1.96√(p(1−p)/n; where p is the observed prevalence and n is the sample size).

Full-genome sequencing

Sample preparation, viral RNA extraction, and amplification of all segments of the genome in a single step multiplex PCR were performed as described by Zhou et al. [28] with a modification of single base in primer MBTuni-12 (the new primer is MBTuni-12(M): 5′- ACGCGTGATCAGCRAAAGCAGG-3′). cDNA libraries were prepared from PCR-purified multiplex amplicons using QIAquick PCR Purification kit (Qiagen, Valencia, California USA), following manufacturer's instructions, and subjected to adaptor ligation and whole genome sequencing using 454 FLX technology using protocols described by Ramakrishnan et al. [29]. Obtained DNA sequences were analyzed as previously described [29]. In brief, all the sequencing reads were aligned against influenza genomes using BlastN algorithm of NCBI BLAST 2.2.16. The ‘non influenza’ sequences were filtered out and only influenza reads were assembled in GS De nova Assembler 2.0.00.20 and mapped in GS Reference Mapper 2.0.00.20. The influenza contigs obtained using the above software were reassembled in Sequencher Version 4.9 (Genecodes) and annotated based on BLAST analyses.

Sequences generated in this study have been deposited in GenBank under the accession numbers CY073707 to CY073714 (A/Surface water/Minnesota/NW1-T/2006) and CY073717 to CY073724 (A/Surface water/Minnesota/W07-2241/2007).

Phylogenetic analysis

Complete nucleotide sequences of the 8 segments of IA viruses isolated in wild waterbirds in North America were retrieved from the Influenza Sequence Database [30]. For computational reasons, we selected only sequences obtained from viruses isolated in Minnesota for internal segments (PB2, PB1, PA, NP, MP and NS). For the HA and NA segments, the dataset was extended to viruses isolated at other sites in North America. For each sequence, the following information was collected: subtype, geographic origin (country and state), strain name, bird host species and year of collection. Analysis was restricted to viruses isolated from wild avian species belonging to the Anseriformes and Charadriiformes orders, as they are recognized to be the natural host reservoir for IA viruses. Sequences for which bird species were not identified (e.g., ‘duck’ or ‘bird’, without any additional information concerning their origin) were not included. Duplicate sequences from the same strains, undefined subtypes and mixed infections, as well as sequences previously identified as reflecting potential laboratory errors [31], also were not included. Sequences were aligned with CLC sequence viewer 6.3 (CLC bio, Aarhus, Denmark) and edited in order to restrict the alignment to the coding region for each gene. The definitive data sets included 2121 nucleotide sequences of viruses isolated between 1976 and 2009 (c.f. Table S2 and S3 for details).

Phylogenetic trees were generated with a Bayesian Markov chain Monte Carlo (MCMC) coalescent approach, implemented in the program BEAST 1.5.3 [32]. The Shapiro-Rambaut-Drummond-2006 (SRD06) nucleotide substitution model was used in all simulations as this model is recognized to provide better resolution for coding regions [33]. Three molecular clock models were tested for each gene: the strict clock (SC) that assumes a single evolutionary rate in the phylogenetic trees and two relaxed clocks: the uncorrelated exponential (UE) and the uncorrelated lognormal (UL) that allows evolutionary rates to vary along branches, within an exponential or lognormal distribution [34]. Molecular clock models were evaluated and tested with the Bayes Factor (BF) [35], [36] implemented in the program TRACER 1.5.0 [37]. The ratio of marginal likelihoods were compared between models and BF significance was determined from the values of 2 ln(BF) as described in Brandley et al. [38]. A Bayesian skyline coalescent tree prior was used in all simulations as it makes the fewest a priori assumptions about the data [39] and has been shown to be more appropriate to describe the population dynamics of IA virus [40]. Three independent analysis were performed (with the SC, UE and UL molecular clocks) for each segment, with a chain length of 140–320 million generations sampled every 1000 iterations. Results were analyzed with TRACER 1.5.0, phylogenetic consensus trees were produced using the program TREEANNOTATOR 1.5.3 and edited for generation of figure captions with the program FigTree 1.3.1.

Persistence in distilled water

Distilled water buffered with 10 mM HEPES was adjusted with 1 N solutions of NaOH or HCl to provide 3 different pH conditions (6.2, 7.2 and 8.2). Five constant temperatures (4°C, 10°C, 17°C, 23°C and 28°C) were tested at pH = 7.2. Two additional pH conditions (6.2, 8.2) were tested at 17°C and 28°C. For both viruses, infective amnio-allantoic fluids were diluted 1∶100 in the water samples. Inoculated water samples were divided into 2 mL aliquots in 5 mL polystyrene round-bottom tubes and placed in incubators set to the appropriate temperature. The effect of temperature changes was also investigated at two conditions (pH = 7.2): −20°C/4°C and 17°C/23°C. Inoculated water samples were transferred daily into the appropriate incubators in order to reflect cyclic temperature changes: samples were maintained 14 hours under low temperatures (−20°C and 17°C) and 10 hours under high temperature (4°C and 23°C).

For each treatment, the viral-inoculated water was sampled at the time of inoculation (day 0) and between 9 to 23 times post-inoculation. Sampling frequency varied between treatments (from 1 to 28 days) and was adjusted during the course of the experiment to ensure at least a 1 log₁₀ decrease of the TCID₅₀/mL. Experiments were conducted during 10 to 365 days post-inoculation, depending on the treatment, with a final sampling performed the last day of the experiment. At the time of sampling, duplicate 0.5 mL samples of IA virus-inoculated water were diluted 1∶1 by addition of 0.5 mL of 2× MEM. Ten-fold dilutions (10⁻¹ to 10⁻⁶) were made in 1× MEM supplemented with antibiotics (10000 U Penicillin G, 10 mg Streptomycin, 25 µg Amphotericin B/mL).

Infectivity assays were performed on MDCK, following the method described by Brown et al. [3]. Results from duplicate titrations were log₁₀ transformed and averaged. Simple linear regressions were used to calculate the time required for a 90% reduction in infectivity (i.e. time required for a decrease of the viral titer by 1 log₁₀ TCID₅₀/mL); previous studies indicated that loss of IA virus infectivity in water over time decreased at a log-linear rate [3], [5]. Statistical analysis was performed in R 2.10.1 [41].

Experimental infections

One-day-old Mallards were purchased from a commercial source (Murray McMurray Hatchery, Webster City, Iowa, USA) and raised for one month under confined conditions until the beginning of the experiments. Three days before inoculation, ducks were separated into two groups of five and transferred to self-contained isolation units ventilated under negative pressure with high-efficiency particulate air-filters. Each duck was weighed before inoculation and samples were collected to test for IA virus shedding (oropharyngeal and cloacal swabs) or antibodies (serum). General care was provided in accordance to an animal use protocol (AUP # A2010 6-101) approved by the Institutional Animal Care and Use Committee at the University of Georgia, Athens, Georgia, USA. All work was conducted at an enhanced Biosafety Level 2 facility at the Poultry Diagnostic Research Center, Athens, Georgia, USA.

Five birds in each group were inoculated in the choanal cleft and the trachea with a volume of 0.2 mL containing an infectious titer of 10^5.63 EID₅₀ for A/Surface water/Minnesota/NW1-T/2006 and 10^6.54 EID₅₀ for A/Surface water/Minnesota/W07-2241/2007 (based on back titers). Oropharyngeal and cloacal swabs were collected from all birds on days 1, 2, 3, 4, 5, 7, 9, 11 and 14 post-inoculation and stored at −80°C until viral isolation and RNA extractions were performed. Blood samples were collected from the right jugular vein on days 4, 7, 10 and 14 post-inoculation, centrifuged for 30 minutes at 1500 rpm, and sera stored at −20°C until testing. Birds were evaluated twice daily and were weighed on days 4, 7, 10 and 14 post-inoculation to ensure that they gained weight during the course of the experiment.

Viral isolation was performed on cloacal and oropharyngeal swabs using 9 to 11 days old SPF embryonating chicken eggs following previously described procedures [42]. For each sample, RNA was extracted the same day as viral isolation was performed, with the MagMAX™-96 AI/ND Viral RNA Isolation kit (Ambion, Austin, Texas, USA) using the Thermo Electron KingFisher magnetic particle processor (Thermo Electron Corporation, Waltham, Massachusetts, USA) according to the modified protocol proposed by Das et al. [43]. Real-time RT-PCR targeting the Matrix gene was conducted with the QIAGEN OneStep RT-PCR kit (QIAGEN Valencia, California, USA) and the Cepheid SmartCycler System (Cepheid, Sunnyvale, California, USA) following the protocol described by Spackman et al. [44]. Samples with a cycle threshold value equal or less than 40 were considered positive.

Serum samples were tested with a commercial ELISA assay (IDEXX FlockCheck IA MultiS-Screen Antibody Test Kit, IDEXX Laboratories, Westbrook, Maine, USA) according to the manufacturer's instructions.