Pone.0016028 1.10

Detection of Resistance Mutations to AntiviralsOseltamivir and Zanamivir in Avian Influenza A VirusesIsolated from Wild Birds Goran Orozovic1,2*, Kanita Orozovic3, Johan Lennerstrand4, Bjo¨rn Olsen2 1 Section for Zoonotic Ecology and Epidemiology, Linneaus University, Kalmar, Sweden, 2 Section of Zoonotic Ecology and Epidemiology, Department of Medical Sciences, Uppsala University, Uppsala, Sweden, 3 Section for Bioorganic and Biophysical Chemistry Laboratory, Linneaus University, Kalmar, Sweden, 4 Section of Clinical Virology, Department of Medical Sciences, Uppsala University, Uppsala, Sweden The neuraminidase (NA) inhibitors oseltamivir and zanamivir are the first-line of defense against potentially fatal variants ofinfluenza A pandemic strains. However, if resistant virus strains start to arise easily or at a high frequency, a new anti-influenza strategy will be necessary. This study aimed to investigate if and to what extent NA inhibitor–resistant mutantsexist in the wild population of influenza A viruses that inhabit wild birds. NA sequences of all NA subtypes available from5490 avian, 379 swine and 122 environmental isolates were extracted from NCBI databases. In addition, a dataset containing230 virus isolates from mallard collected at Ottenby Bird Observatory (O ¨ land, Sweden) was analyzed. Isolated NA RNA fragments from Ottenby were transformed to cDNA by RT-PCR, which was followed by sequencing. The analysis ofgenotypic profiles for NAs from both data sets in regard to antiviral resistance mutations was performed usingbioinformatics tools. All 6221 sequences were scanned for oseltamivir- (I117V, E119V, D198N, I222V, H274Y, R292K, N294Sand I314V) and zanamivir-related mutations (V116A, R118K, E119G/A/D, Q136K, D151E, R152K, R224K, E276D, R292K andR371K). Of the sequences from the avian NCBI dataset, 132 (2.4%) carried at least one, or in two cases even two and three,NA inhibitor resistance mutations. Swine and environmental isolates from the same data set had 18 (4.75%) and one (0.82%)mutant, respectively, with at least one mutation. The Ottenby sequences carried at least one mutation in 15 cases (6.52%).
Therefore, resistant strains were more frequently found in Ottenby samples than in NCBI data sets. However, it is stilluncertain if these mutations are the result of natural variations in the viruses or if they are induced by the selective pressureof xenobiotics (e.g., oseltamivir, zanamivir).
Citation: Orozovic G, Orozovic K, Lennerstrand J, Olsen B (2011) Detection of Resistance Mutations to Antivirals Oseltamivir and Zanamivir in Avian Influenza AViruses Isolated from Wild Birds. PLoS ONE 6(1): e16028. doi:10.1371/journal.pone.0016028 Editor: Ralph Tripp, University of Georgia, United States of America Received October 30, 2010; Accepted December 9, 2010; Published January 6, 2011 Copyright: ß 2011 Orozovic et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was financially supported by the Swedish Research Council and the Swedish Research Council Formas. The funders had no role in studydesign, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
during influenza pandemics. Thus, many countries stockpile twocommercially available influenza A NA inhibitors, oseltamivir The scientific community has frequently expressed concern (TamifluH; active substance after oseltamivir processing in the liver about the potential of influenza A virus to evolve into novel strains is oseltamivir carboxylate (OC)) and zanamivir (RelenzaH), as the that can spread globally and induce pandemics [1–3]. These main defenses against pandemic strains [7–9].
warnings were proven justified 2009 when the world experienced For both OC and zanamivir, high-level drug resistance is the last influenza A pandemic induced by strain H1N1, also conferred by single or multiple nucleotide changes in the NA gene, known as swine influenza or new influenza. Fortunately, the new as influenza A displays a high mutation rate and high viral influenza was mild, as the viral infections in the majority of replication. Long-term seasonal use of amantadine, a previously infected humans did not end with serious complications [4,5]. All used antiviral of another class, has led to natural amantadine influenza A viruses originate from the avian influenza A viruses resistance in epidemic H3N2 and H1N1 viruses. Thus, human that naturally occur in waterfowl. The influenza genome encodes influenza A can develop resistance against both OC and zanamivir 11 proteins, of which one is non-structural. Avian influenza A is [10–18]. As both inhibitors bind to the catalytic site of NA, cross- classified according the presence of two membrane proteins, resistance mutations are also found. Perhaps the most alarming hemagglutinin (HA) and neuraminidase (NA). There are 16 HA news is the emergence of drug-resistant strains of the H5N1 and nine NA identified subtypes, and the majority of subtypes (96 subtype that cause high rates of mortality in humans [19–21]. The of 144) are found in the mallard duck (Anas platyrhynchos), assumed use of only OC and zanamivir as the first line of defense against to be the major host and source of the influenza A viruses [6].
pandemic strains has been disputed, and the need for new Vaccination is the most effective, cheapest and safest way to strategies or/and new antivirals has been proposed [22–24].
protect the majority of a population against influenza A, but Furthermore, the use of both antivirals increases during seasonal vaccines can be difficult to rapidly produce in sufficient quantities influenza, especially during a pandemic, which results in higher January 2011 | Volume 6 | Issue 1 | e16028 Influenza Antiviral Resistance in Wild Birds concentrations of these substances in the environment. It is isolates) or according to Orozovic [31] (120 isolates). In both cases, possible that wild birds (waterfowl) and, subsequently, influenza A the PCR products were electrophoresed in 1.5% agarose and viruses may come in contact with sewage water enriched with the visualized with ethidium bromide [31]. The bands of 1,400 bp antivirals. This may cause selection pressure on existing virus were cut out from the agarose gel, and the gel slices were purified populations, and as a consequence, resistant mutants may be using a QIAquick Gel Extraction Kit (Qiagen, Valencia, CA, USA). These 1,400-bp DNA fragments, representing NA genes, The aim of this study was to screen influenza A cDNA were sequenced by Macrogen (Seoul, Korea). The obtained sequences for resistance mutations against the two existing sequence readings were assembled and processed using either neuraminidase inhibitors to determine the prevalence these DNASTARH (DNASTAR, Inc., USA) or Vector NTI Advanced mutations in the wild bird population infected with influenza A.
version 10.3.0 (Invitrogen. Co., USA). The whole sequences were In this investigation two data sets were used. The first data set was aligned by BLAST, after which NA subtypes were identified [31].
collected between 2002 and 2008 at Ottenby Bird Observatoryand included 230 virus isolates from mallard. The second data set was obtained from the NCBI database and contained all bird, Published literature was researched for mutations connected to swine and environmental isolates of NAs of different subtypes.
both antivirals. NAs of all subtypes have eight conserved aminoacids involved in the contact with substrates as well as in the function of the active site, and these are defined as catalyticresidues: R118, D151, R152, R224, E276, R292, R371, and Y406. Ten additional amino acids, also well conserved, which are Ethical approval for trapping, sampling, and keeping of birds so-called framework residues (E119, R156, W178, S179, D198, was obtained from the Malmo¨/Lund Animal Research Ethics 1222, E227, E277, N294, and E425), are involved in stabilization Nine mutations against OC (V116A, I117V, E119V, D198N, I222V, H274Y, R292K, N294S and I314V) and 10 mutations Between 2005 and 2008 numerous cloacal samples from against zanamivir (V116A, R118K, E119G/A/D, Q136K, mallard ducks (Anas platyrhynchos) were collected using the cotton D151E/G/N, R152K, R224K, E276D, R292K and R371K) swab method. The sampling is a part of an ongoing surveillance at have been identified (N2 numbering; Table 1) [33–42]. Viruses Ottenby Bird Observatory on the Swedish island O with mutations R292K and V116A show resistance to both swabs were placed in 2-ml tubes with virus transport media inhibitors. All mutations have been detected in human influenza A [Hanks’ Balanced Salt Solution containing 0.5% lactalbumin, strains, with the exception of D198, which is found in an influenza 10% glycerol, 200 U/ml penicillin, 200 mg/ml streptomycin, B strain. As a reference, human N2 (accession number 100 U/ml polymyxin B sulfate, 250 mg/ml gentamicin, and CAD35677) from the NCBI database was used.
50 U/ml nystatin (ICN, Zoetermeer, Netherlands)] that were The alignments using ClustalW were performed in BioEdit immediately frozen at 270uC (at the latest, 30 min after 7.0.8.0 which was also used to scan all of the above-mentioned sampling). The 100 ml of virus transport media was used for mutations. The 230 mallard sequences from Ottenby as well as the RNA extraction, which was performed using an EZ1 Virus Mini 5490 avian, 379 swine and 122 environmental sequences obtained Kit (QIAGEN, Germantown, MD, USA) and extraction Biorobot from the NCBI database (Table 2) were included in the analyses.
EZ1 kit (QIAGEN), to yield a final volume of 75 ml of extracted Altogether, 6221 NA sequences were analyzed. The number of mutations for each subtype is expressed as a proportion of the total The presence of virus in the samples was confirmed using one- number analyzed sequences for that particular subtype (Table 2).
step q-PCR that targeted a conserved region of the avian influenza The proportions of mutations for avian isolates of both NCBI and A matrix gene. Extracted RNA (2 ml) was used as template in the Ottenby sequences were pooled separately, which resulted in six final reaction volume of 20 ml using a FastStart DNA Master replicates in the NCBI group and five replicates in the Ottenby SYBR Green I kit (Roche Diagnostics GmbH, Roche Applied group. To investigate if the proportion of mutants differed Science, Mannheim, Germany). The amplification procedure was between the two data sets, the unpaired t-test was performed performed in a LightCycler 1.5 (Roche Diagnostics GmbH) under the following conditions: activation of polymerase for 10 min at Also all sequences from mallard isolates (795 sequences) were 95uC and 43 cycles of 10 s at 95uC, 10 s at 60uC, and 10 s at extracted from the NCBI avian data set to form a new NCBI 72uC. Finally, melting curve analysis was performed via a stepwise mallard group. Another group was made of the same Ottenby temperature increase from 65uC to 95uC, which identified the virus isolates from mallard. The frequency of wild-type isolates (no melting temperature of the reaction product [30].
mutations) and mutants was organized as a contingency table andanalyzed by chi-square test (Figure 1B). The null hypothesis was that the proportion of mutants was the same in both sources of All positive samples from q-PCR were grown in 11-day-old pathogen-free chicken eggs (allantoic fluid). Each sample wasinjected into two eggs and left at 37uC for 2 days, upon which the allantoic fluid was removed by syringe. The presence of virus wasdetermined by hemagglutination assay using turkey erythrocytes.
HA subtyping was performed by hemagglutination inhibition The analyses of 5490 avian NCBI annotated sequences revealed assay with subtype-specific hyperimmune rabbit sera [30].
132 sequences carrying OC or zanamivir resistance mutations The NA gene was sequenced to subtype the viruses. Total RNA (Table 2 and Table S1). Eight OC-related mutations (I117V, was extracted from all hemagglutination-positive samples (High E119V, D198N, I222V, H274Y, R292K, N294S and I314V) and Pure RNA Isolation Kit; Roche Diagnostics GmbH, Germany) six zanamivir-related mutations (V116A, R118K, E119G/A/D, [31]. RT-PCR was done either according Hoffman [32] (110 R152K, R224K and R371K) were identified (Table 3). The January 2011 | Volume 6 | Issue 1 | e16028 Influenza Antiviral Resistance in Wild Birds Table 1. Overview on published oseltamivir and zanamivir related mutations.
Sensitivity in regard to NA subtype and acquisition 3) (Nt/R) N2 a, (R/R) NB a, (R/Nt) N1 a, (R/R) N2 c (Nt/S) N2 a, (R/R) NB a, (R/Nt) N1 a, (I/S) N2 c Mainly adopted from Ferraris and Lina [36].
1)Z - selected by zanamivir; O - selected by oseltamivir.
2)F - Framework residue; C - Catalytic residue.
3)Within bracket: R - resistant, I - intermediate, LR - low resistant, S - susceptible; nr - not recovered, Nt - not tested. Out of bracket: virus NA subtypes (N1 - 9 or B). Origin a)reverse genetic,b)In clinic,c)In vitro;*- mutations included in double mutant.
doi:10.1371/journal.pone.0016028.t001 majority of mutants were found in the N1 and N2 subtypes, which one N3 isolate (11.1% of subtype and 0.82 of total isolates; Table 2 showed 57 (2.7%) and 55 (3.2%) mutants, respectively. The N5 and Table S1) had R152K (zanamivir-related) (Table 4).
subtype had the largest proportion of mutants (6.25%), while theN4, N7 and N9 subtypes did not have any mutants (Table 2).
Subtype N1 showed one double mutant with I117V (OC-related) Mutants were found in 15 of 230 (6.52%) NA sequences from and E119G/A/D (zanamivir-related) mutations, and subtype N2 the Ottenby data set. Most of the mutants were found in the N6 had one triple mutant with the zanamivir-related R118K and subtype, which included six mutants (10.91%), followed by N1, R152K and OC-related D198N mutations (Table 3). The most with four (10.26%), and N9, with three mutants (27.27%). The N3 common mutation in N1 was the OC-related I117V, while the and N5 subtypes had only one mutant (4.35% and 8.33%, most common mutation for N2 was the OC-related I314V.
respectively) each, while the N2 and N8 subtypes did not show any H274Y (the most frequent resistance mutation in human influenza mutants. Subtypes N4 and N7 were not part of this collection of A) was seen in four isolates in N1 and one isolate in N2. Subtype sequences (Table 2). In the N1 subtype, I222V (OC-related) was N3 had three types of mutations, of which the OC-related I222V, found in one isolate. One isolate (68556) had only an R118K which was found in five isolates, was the most frequent. Only one mutation, and isolate 68557 was a double mutant with R118K and type of mutation was found in the N5 or N6 subtype. D198N was D151N mutations. The fourth isolate (79959) was triple mutant present in seven N5 isolates, and I222V was present in two N6 with the zanamivir-related R118K, D151N and OC-related isolates. Subtype N8 had two different mutations: one V116A and D198N mutations. This isolate had an additional R156K change in its sequence. However, this change is not related to resistance In the swine isolates, most of the subtypes did not have any NA [39]. The NCBI collection of avian sequences had 2 out of 5991 inhibitor resistance mutations (i.e., subtypes N1, N3, N6, N8 and (0.03%) mutants with more than a single mutation, while in the N9). The mutations were found only in the N2 subtype, which had Ottenby collection the same type of mutants was found in 4 out of 18 isolates (8.87% of the N2 subtype and 4.75% of all subtypes) with single mutations (Table 2 and Table S1). The majority of Mutants from both N3 and N5 subtypes had the R118K isolates had the OC-related I314V mutation, while the zanamivir- inhibitor resistance change. In all mutants belonging to the N6 related Q136K mutation was found only in swine sample set subtype, R152K and the subtype-conserved D198N mutation (Tables 1, 2, 3). In total, 122 isolates from the environment and were found. The OC-related mutation D198N has been observed January 2011 | Volume 6 | Issue 1 | e16028 Influenza Antiviral Resistance in Wild Birds Table 2. Summary of all virus isolates screened for antiviralsmutations.
Figure 1. Comparison of mutant proportions from the NCBIand Ottenby databases carrying zanamivir and OC resistance mutations. A) The mean percentages of six replicates from the NCBI and five replicates from the Ottenby data set. All percentages representsubtypes containing mutant virus isolates. Subtypes without mutants are not represented in this analysis. The NCBI mean 6 SEM percentage of mutants was 2.6660.865; N = 6. The Ottenby mean 6 SEM percentage of mutants was 12.2263.931; N = 5. p,0.05, unpaired t-test. B) Frequencies of wild-type isolates in the NCBI and Ottenby data sets were 716 (98.76%) and 215 (93.48%) isolates, respectively.
Frequencies of mutant isolates in the NCBI and Ottenby data setswere 9 (1.24%) and 15 (6.52%), respectively. In the NCBI data set, only viruses isolated from mallard were counted. Bars represent the percentage of wild-types and mutants in each data set. p,0.0001, chi-squared test.
doi:10.1371/journal.pone.0016028.g001 data sets (1.24% in NCBI and 6.52% in Ottenby) were different (p,0.0001), i.e., mutants were more frequent in the Ottenby data Resistance mutation patterns depend on the drug and the virus subtype. Additionally, some subtypes (e.g., influenza N2) are more sensitive to OC than to zanamivir, while the opposite is observed with other subtypes (e.g., N1) [14,43].
In this study 6221 NA sequences were scanned for published anti-OC and anti-zanamivir mutations (Table 1). When the sequences from the NCBI database were compared with Ottenby sequences, some differences emerged. The subtypes N4 and N7 were absent in the Ottenby sequence collection, but these subtypesdid not show mutations in the NCBI sequence collection. Subtypes only in influenza B virus. This study revealed that all NA N2 and N8 from the NCBI data set had isolates with mutations, sequences from subtypes N6, N7 and N9 had this change as a while mutants were absent in the Ottenby N2 and N8 sequences.
conserved feature. Two of the mutants from N9 carried one This could be explained by the small number of sequences for R118K mutation, and one isolate was a double mutant with these subtypes in the Ottenby set (Table 2). In contrast, the N9 R118K and D151N mutations. Furthermore, all isolates from this subtype collected at Ottenby had three mutants, while none was subtype had the conserved D198N change (Table 5).
seen in the NCBI date set. In the case of the N9 subtype, factorssuch as host species difference and the location where the isolation was carried out could be important. However, it is also possible The unpaired t-test showed that the mean proportions of that the N9 subtype was more sensitive to selective forces, such as mutations in the NCBI (2.66%) and Ottenby data sets (12.22%) immunity, natural NA inhibitors [44] or even different xenobiotics were different (p,0.05). Ottenby sequences had a higher distributed in the environment, including OC [25].
proportion of mutants than NCBI sequences. In the succeeding The largest proportion of mutants, 7 out of 112 (6.25%), in the analyses, frequencies of mutants within both data sets were NCBI database was in the N5 subtype, which could indicate that compared. Here, only sequences belonging to mallards from this subtype is the most prone to develop inhibitor resistance NCBI data set (13.20% of all avian sequences) were included. The mutations. However, six isolates included the same species and chi-square test showed that the frequencies of mutants in the two were sampled at the same place (Table S1). These isolates were January 2011 | Volume 6 | Issue 1 | e16028 Influenza Antiviral Resistance in Wild Birds Table 3. Overview of antivirals mutations virus isolates from Table 4. Overview of antivirals mutations in virus isolates 1)(z) - zanamivir related mutations; (o) - OC related mutation.
2)C - catalytic residue; F - framework residue.
collected for subtypes N1 and N2, respectively, in the NCBI group. Only 25% of NCBI N3 isolates were from that period, whereas none of the N5, N6 or N8 isolates was. However, in the subsequent analyses, the time of data collection was ignored as afactor that potentially influenced the quantity of mutants.
The lowest proportion of mutants from the Ottenby data set was 4.35% for the N3 subtype, which was still higher than the proportions of mutants from all other avian subtypes from the NCBI data set, except for N5 (Table 2). The N5 subtype from Ottenby had 8.33% mutants (1 out of 11), which was higher than the same subtype from the NCBI data set. The highest proportionof mutants from Ottenby isolates was found in the N9 subtype, with 27.27% mutations (3 out of 11). The same subtype from the NCBI database did not show any mutations. Another remarkable difference between the two data sets was observed in the N6 Table 5. Overview of antivirals mutations in virus isolates 1)(z) - zanamivir related mutations; (o) - OC related mutation.
2)C - catalytic residue; F - framework residue.
probably sampled at the same time form different individuals (same group of birds). Such relationships were observed in themany of the sequences that showed mutant genotypes, and information of this source is important when a detected mutation Comparison of the subtypes present in both data sets showed that the Ottenby samples had a higher proportion of mutants than the samples from the NCBI database (Figure 1A). Again, this could simply have been an effect of a non-random grouping of data,which was characteristic for both data sets. All Ottenby isolates were collected between 2002 and 2008. The period when 1)(z) - zanamivir related mutations; (o) - OC related mutation.
mutation-carrying NCBI isolates were collected varied depending 2)C - catalytic residue; F - framework residue.
on subtype. During 2002–2008, 93% and 84% of mutants were January 2011 | Volume 6 | Issue 1 | e16028 Influenza Antiviral Resistance in Wild Birds subtype, where the proportion of mutants was almost 20 times To summarize, a possible explanation for the lack of Q136K, higher in the Ottenby than the NCBI data set (Table 2).
D151E and E276D mutations in avian influenza A could be that These results indicate an increase of mutant frequency in the virus variants with such changes in the NA gene are not part of the population of avian influenza A viruses in mallard duck from natural variation. However, even if they were, it is probable that Ottenby. The reason for this can only be speculated. Given the selection pressure in the form of competition with other virus migratory routes of mallard duck that include many populated variants reduces the fitness of the virus so severely that these areas in northern and western Europe [45,46], it is appropriate to assume that mallards and consequently viruses could encounterwater-borne OC or zanamivir more frequently than viruses in Detected primary and secondary mutations in this study avian species collected from the NCBI data set. The possibility for It has long been thought that reductions in viral fitness recombination between avian and human virus strains already conferred by NA inhibitor resistance mutations would prevent carrying mutations in the NA gene cannot be ruled out [3].
transmission and the spread of resistance. However, recently NA In swine, only the N2 subtype sequences contained mutants inhibitor resistance has become apparent and has gradually spread (8.87%), while in viruses isolated from the environment, only one amongst circulating seasonal influenza viruses worldwide. This mutant in the N3 subtype was observed. Among swine mutants, could occur during prolonged treatment in, e.g., immunocom- seven belong to H3N2 virus subtype, five to H1N2 and six to promised patients. Therefore, ‘‘permissive’’ secondary mutations H9N2 virus subtype (Table S1). Two of the NA mutants (Q136K emerge that compensate for the reduced fitness of the primary NA and D198N) from H3N2 subtypes have sequence similarity with inhibitor resistance mutations [16,18,57].
human N2 which indicates that they probably originate from humans. The rest of NAs of H3N2 and all NAs from H1N1 virus subtype were categorized relative to two inhibitors (R292K subtype are most similar to swine N2 indicating origin from the excluded), 96% of all mutations were OC-related. Only one same organism. On the other hand each NA mutant from H9N2 double mutant, with I117V and E119A mutations, was found in subtype shows similarity with NA isolated from chicken i.e. duck the same subtype. This mutant probably did not show a reduction (A/Duck/Hong Kong/Y280/97) which implies avian origin.
in fitness, as it persisted in competition with other N1 versions that Adaptation of avian [47,48] or human [49] N2 to the new host lacked those changes. The OC-related mutation I117V was the (swine) that led to changes in the swine N2 subtype might haveprovided conditions favoring a more frequent occurrence of most frequent mutation, and it accounted for 63% of all mutants.
antiviral resistance genotypes than in the avian N2 subtype It has been detected in NA of an H5N1 virus strain (A/Chicken/ Indonesia/Wates/77/2005) that also had the I314V mutation, Regarding mutants isolated from environmental samples, only which made this strain a OC resistant double mutant I117V/ one with the zanamivir- related R152K mutation was found, I314V (Table 1) [34]. On the background of human strain H1N1 dating from 2004 (Table 4 and Table S1). Altogether, 122 A/WSN/33, mutation I117V alone is sensitive to OC but weakly environmental NA sequences are available in the NCBI database, resistant to the NA inhibitor A-315675 [58]. The capability of which may indicate difficulties in recovering virus genomic RNA I117V to reduce viral sensitivity to NA antivirals would probably from environmental samples or restricted efforts in doing so. The depend on the presence and identity of secondary mutations [34].
limited number of sequences from this analysis might explain why I117V was, in 23 of 36 cases (64%), isolated at the same time, at there were not more mutants within this group of sequences. The the same place and from the same species (open-billed stork), R152K mutation also occurred in the N2 subtype from the NCBI which indicated that the same group of birds has been infected data set, as well as in the N6 subtype from the Ottenby data set.
with the same virus subtype. The next most frequent mutations in NA RNA from the environmental sample originating from the N1 subtype were I222V, H274Y and N294S (Table 3). I222V Canada (Table S1) had the greatest similarity with chicken NA has been detected by reverse genetics (RG) in both N1 and N2 RNA that was also isolated in Canada. It is likely that the NA subtype, while H274Y and N294S have been found in human isolated from the environment originated from local poultry farms.
clinical isolates in the N1 and in the N1 and N2 subtypes,respectively (Table 1) [36]. The highly OC-resistant mutation Undetected resistance mutations in the study H274Y is the most frequent mutation in human isolates of H1N1 Scanning of all avian NA sequences from the NCBI database viruses [16,18] and has even been discovered in highly pathogenic showed that all known OC-related mutations were present in this avian H5N1 strains [20]. The ability of avian influenza A viruses data set [36,52]. On the contrary, the zanamivir-related mutations to carry OC-resistant mutations was revealed in this sequence Q136K [42], D151E and E276D [36,40] were not observed in the screening. This implies that these viruses possess enough fitness to cope with all the challenges imposed on them by the environments The measured concentration maxima (Cmax) of OC [53,54] (different hosts, different types of open waters) in which they exist.
and zanamivir [55] in the blood plasma can vary in the ranges of Of all mutations in the N2 subtype, the OC-related I314V 1.4–1.9 mM and 0.05–0.43 mM, respectively. On the other hand, mutation was dominant (73%) compared to the total proportion of concentrations of both inhibitors that induce mutations in vitro are all mutations (Table 3). This mutation has not been reported in the well above 1.0 mM [41,56]. The peak concentration of OC in two literature as a single mutation but only as a paired one with I117V studies from Japan was reportedly 0.001 mM [28,29], but in [34]. Therefore, it is not clear whether it could alone influence studies from the UK and U.S., it was predicted to be as high as changes in susceptibility to NA inhibitors. Still, its presence in 0.05 mM and 0.1 mM, respectively [27]. Thus, it is apparent that viruses of wild populations could be potentially harmful if such in cases of induced antiviral resistance, the concentrations of NA viruses obtain additional mutation(s). One triple mutant was found inhibitors in vitro are well above the values detected or predicted in within N2 sequences, carrying the R118K, R152K and D198N the environment. Even if OC or zanamivir had bioaccumulated in mutations. In this triple mutant, as in the case of the double N1 the waterfowl, it is likely that their concentrations would be much mutant, the fitness did not appear to be reduced dramatically, and lower than OC concentrations known to select for resistant virus it is possible that in both cases these changes in sequence could be compensatory regarding viral fitness [33].
January 2011 | Volume 6 | Issue 1 | e16028 Influenza Antiviral Resistance in Wild Birds Individual mutations R118K, E119V and S/R371K were number analyzed sequences for N1 subtype, while the corresponding specific for the N1 subtype, while V116A, R152K, R224K, and mutation rate for the NCBI data set was only 0.04% of total I314V were specific for the N2 subtype (Table 3). There were also sequences for the same subtype (Table 2). Two R118K mutants differences in the frequencies of some mutations. I222V, H274Y originated from successive years (2007–2008), which could indicate and N294S were found in a higher number in the N1 subtype, that this mutation was established in the wild virus population.
while E119G/A/D was more frequent in the N2 subtype. The It is tempting to speculate that the wild virus population of the majority of mutations in the N1 subtype were OC-related, while N1 subtype has gone through evolutionary changes driven by most mutations in the N2 subtype were zanamivir-related selective forces (antigenic drift, natural NA inhibitors or xenobi- (provided that I314V was not treated as a resistance-related otics) and that it is not driven by evolutionary fidelity. If one such mutation). In regards to the most frequent mutations for N1 virus population carries such a resistance mutation, it could (I117V) and N2 (I314V), it appears that they could be simply a potentially be harmful if it obtains other resistance-related genetic natural form of NA, i.e., they were neutral mutations. The same shifts. Thus, contact between a wild bird virus population and a was true for all N6, N7 and N9 subtype viruses that had D198N, human strain is the only step necessary in this scenario. Such an which is otherwise related to OC resistance in influenza B viruses event is possible either via direct transmission from birds to humans or via transmission involving a mixing vessel, such as pigs Of all sequences tested, only the avian N3 subtype had one [3]. Furthermore, the mutation R118K (Table 5) has been isolate with the zanamivir-related V116A mutation [35,58], and experimentally induced (RG) only in the N2 subtype [36], and its the rest of the mutations were I222V and S/I314V (Table 3). The instability seriously impacts viral fitness [40]. R118K was N5 and N6 subtypes each had only one type of mutation.
associated with N1, N3, N5 and N9 subtypes in the Ottenby data Moreover, the N6 and N8 subtypes had only two isolates with set. It could be that the spread of R118K was a result of a mutations. However, the low number of NA sequences for these recombination event, i.e., R118K was transmitted from one NA subtypes compared to N1 and N2 (Table 2) made it difficult to subtype to the other. This would be possible when more virus draw any firm conclusions on the frequency and type of mutations.
strains infect the same host simultaneously, in this case the Only five types of mutations were detected within swine NA mallards. Mallards are birds that gather in high numbers at sequences. The zanamivir-related Q136K mutation was only Ottenby. Still, not all of the birds are infected with the same virus found here (Table 4). I314V was the most frequent mutation, and strain at the same time, which significantly increases the chance if its frequency is compared to frequency of avian N2, where the that they could be infected with different virus strains.
same mutation was also prevalent (Table 3), then this NA variant In the Ottenby data set, besides the D198N found in the N1 might have originated from avian NA, where it was likely a part of triple mutant, I222V was another OC-related mutation. This a normal gene pool variation for the N2 subtype. Furthermore, if mutation has been obtained by RG [36,38] involving the N1 and I314V was ignored in both the avian and swine N2 subtype, then it N2 subtypes (Table 1), which did not show a resistant phenotype.
appears that the rest of the mutations constituted 0.9% and 2.0% However, in combination with mutation H274Y, its IC50 of the avian and swine N2 subtype populations, respectively.
increased almost 2000 times for one H5N1 strain [41]. The According to this, the swine N2 subtype was more receptive for mutation I222V was detected in one (2.56%) isolate in the N1 antiviral resistance mutations than the avian N2 subtype.
subtype from the Ottenby date as well as in five (0.23%) and one (0.06%) isolates in the N1 and N2 subtypes from the NCBI data, Ottenby (Table 5), only the N1 subtype had different mutation respectively. Thus, this ‘‘permissive and secondary’’ mutation was types, i.e., R118K, D151N, R156K, D198N and I222V. The rest found in a much higher proportion in the Ottenby set.
of the subtypes had either the zanamivir-related R118K or R152K Mutation D151N was present in two isolates from the N1 subtype mutations only. The zanamivir-related mutation R118K obtained as the second mutation besides R118K. This mutation was, until only by RG has not been possible to analyze by enzyme assay due recently, only associated with RG experiments involving the human to a complete lack of NA activity (Table 1) [40]. In the Ottenby N2 subtype, where it showed resistance against zanamivir (Table 1) data set R118K was present in all subtypes except in N6. In the N9 [36]. In a recent publication D151N was found alone or together subtype from Ottenby, R118K showed the highest frequency of all with H274Y in human N1 isolates [38]. It did not influence mutations in all subtypes from both data sets. Interestingly virus sensitivity to any inhibitors alone, but in combination with H274Y it H3N2 with the R118K mutation has been difficult to isolate from increased resistance to OC and to another NA inhibitor, peramivir.
in vitro culture [40], but in the Ottenby set such mutants existed as D151N can also result as adaptation to a new host, i.e., MDCK a part of the natural population in mallard. The N1 subtype had cells. The isolate 79959, with R118K and D151N mutations, also one double and one triple mutant, with R118K/D151N and has R156K and D198N mutations. R156K is not related to inhibitor resistance (and therefore is not shown in Table 3 or 4), but mutant had an additional mutation, R156K, but that mutation D198N has been detected in influenza B isolates as an OC- is not considered resistance-related. The presence of multiple resistance mutation [36]. D198N was also found in one avian isolate mutations could be considered secondary compensatory mutations of N2 and all avian isolates of N5 from the NCBI data set (Table 3).
that work in synergy together with the first resistance mutation, The occurrence of D198N could be the result of either a which usually reduces the fitness of the virus. In the N1 subtype, compensatory change to an already present primary resistance half of the mutants carried multiple changes in NA sequences. If mutation that reduced fitness or the result of a recombination with the relationship between the mutation frequencies and non- virus subtypes that had it as a conserved residue.
mutant frequencies were counted, then N1 had 18.00% of The N3 and N5 subtypes had the R118K mutation as the only mutations instead of 10.26%. Such a high mutation rate could mutation in the Ottenby set. The R118K mutation was not found be an indication of the existence of additional selection forces not in N3 from the NCBI data set. However, in the NCBI data set this mutation was detected in two avian isolates: one from the N1 and Comparisons of N1 mutations from both Ottenby and NCBI one from the N2 subtype (Tables 2 and 4). A similar trend was revealed some interesting details (Tables 1, 2 and 4). For example, observed even in the case of the N5 subtype. In the NCBI data set the proportion of R118K mutations in Ottenby was 7.7% of the total the N5 subtype exclusively had the OC-related D198N mutation, January 2011 | Volume 6 | Issue 1 | e16028 Influenza Antiviral Resistance in Wild Birds while the N5 subtype from Ottenby had only the zanamivir- Forces triggering such changes could be immune defense, natural NA inhibitors and adaptation to diverse transmission directions.
The N6 subtype had six mutants (10.91%) with the catalytic Additionally, these changes could be a consequence of regulation residue change R152K. No such mutation was detected in the of the balance between HA and NA activities subsequent to same subtype in the NCBI data set, which had two mutants changes in HA affinity towards its cell receptor. Such an (0.53%) with I222V (Tables 1 and 2). Interestingly the zanamivir- adaptation of NA as a response to changes in HA could lead to related mutation R152K, found in the bird NCBI data collection, was characteristic for the N2 subtype only, comprising 0.12% of The second question is related to the need to investigate whether the total number of sequences for the subtype.
those mutations actually reduce NA sensibility to inhibitors. They Subtype N9 from the NCBI data set did not have any mutation at could, in the light of differences in amino acids between avian and all. However, the same subtype from the Ottenby data set had the human virus strains, be neutral, i.e., they might not reduce highest proportion of mutants (3 out of 11; 27.27%) relative to all sensibility to NA inhibitors. It would be interesting to investigate subtypes of both data sets (Table 2). These three mutants had the the naturally produced mutants by NA enzymatic inhibition assays same R118K change, which has been observed by RG on the N2 to find out if they behave in the same way as their human counter- subtype [40]. However, a residue switch in one subtype does not parts. Such studies are currently in progress in our laboratory.
necessarily relate to NA inhibitor resistance in another subtype. Still, The third question concerns the source of transmissibility and this N9 catalytic residue change could indicate (as already antigenic shift of influenza A, which could be an important issue mentioned in the case of the N1 subtype) that certain forces select facing the next influenza outbreak. It is essential to investigate the for the accumulation of such resistance mutations in the NA gene.
transmission potential of those avian mutant strains to humans, as The Ottenby isolates carrying the catalytic residue change these strains might already be equipped with resistance against the R118K (N1, N3, N5 and N9) or R152K (N5 and N6) would be currently used inhibitors OC and zanamivir. Therefore, the interesting to study in NA enzymatic inhibition assays, and such strategy of stockpiling influenza NA antivirals as a first line of research is currently in progress in our laboratory. R118K is defense against new pandemic strains could be endangered. In especially interesting because RG viruses carrying this mutation do another study by us, preliminary results indicate that H1N1- not propagate well. There is also the possibility that these mutations infected mallards exposed to environmental concentrations of OC were obtained as an adaptation in chicken eggs. However, it should can develop the H274Y mutation (unpublished data).
be emphasized that almost all virus isolates in our study went The fourth and final question deals with the possibility of an through only one passage, so this event was less likely.
increased frequency of NA mutations in the wild populations ofviruses, as well as the emergence of novel and so far dormant subtypes (N9) in potentially harmful virus strains. Based on the NA subtypes that did not have any mutants were excluded from results from the Ottenby data set, this possibility is not unreasonable the total number of sequences. The test showed (Figure 1A) that (Figure 1A–B). It appears that certain selective forces have pushed the proportion of mutants was different between those two data the virus towards phenotypes that could be better equipped to infect sets (p,0.05). This analysis included all avian species from the a larger number of hosts. The threat of antiviral resistance in future NCBI data set, and it provided insight into how mean percentages influenza outbreaks warrants further exploration of alternative of virus mutants were related in a more general fashion within therapeutic strategies, e.g., new classes of drugs used in combination Within the analyzed NCBI sequences, the majority of isolates therapy with NA inhibitors [22,23] or alternative technologies for with mutations did not come from waterfowl (ducks, geese and faster production of influenza vaccine [62].
swans) but from chicken, stork, turkey, quail, tern and herring.
Ottenby samples were exclusively isolated from mallards.
Transmission of influenza A from its natural host, mallard [59], List with mutants from NCBI dataset includ- to another bird species, such as chicken [51], or to another animal ing protein ID, subtype, mutation, mutation sequence species in general [1,3,4,50] could have led to changes in the NA gene that were the result of an adaptation to a new host. Thus, in order to avoid influence of NA gene changes arisen due toadaptation to a new host, all sequences from each NA subtypefound only in mallards were extracted from NCBI data set and compared with Ottenby data set. The two data sets were put in acontingency table and analyzed by chi-squared test (Figure 1B), We would like to thank the staff at the Ottenby Bird Observatory, O which gave the same result as the unpaired t-test. The frequency of Sweden for the collection of the bird samples and providing us with mutations found in the Ottenby data set was higher than in the material for this study. We also wish to thank Conny Tolf, Neus Latorre- Margalef and Abbtesaim Jawad (Linnaeus University, Sweden) for theirassistance in laboratory analyses.
In summary, antiviral resistance–related mutations already exist in populations of avian influenza A viruses isolated from theirnatural hosts, i.e., mallard duck, other waterfowl as well as domestic poultry, and domestic swine. Therefore, several questions Conceived and designed the experiments: GO KO BO. Performed the experiments: GO KO. Analyzed the data: GO KO JL BO. Contributed The first question is whether virus strains carrying resistance reagents/materials/analysis tools: GO BO. Wrote the paper: GO KO JL mutations are natural fluctuations of different virus versions [60].
1. Cox NJ, Subbarao K (2000) Global epidemiology of influenza: Past and present.
2. Ferguson NM, Cummings DAT, Fraser C (2006) Strategies for mitigating an Annual Review of Medicine 51: 407–421.
influenza pandemic. Nature 442: 448–453.
January 2011 | Volume 6 | Issue 1 | e16028 Influenza Antiviral Resistance in Wild Birds 3. Taubenberger JK, Kash JC (2010) Influenza Virus Evolution, Host Adaptation, 31. Orozovic G, Latorre-Margalef N, Wahlgren J, Muradrasoli S, Olsen B (2010) and Pandemic Formation. Cell Host & Microbe 7: 440–451.
Degenerate primers for PCR amplification and sequencing of the avian 4. Deshpande JD, Phalke DB, Phalke VD (2009) ‘‘Swine Flu’’: The return of influenza A neuraminidase gene. Journal of Virological Methods 170: 95–99.
pandemic. Pravara Medical Review 4: 4–7.
32. Hoffmann E, Stech J, Guan Y, Webster RG, Perez DR (2001) Universal primer 5. Malik Peiris JS, Poon LLM, Guan Y (2009) Emergence of a novel swine-origin set for the full-length amplification of all influenza A viruses. Archives of influenza A virus (S-OIV) H1N1 virus in humans. Journal of Clinical Virology 33. Bloom JD, Gong LI, Baltimore D (2010) Permissive Secondary Mutations 6. Steinhauer DA, Skehel JJ (2002) Genetics of influenza viruses. Annual Review of Enable the Evolution of Influenza Oseltamivir Resistance. Science 328: 7. Monto AS (2006) Vaccines and Antiviral Drugs in Pandemic Preparedness.
34. Hurt AC, Selleck P, Komadina N, Shaw R, Brown L, et al. (2007) Susceptibility Emerging Infectious Diseases 12: 55–61.
of highly pathogenic A(H5N1) avian influenza viruses to the neuraminidase 8. Oshitani H (2006) Potential benefits and limitations of various strategies to inhibitors and adamantanes. Antiviral Research 73: 228–231.
mitigate the impact of an influenza pandemic. Journal of Infection and 35. Ilyushina NA, Seiler JP, Rehg JE, Webster RG, Govorkova EA (2010) Effect of Neuraminidase Inhibitor-Resistant Mutations on Pathogenicity of Clade 2.2 A/ 9. Germann TC, Kadau K, Longini IM, Macken CA (2006) Mitigation Strategies Turkey/15/06 (H5N1) Influenza Virus in Ferrets. Plos Pathogens 6.
for Pandemic Influenza in the United States. Proceedings of the National 36. Ferraris O, Lina B (2008) Mutations of neuraminidase implicated in Academy of Sciences of the United States of America 103: 5935–5940.
neuraminidase inhibitors resistance. Journal of Clinical Virology 41: 13–19.
10. Baigent SJ, Bethell RC, McCauley JW (1999) Genetic analysis reveals that both 37. Ho HT, Hurt AC, Mosse J, Barr I (2007) Neuraminidase inhibitor drug haemagglutinin and neuraminidase determine the sensitivity of naturally susceptibility differs between influenza N1 and N2 neuraminidase following occurring avian influenza viruses to zanamivir in vitro. Virology 263: 323–338.
mutagenesis of two conserved residues. Antiviral Research 76: 263–266.
11. Govorkova EA, Leneva IA, Goloubeva OG, Bush K, Webster RG (2001) 38. Okomo-Adhiambo M, Nguyen HT, Sleeman K, Sheu TG, Deyde VM, et al.
Comparison of efficacies of RWJ-270201, zanamivir, and oseltamivir against (2010) Host cell selection of influenza neuraminidase variants: Implications for H5N1, H9N2, and other avian influenza viruses. Antimicrobial Agents and drug resistance monitoring in A(H1N1) viruses. Antiviral Research 85: 381–388.
39. Richard M, Deleage C, Barthelemy M, Lin YP, Hay A, et al. (2008) Impact of 12. Gubareva LV (2004) Characterization of influenza A and B viruses recovered influenza A virus neuraminidase mutations on the stability, activity, and from immunocompromised patients treated with antivirals. In: Kawaoka Y, ed.
sensibility of the neuraminidase to neuraminidase inhibitors. Journal of Clinical Options for the Control of Influenza V. pp 126–129.
13. Gubareva LV (2004) Molecular mechanisms of influenza virus resistance to 40. Yen HL, Hoffmann E, Taylor G, Scholtissek C, Monto AS, et al. (2006) neuraminidase inhibitors. Virus Research 103: 199–203.
Importance of neuraminidase active-site residues to the neuraminidase inhibitor 14. Gubareva LV, Webster RG, Hayden FG (2001) Comparison of the activities of resistance of influenza viruses. Journal of Virology 80: 8787–8795.
zanamivir, oseltamivir, and RWJ-270201 against clinical isolates of influenza 41. Hurt AC, Holien JK, Barr IG (2009) In Vitro Generation of Neuraminidase virus and neuraminidase inhibitor-resistant variants. Antimicrobial Agents and Inhibitor Resistance in A(H5N1) Influenza Viruses. Antimicrobial Agents and 15. McKimm-Breschkin JL (2000) Resistance of influenza viruses to neuraminidase 42. Hurt AC, Holien JK, Parker M, Kelso A, Barr IG (2009) Zanamivir-Resistant inhibitors - a review. Antiviral Research 47: 1–17.
Influenza Viruses with a Novel Neuraminidase Mutation. Journal of Virology 16. Dharan NJ, Gubareva LV, Meyer JJ, Okomo-Adhiambo M, McClinton RC, et al.
(2009) Infections With Oseltamivir-Resistant Influenza A(H1N1) Virus in the 43. Ferraris O, Kessler N, Lina B (2005) Sensitivity of influenza viruses to zanamivir United States. Jama-Journal of the American Medical Association 301: and oseltamivir: A study performed on viruses circulating in France prior to the introduction of neuraminidase inhibitors in clinical practice. Antiviral Research 17. Harvala H, Gunson R, Simmonds P, Hardie A, Bennett S, et al. (2010) The emergence of oseltamivir-resistant pandemic influenza A(H1N1) 2009 virus 44. Matrosovich M, Klenk HD (2003) Natural and synthetic sialic acid-containing amongst hospitalised immunocompromised patients in Scotland, November- inhibitors of influenza virus receptor binding. Reviews in Medical Virology 13: December, 2009. Eurosurveillance 15: 2–4.
18. Weinstock DM, Zuccotti G (2009) The Evolution of Influenza Resistance and 45. Munster VJ, Baas C, Lexmond P, Waldenstrom J, Wallensten A, et al. (2007) Treatment. Jama-Journal of the American Medical Association 301: 1066–1069.
Spatial, temporal, and species variation in prevalence of influenza A viruses in 19. Boltz DA, Douangngeun B, Phommachanh P, Sinthasak S, Mondry R, et al.
wild migratory birds. Plos Pathogens 3: 630–638.
(2010) Emergence of H5N1 avian influenza viruses with reduced sensitivity to 46. Munster VJ, Wallensten A, Baas C, Rimmelzwaan GF, Schutten M, et al. (2005) neuraminidase inhibitors and novel reassortants in Lao People’s Democratic Mallards and highly pathogenic avian influenza ancestral viruses, northern Republic. Journal of General Virology 91: 949–959.
Europe. Emerging Infectious Diseases 11: 1545–1551.
20. Hill AW, Guralnick RP, Wilson MJC, Habib F, Janies D (2009) Evolution of 47. Cong YL, Pu J, Liu OF, Wang S, Zhang GZ, et al. (2007) Antigenic and genetic drug resistance in multiple distinct lineages of H5N1 avian influenza. Infection characterization of H9N2 swine influenza viruses in China. Journal of General Genetics and Evolution 9: 169–178.
21. Yen HL, Ilyushina NA, Salomon R, Hoffmann E, Webster RG, et al. (2007) 48. Yu H, Hua RH, Wei TC, Zhou YJ, Tian ZJ, et al. (2008) Isolation and genetic Neuraminidase inhibitor-resistant recombinant A/Vietnam/1203/04 (H5N1) characterization of avian origin H9N2 influenza viruses from pigs in China.
influenza viruses retain their replication efficiency and pathogenicity in vitro and Veterinary Microbiology 131: 82–92.
in vivo. Journal of Virology 81: 12418–12426.
49. Vincent AL, Ma WJ, Lager KM, Janke BH, Richt JA (2008) Swine Influenza 22. Govorkova EA, Webster RG (2010) Combination Chemotherapy for Influenza.
Viruses: A North American Perspective. Advances in Virus Research 72: 23. Ilyushina NA, Bovin NV, Webster RG, Govorkova EA (2006) Combination 50. Dugan VG, Chen R, Spiro DJ, Sengamalay N, Zaborsky J, et al. (2008) The chemotherapy, a potential strategy for reducing the emergence of drug-resistant evolutionary genetics and emergence of avian influenza viruses in wild birds.
influenza A variants. Antiviral Research 70: 121–131.
24. Poland GA, Jacobson RM, Ovsyannikova IG (2009) Influenza Virus Resistance 51. Li J, Dohna Hz, Anchell NL (2010) Adaptation and transmission of a duck- to Antiviral Agents: A Plea for Rational Use. Clinical Infectious Diseases 48: origin avian influenza virus in poultry species. Virus Research 147: 40–47.
52. Carr J, Ives J, Kelly L, Lambkin R, Oxford J, et al. (2002) Influenza virus 25. Fick J, Lindberg RH, Tysklind M, Haemig PD, Waldenstrom J, et al. (2007) carrying neuraminidase with reduced sensitivity to oseltamivir carboxylate has Antiviral Oseltamivir Is not Removed or Degraded in Normal Sewage Water altered properties in vitro and is compromised for infectivity and replicative Treatment: Implications for Development of Resistance by Influenza A Virus.
ability in vivo. Antiviral Research 54: 79–88.
53. Jhee SS, Yen M, Ereshefsky L, Leibowitz M, Schulte M, et al. (2008) Low 26. Sacca ML, Accinelli C, Fick J, Lindberg R, Olsen B (2009) Environmental fate penetration of oseltamivir and its carboxylate into cerebrospinal fluid in healthy of the antiviral drug Tamiflu in two aquatic ecosystems. Chemosphere 75: Japanese and Caucasian volunteers. Antimicrobial Agents and Chemotherapy 27. Singer AC, Nunn MA, Gould EA, Johnson AC (2007) Potential risks associated 54. Morrison D, Roy S, Rayner C, Amer A, Howard D, et al. (2007) A randomized, with the proposed widespread use of Tamiflu. Environmental Health crossover study to evaluate the pharmacokinetics of amantadine and oseltamivir administered alone and in combination. Plos Clinical Trials 2.
28. Soderstrom H, Jarhult JD, Olsen B, Lindberg RH, Tanaka H, et al. (2009) 55. Hata K, Koseki K, Yamaguchi K, Moriya S, Suzuki Y, et al. (2008) Limited Detection of the Antiviral Drug Oseltamivir in Aquatic Environments. Plos One inhibitory effects of oseltamivir and zanamivir on human sialidases. Antimicro- bial Agents and Chemotherapy 52: 3484–3491.
29. Ghosh GC, Nakada N, Yamashita N (2010) Oseltamivir Carboxylate, the Active 56. Gubareva LV, Robinson MJ, Bethell RC, Webster RG (1997) Catalytic and Metabolite of Oseltamivir Phosphate (Tamiflu), Detected in Sewage Discharge framework mutations in the neuraminidase active site of influenza viruses that and River Water in Japan. Environmental Health Perspectives 118: 103–108.
are resistant to 4-guanidino-Neu5Ac2en. Journal of Virology 71: 3385–3390.
30. Wallensten A, Munster VJ, Latorre-Margalef N (2007) Surveillance of Influenza 57. Hurt AC, Barr IG, Hartel G, Hampson AW (2004) Susceptibility of human A Virus in Migratory Waterfowl in Northern Europe. Emerging Infectious influenza viruses from Australasia and South East Asia to the neuraminidase inhibitors zanamivir and oseltamivir. Antiviral Research 62: 37–45.
January 2011 | Volume 6 | Issue 1 | e16028 Influenza Antiviral Resistance in Wild Birds 58. Abed Y, Nehme B, Baz M, Boivin G (2008) Activity of the neuraminidase 61. Aoki FY, Boivin G, Roberts N (2007) Influenza virus susceptibility and resistance inhibitor A-315675 against oseltamivir-resistant influenza neuraminidases of N1 to oseltamivir. Antiviral Therapy 12: 603–616.
and N2 subtypes. Antiviral Research 77: 163–166.
62. Musiychuk K, Stephenson N, Bi H, Farrance CE, Orozovic G, et al. (2007) A 59. Olsen B, Munster VJ, Wallensten A, Waldenstrom J, Osterhaus A, et al. (2006) launch vector for the production of vaccine antigens in plants. Influenza and Global patterns of influenza A virus in wild birds. Science 312: 384–388.
Other Respiratory Viruses 1: 19–25.
60. Lauring AS, Andino R (2010) Quasispecies Theory and the Behavior of RNA January 2011 | Volume 6 | Issue 1 | e16028

Source: http://www.aitoolkit.org/site/DefaultSite/filesystem/documents/Antiviral%20reeistance%20H5N1%20wild%20birds%20Orazovic%20PLoS%20ONE%206Jan2011.pdf