Logo: to the web site of Uppsala University

Oseltamivir (Tamiflu) is the most important antiviral drug available and a cornerstone in the defence against a future influenza pandemic. Recent publications have shown that the active metabolite, oseltamivir carboxylate (OC), is not degraded in sewage treatment plants and is also persistent in aquatic environments. This implies that OC will be present in aquatic environments in areas where oseltamivir is prescribed to patients for therapeutic use. The country where oseltamivir is used most is Japan, where it is used to treat seasonal flu. We measured the levels of OC in water samples from the Yodo River system in the Kyoto and Osaka prefectures, Japan, taken before and during the flu-season 2007/8. No OC was detected before the flu-season but 2-58 ng L(-1) was detected in the samples taken during the flu season. This study shows, for the first time, that low levels of oseltamivir can be found in the aquatic environment. Therefore the natural reservoir of influenza virus, dabbling ducks, is exposed to oseltamivir, which could promote the evolution of viral resistance.

Oseltamivir (Tamiflu (R)) is the most widely used drug against influenza infections and is extensively stockpiled worldwide as part of pandemic preparedness plans. However, resistance is a growing problem and in 2008-2009, seasonal human influenza A/H1N1 virus strains in most parts of the world carried the mutation H274Y in the neuraminidase gene which causes resistance to the drug. The active metabolite of oseltamivir, oseltamivir carboxylate (OC), is poorly degraded in sewage treatment plants and surface water and has been detected in aquatic environments where the natural influenza reservoir, dabbling ducks, can be exposed to the substance. To assess if resistance can develop under these circumstances, we infected mallards with influenza A/H1N1 virus and exposed the birds to 80 ng/L, 1 mu g/L and 80 mu g/L of OC through their sole water source. By sequencing the neuraminidase gene from fecal samples, we found that H274Y occurred at 1 mu g/L of OC and rapidly dominated the viral population at 80 mu g/L. IC(50) for OC was increased from 2-4 nM in wild-type viruses to 400-700 nM in H274Y mutants as measured by a neuraminidase inhibition assay. This is consistent with the decrease in sensitivity to OC that has been noted among human clinical isolates carrying H274Y. Environmental OC levels have been measured to 58-293 ng/L during seasonal outbreaks and are expected to reach mu g/L-levels during pandemics. Thus, resistance could be induced in influenza viruses circulating among wild ducks. As influenza viruses can cross species barriers, oseltamivir resistance could spread to human-adapted strains with pandemic potential disabling oseltamivir, a cornerstone in pandemic preparedness planning. We propose surveillance in wild birds as a measure to understand the resistance situation in nature and to monitor it over time. Strategies to lower environmental levels of OC include improved sewage treatment and, more importantly, a prudent use of antivirals.

Low-pathogenic avian influenza (LPAI) viruses in wild birds are important as they can constitute the basis for the development of high-pathogenic avian influenza (HPAI) viruses or form part of human-adapted strains with pandemic potential. However, the LPAI infection as such is not very well characterized in the natural reservoir, dabbling ducks, and results are in part contradictory. The effects on the infection by artificial versus natural infection, exposure to antiviral drugs or development of resistance have not been studied. Therefore, we used q-PCR, histopathology and immunohistochemistry (IHC) to study mallards infected with an influenza A/H1N1 virus isolated from a wild mallard in Sweden. The mallards were either inoculated intra-esophageally or infected by virus shed by other ducks in the experiment. The birds were subjected to low levels of the active metabolite of oseltamivir (Tamiflu^®) and the resistance mutation H274Y developed during the course of the experiment.

All mallards but one had a strictly intestinal localization of the LPAI infection. The exception was a bird euthanized one day post artificial inoculation whose infection was located solely in the lung, possibly due to intra-tracheal deposition of virus. The intestinal infection was characterized by degenerating cells in the lamina propria, infiltrating heterophils and lymphocytes as well as positivity of IHC and q-PCR on samples from feces and intestinal contents. Histopathological changes, IHC positivity and viral shedding all indicate that the infection peaked early, around two days post infection. Furthermore, the infection had a longitudinal progression in the intestine with more activity in the proximal parts early in the infection and vice versa as observed both by IHC and by q-PCR. There was no obvious difference in the course of the infection in artificial versus natural infection, when the level of OC was increased from 80 ng/L to 80 µg/L or when the resistance mutation H274Y developed.

Resistance to neuraminidase inhibitors (NAIs) is a growing problem in the battle against influenza A viruses. However, little is known about the resistance of viruses isolated from dabbling ducks, the natural reservoir of the influenza virus. To date, no low-pathogenic avian influenza (LPAI) virus functionally resistant to NAIs has been detected. The aim of this study was to investigate mallard isolates of influenza A virus previously identified to carry oseltamivir carboxylate (OC)- or zanamivir (ZA)-related resistance mutations. In this work, 21 viruses belonging to the N1, N3, N6 and N9 subtypes were analyzed using a colorimetric NA inhibition assay. The R118K mutation was the most frequently observed; it was seen in all subtypes except for N6. IC₅₀ values confirmed the differences in sensitivity to OC or ZA previously observed in the N1 and N2 groups of NAs. Furthermore, both negative controls (NCs) in the N6 and one NC in the N9 subtype were less sensitive to ZA than were genotypically related mutants of the respective subtypes. The presence of OC- and ZA-related mutations in the NA of viruses isolated from wild birds did not result in a NAI-resistant phenotype in this study. The R118K and R152K mutations seemed to somewhat increase the sensitivity to both NAIs compared to WT viruses which supports the thought that the mutations were a result of natural NA variation and not induced by NAI residuals in the environment. Although not resulting in a NAI resistant phenotype, the finding of the mutations D151N and I222V shows that mutations with the potential to enhance NAI resistance exist in the pool of LPAI viruses which stresses the need for surveillance of NAI resistance in wild birds.