Two classes of antivirals against IFV are currently on the market, the adamantanes (amantadine and rimantadine), and the NAIs (oseltamivir and zanamivir). Antivirals play an important role in the treatment of severe seasonal and pandemic influenza, but their effectiveness is limited by the potential for IFV to develop drug resistance. Fortunately, the current pandemic IFV strain of swine origin (2009 A (H1N1)) is still largely sensitive to oseltamivir, although it is resistant to adamantanes 
. However, given the presence of rampant oseltamivir resistance among the current seasonal H1N1 IFV strains, it could be just a matter of time before the pandemic H1N1 virus acquires resistance to oseltamivir 
. Indeed, 28 cases of 2009 A (H1N1) resistance to oseltamivir have been reported as of October 2, 2009 (http://www.who.int/csr/don/2009_10_02/en/index.html
). Thus development of novel therapeutics with unique mechanism of action is of great importance to public health.
In our analysis, all eleven 2007 and 2009 seasonal IFV clinical isolates (11/11) exhibited resistance to oseltamivir when tested in MDCK cells. The high frequency of observed oseltamivir resistance and H274Y mutation in our 2009 H1N1 isolates correlates with worldwide reports of resistance in this season (http://www.cdc.gov/flu/weekly/weeklyarchives2008-2009/weekly15.htm
). Here we report that DAS181 is a potent inhibitor of the recent IFV clinical isolates. There is no reduction in DAS181 sensitivity amongst all the tested IFV clinical isolates from 2004, 2007 and 2009, demonstrating that the H274Y mutation does not reduce DAS181 sensitivity.
It has been well established that IFV develops resistance to NAIs through two mechanisms: one, mutations in NA that decrease binding affinity to NAI drugs, such as the H274Y mutation in the current seasonal IFV; two, mutations in HA which decrease virus receptor binding affinity, thereby reducing dependence on NA 
. While NA mutations tend to make IFV resistant to one NAI drug, HA mutations have the potential to confer resistance to all NAI drugs. However, it is essential to recognize that NAI resistance caused by HA mutations is host-specific. Changes in HA that reduce the virus binding to receptors in one host (e.g., MDCK cells) may not be the same that produce such effect in another host (e.g., humans). Therefore, viruses that show “pan-NAI resistance” in MDCK cells could be fully susceptible to NAI in humans and vice versa 
. HA mutations may play a role in reduced susceptibility to NAIs in drug patients, although direct experimental evidence is lacking at this time. In addition to resistance issues, HA mutations can change the tropism of viral binding (upper vs. lower respiratory tract binding) and the antigenicity of the virus.
When HA sequence analysis was performed on viruses recovered from oseltamivir treated children, an HA S262N mutation was found in 1 out of 50 cases, but NA mutations were detected in 18% of the cases 
. Additional HA mutations, with or without accompanying NA mutations, have also been reported in other studies 
. In spite of numerous HA mutations reported in clinical samples, there is a lack of documentation on HA-mediated resistance to NAI drugs in seasonal IFV due to the lack of an acceptable phenotypic assay that can be used for surveillance purpose and resistance diagnosis 
. The commonly used MDCK cell PRA is considered unreliable for detecting NAI resistance in humans because it is prone to give false positive and false negative results 
. The well differentiated human airway epithelium culture (HAE) is an experimental model system that better mimics the human airway, but it is cumbersome to grow and IFV clinical isolates tend to grow poorly in HAE cultures. Thus, the best option to identify HA mutations that may potentially give rise to NAI resistance may be to perform sequence analysis of the HA gene to specifically look for mutations at or near sialic acid binding site 
In the 2007 and 2009 clinical isolates that appear to have decreased sensitivity to zanamivir based on PRA, we have identified mutations involving two amino acid residues in the HA gene that are proximate to the sialic acid binding site, N163 and D225. Mutation of N163 may be highly relevant for NAI-resistance because glycosylation of asparagine (N) near the sialic binding site directly impacts the HA receptor binding affinity 
. Mishin et al. reported that introduction of an N163G mutation made a recombinant laboratory IFV strain more sensitive to NAIs in MDCK cells 
. However in our analysis, the N163H/K/T mutation was primarily observed in isolates with reduced zanamivir sensitivity, which seems to indicate an opposite effect of mutation involving N163. This contradiction may be due to differences in the HA backbone sequence between the isolates tested here and the virus in Mishin et al. It also reflects the complexity of interpreting HA mutations. Nevertheless, these results together suggest significance of N163 residue in NAI sensitivity in MDCK cells.
Examination of the HA sequence of pandemic
2009 IFV finds that nearly all isolates have a lysine (K) at amino acid 163, yet exhibit normal zanamivir sensitivity. However, the framework sequence of the HA in the pandemic 2009 IFV is radically different from that of seasonal H1N1 IFV, and thus inferences about a single point mutation amidst the sea of other changes is tenuous at best. For seasonal
influenza H1N1 an alignment of 2009 sequences prior to the swine flu outbreak (01/2009 to 03/2009) reveals that 85/89 have an N and 2/89 have K/H/T at amino acid 163, indicating that the N163K/H/T mutation is rare, but does exist at some low level in seasonal influenza (Influenza Virus Resources, 
). Further, most testing of NAI sensitivity currently occurs with the NI assay, which would miss NAI-resistance due to HA mutations, and limited prescription of zanamivir might reduce the chance of observing clinical zanamivir resistance. Hence, N163K/H/T-mediated reduction in NAI sensitivity in humans is unlikely to have been observed as yet.
An NA mutation (I222V) was observed in two of the zanamivir-resistant 2009 seasonal IFV strains, in addition to the well described H274Y mutation. Interestingly this mutation has recently been observed, in tandem with the H274Y mutation, in 2 patients with pandemic 2009 IFV. Drug sensitivity analysis was not performed on these viral isolates, therefore the significance of the I222V mutation in patients is unclear 
. Previous in vitro
selection studies have indicated that the I222V mutation exacerbates oseltamivir- and peramivir-resistance caused by H274Y, but has only modest affect on zanamivir sensitivity 
Broad-spectrum resistance to NAIs caused by a combination of HA and NA mutations has been previously reported when laboratory IFV strains were subjected to in vitro
passages in the presence of oseltamivir 
or peramivir 
. In the first case, a combination of A28T(HA) and R292K(NA) mutations resulted in 3230-fold and 60-fold resistance by A/Victoria/3/75 (H3N2) to oseltamivir and zanamivir, respectively 
. In the second case, B/Yamagata/16/88 acquired 100 to 700-fold resistance to oseltamivir, zanamivir, and peramivir due to a H274Y mutation in NA in combination with six HA mutations (G141E, D195N, T197N, T139N, R162M, and Y319H) 
. Similarly in our study, the 2007 and 2009 isolates that are resistant to oseltamivir and have reduced sensitivity to zanamivir carry mutations in both NA (H274Y) and HA (N163K/T/H and D225G), but the observed HA mutations have not been reported previously in drug-resistant IFV strains. The N163 and D225 amino acids are highly conserved amongst published H1N1 IFV strains (Figure S6
), further indicating their potential importance in viral infection. Further elucidation in in vivo
models is required to confirm the clinical significance of these mutations. Finally, an IFV A/Victoria/3/75 passaged with peramivir and found to be strongly pan-NAI resistant (oseltamivir, zanamivir, peramivir) in vitro 
, correlating with a single HA mutation (K186E), has been subsequently shown to be highly sensitive to DAS181 in MDCK cells and mice (unpublished data).
Distinct from the adamantanes and NAIs, DAS181 targets the host cells, not the virus. It functions by removing sialic acid receptors for IFV and thus rendering the host cells unable to be infected by IFV. Given the unique mechanism of action of DAS181, it is not unexpected that IFV strains that are resistant to antivirals remain sensitive to DAS181. Our results suggest that DAS181 could be an effective treatment for NAI-resistant IFV, whether the drug resistance is due to mutations in NA or HA. We have previously demonstrated that DAS181 potently inhibits the current pandemic H1N1 virus of swine origin in vitro
and in vivo 
. Thus, DAS181 may offer a potential therapeutic option for pandemic 2009 A(H1N1) virus resistant to NAI drugs.