Using influenza vRNPs purified under acidic conditions, and thus mimicking physiologically-relevant influenza infections as closely as possible, we have found that inhibition of either NLS1 or NLS2 on influenza NP significantly inhibited the nuclear import of influenza vRNP complexes. Therefore, both NLS1 and NLS2 on NP are involved in mediating the nuclear import of incoming vRNP complexes. These two sequences act independently of each other, as peptide competition with or antibody inhibition of only one of these sequences still resulted in a certain, though less pronounced, degree of nuclear import of vRNPs. Furthermore, when both NLS1 and NLS2 were competed with peptides or blocked by antibodies, the nuclear import of the vRNPs was even more drastically reduced.
Some differences, however, existed in the ability of the NLS peptides versus the anti-NLS antibodies to inhibit nuclear import of the vRNPs. These differences were likely due to the nature of the competition experiments, since peptide competition with the vRNPs for the cytosolic nuclear import receptors is less specific than direct inhibition of the vRNP NLSs with antibodies. The antibody inhibition experiments may hence provide a more accurate picture of the relative contributions of NLS1 and NLS2 to influenza vRNP nuclear import. From these results, it appears that peptides mimicking or antibodies against these conserved NLS regions on NP may be an effective means of interrupting a critical stage in the influenza A life cycle.
Interestingly, performing a sequence alignment (using Clustal W ) of NP from different influenza A strains, we can observe that each of NLS1 and NLS2 on NP are highly conserved among influenza A strains. However, NLS1 and NLS2 do not share much similarity with any region on NP of influenza B or C, as analyzed from a Clustal W alignment. This is in agreement with previous studies that have not been able to pinpoint the NLS in NP for influenza B , so it appears that influenza B and C may utilize NLSs that are different from those of influenza A. With respect to NLS1, the only residue conserved among influenza A, B, and C is a conserved arginine at position 8 of the influenza A NP, which agrees with previous studies indicating that that residue is one of the most important residues involved in mediating the nuclear import of recombinant NP [29, 30]. For NLS2, residue 214 on influenza A NP is probably one of the most important residues as an arginine or lysine is found in that position in influenza A, B, and C.
The exact length of NLS2 also remains to be determined. Previous studies had implicated that NLS2 was a bipartite NLS of 19 amino acids long, spanning residues 198 to 216 . However, recent structural data has questioned that this NLS functions as a bipartite classical NLS since the crystal structure of NP showed that the two clusters of basic residues of the bipartite NLS2 on NP were located too close together in space to be functional as a bipartite NLS . Even though NLS2 may not be a bipartite NLS, the relevance of residues 213, 214, and 216 on NP in mediating the nuclear import of recombinant NP appears to be significant . However, which other residues in NLS2 are important in mediating influenza nuclear import remains to be determined.
Our studies here with influenza vRNPs also confirm findings with recombinant NP that NLS1 is the stronger of the two NLSs [29, 31, 37]. From the crystal structure of NP , it is probably reasonable to assume that NLS1, being an N-terminal sequence near the edge of NP, may be more accessible to the binding of cytosolic nuclear import factors than NLS2. However, having both NLS1 and NLS2 as functional NLSs on influenza vRNPs could serve a vital purpose to the virus. If, in the event that the N-terminal NLS1 is inadvertently cleaved off by any proteases in its host cell, the vRNP may still have an extra NLS (NLS2) to mediate its nuclear import. Nonetheless, it appears that with so many copies of NP, it does seem to be a redundant function. However, it is not clear how many of these NLSs are exposed when NP oligomerizes and associates with the vRNA. Therefore, further studies at understanding the kinetics, cellular targets, conformational states, and role in viral replication of NLS1 and NLS2 will be required to bring further light to their roles in influenza cellular trafficking and replication.
Previous work by other groups have concentrated mainly on the nuclear import of recombinant NP [5, 27–31, 37]. These studies have provided a better understanding of the role of the various NLSs on NP in the nuclear import of newly-synthesized NP, which occurs after the initial nuclear import of the entire vRNP complex and the subsequent synthesis of new NP in the cytoplasm [12, 14]. To study the nuclear import of influenza vRNPs, O'Neil et al. (1995) , and more recently Cros et al. (2005) , formed in vitro-assembled NP-RNA complexes by incubating recombinant NP with in vitro-synthesized influenza vRNA. To study the nuclear import of the influenza genome, however, vRNPs purified from influenza virions would likely be the preferred substrates over in vitro-formed vRNA-NP complexes. This is because the actual assembly of NP into an oligomeric structure, and the interactions of NP with the vRNA in actual influenza infections, would likely result in structural differences between in vitro-formed RNA-NP complexes versus virally-produced and assembled vRNPs. For example, certain NLSs may be exposed or hidden according to how NP actually interacts with itself in the oligomer and how NP interacts with the vRNA. In addition, NP molecules within actual influenza vRNPs that are produced in mammalian cells may have differences in their post-translational modifications compared to recombinant NP molecules produced in bacteria. Furthermore, any conformational changes in the structure of the vRNPs after influenza export from the cell, viral entry into a new cell, and during or after their exit from endosomes are not taken into account by in vitro-formed RNA-NP complexes. The methodological differences in the preparation of vRNPs may therefore explain the differences observed in studies using in vitro-formed RNP and our results reported here using naturally-occurring, influenza-derived vRNP complexes. For example, Cros et al.  found that disruption of NLS2 on NP has no effect on the nuclear accumulation of in vitro-formed vRNA-NP complexes, while we showed here that interfering with NLS2 on NP diminished the nuclear import capability of influenza-purified vRNPs (Figs. 4 and 5). Likewise, Ozawa et al.  found that NLS2, but not NLS1, deletion mutants of NP were unable to target effectively to nucleolar regions, while we showed here that nucleolar localization still occurred whether NLS1 or NLS2 on NP was inhibited.
Much work on understanding the nuclear import of the influenza genome still remains. For example, the nuclear accumulation sequence (NAS) within influenza-assembled vRNP complexes is still a mystery. The NAS on NP was originally found to mediate nuclear accumulation of NP in Xenopus oocytes , but more recently has been found to be a cytoplasmic retention signal in mammalian cells [30, 31, 46]. It would therefore be useful to understand in greater detail the role of the NAS and how its function relates to that of NLS1 and NLS2. In addition, the role of M1 in preventing nuclear import of vRNP is still unclear [11, 39]. For example, M1 may be acting indirectly, where interaction of M1 with the vRNPs causes the vRNPs to change their structural conformation  and thus expose or hide certain NLSs. Alternatively, M1 may be binding directly to the NLSs of vRNPs to inhibit their nuclear import. The studies completed to date also raise the question as to under what conditions and in what conformational states of NP would NLS1 and NLS2 act to mediate the nuclear import of vRNPs. Further unraveling the answers to these questions may give us a more detailed understanding of how the various NLSs on the influenza vRNPs work together or independently to mediate influenza A nuclear import.