| |
|
Introduction for the Public:
The current flu outbreak that is extensively covered in the media is caused by a new strain, the "2009 H1N1" (swine or Mexican) flu virus, that developed as a result of reassortment and multiple mutations from previously existing influenza A virus strains. The possible medical impact of these mutations can be partially evaluated with computational methods.
We were the first to publish the evolutionary analysis and a 3D structural model of a critical protein, the enzyme neuraminidase, of this flu virus in the journal Biology Direct. With this model, we and other scientists can quickly reveal if new mutations occur at strategic places such as the region where the drug interacts with the virus or the antigenic sites that are important for developing immunity.
While the virus has mutated extensively compared to related strains including the H5N1 avian flu and 1918 H1N1 Spanish flu, the mutations do not yet touch the binding site of the available drugs such as Tamiflu® or Relenza®, which means that they should still remain effective.
At the same time, mutations observed in patients have already changed possible sites at the enzyme's surface; thus, vaccines to previous strains and acquired immunity from preceding flu infections are expected to be less effective. According to the CDC and WHO, a new vaccine is expected soon.
Our molecular model also allows quickly assessing new mutations that will occur in the future and the developments will be watched closely.
News: Isolated oseltamivir (Tamiflu) resistant cases are reported in the news from Denmark, Osaka (Japan) and Hong Kong.
The Hong Kong and Osaka variants both have a H275Y mutation (H1 numbering) and are available from GenBank and are hence included
here. This mutation is known to cause Tamiflu but not Relenza resistance in previous influenza strains, including H5N1 bird flu.
|
|
|
Viewing residue(s) of interest:
If you want to map a new mutation to this structure or are interested in the location of other residues, right-click on the applet and select "Console". In the lower panel of the "Jmol Script Console", type:
"select $residue_num:$chain; spacefill 200; color pink".
Example: select 127:A; spacefill 200; color pink
In this example, residue 127 in chain A of the PDB file is selected, rendered with VdW radius of 200 Angstrom and colored pink. Typically, the neuraminidase is chain A and the drug chain B, except for the dimer where chains A and B are the neuraminidases and chain C the drug and other heteroatoms.
|
| |
| Download supplementary files: |
| |
| Alignments and phylogenetic trees |
| |
| Aln_all_NA_subtypes_nr90.fasta | Alignment of all NA subtypes, redundancy removed at 90% sequence identity |
| Tree_all_NA_subtypes_nr90.jpg | Phylogenetic tree of above alignment |
| Aln_N1_subtype_nr95+2009.fasta | Alignment of N1 subtype, redundancy removed at 95% sequence identity but keeping all new sequences from 2009 (as of April 29th) |
| Tree_N1_subtypes_nr95_JTTvarrate.jpg | Phylogenetic tree of above alignment using an alternative method of neighbor joining with JTT distance and gamma-distributed variable rates (gamma = 1.0) |
| |
| Structural models |
| |
| BII_H1N1_2009_zanamivir.pdb | Homology model minimized together with drug zanamavir |
| BII_H1N1_2009_oseltamivir.pdb | Homology model minimized together with drug oseltamivir |
| BII_H1N1_2009_peramivir.pdb | Homology model minimized together with drug peramivir |
| BII_H1N1_2009_dimer.pdb | Homology model of the homodimeric complex |
| |
| Epitope mapping |
| |
| H1N1_epitope_mapping.jpg | Epitopes from recent literature and IEDB mapped to alignment of NAs from the 2009 H1N1, 1918 H1N1 and 2004 H5N1 strains (reported antigen required to be at least 80% identical to the 2009 H1N1 NA sequence). |
| |
| Coevolution analysis |
| |
| NA_coevo_mafft.fasta.gz | Input alignment of 6182 non-identical neuraminidase sequences of influenza A viruses aligned with MAFFT |
| NA_coevo_pairSCA.scores.gz | Pairwise SCA scores (normalization over whole database) using the implementation of Ranganathan's SCA algorithm at http://coevolution.gersteinlab.org |
| NA_coevo_2sigma.jpg | Figure of coevolving residues. Residues that take part in pairwise correlations with SCA scores >2 times the standard deviation of all scores are colored cyan. All other residues are colored grey and the drug (zanamivir) is colored red. |
| |
| Animated Movie & Press Figure |
| |
| BII_swineflu_biologydirect.mpg (18 MB!) | MPEG movie of the 3D model of the 2009 H1N1 neuraminidase. The drug zanamivir is shown in green. Regions differing from the H5N1 avian flu and the 1918 H1N1 Spanish flu are shown in yellow. Mutations occurring among different patients within the first weeks of the 2009 outbreak appear red. So far, the drug binding site of the new virus appears unaffected and current treatment options, therefore, effective. But, surface regions important for vaccine recognition and immunity are already changing.
The movie was created with Yasara. |
| |
| BII_swineflu_NA.png | Static picture of the 3D model of the 2009 H1N1 neuraminidase and coloring as described above. |
| |
| BII_swineflu_NA_transparent.png | Same as above but with transparent background. |
| |
| These files are free to use if the sources (BII, A*STAR Singapore and the Biology Direct article) are properly cited. |
| |
