Figure 3

Nucleotide and amino acid sequence alignments and predicted RNA structures. (A1) Nucleotide alignment of ORFX and flanking regions for the sequences [GenBank:NC_009025] (IAPV), [GenBank:NC_004807] KBV, and [GenBank:NC_002548] (ABPV). Spaces separate +0/CDS2-frame codons. Colour coding is as follows: light blue - CDS2 IGR-IRES-mediated initiation site; red - ORFX termination codon; green - potential +1/ORFX-frame AUG initiation codons if ORFX is AUG-initiated (there are no intervening +0 or +2 frame AUG codons). Black arrows indicate the approximate expected initiation site if ORFX is IGR-IRES initiated (see text). Symbols '*' and 'x' represent completely conserved columns (based on a larger alignment comprising GenBank accession numbers NC_009025, EU436455, EU436456, EU436423, NC_004807, AY053375, AY053374, AY053372, AF486072, AY053367, AY053370, AY053366, AY053368, AF486073, AY053371 and NC_002548). (A2) The corresponding region in [GenBank:NC_006559] (SINV-1). (B) Amino acid alignment of the translated ORFX assuming initiation at the normal IGR-IRES initiation site but in the +1 reading frame. Methionine residues are highlighted in green. (C) Representative sequences showing a potential RNA hairpin structure directly upstream of the predicted IGR-IRES in the bee paralysis dicistroviruses. The CDS1 termination codons are underlined and in bold. The 5' end of the IGR-IRESs (as summarized in Ref. [11]) are underlined. Predicted base pairings are indicated by paired parentheses and coloured background shading. Substitutions that maintain the predicted base pairings are highlighted in blue (for single substitutions involving G:U pairings) or pink (for compensatory paired substitutions).