Ribosomal frameshifting used in influenza A virus expression occurs within the sequence UCC_UUU_CGU and is in the +1 direction

Abstract

Programmed ribosomal frameshifting is used in the expression of many virus genes and some cellular genes. In eukaryotic systems, the most well-characterized mechanism involves -1 tandem tRNA slippage on an X_XXY_YYZ motif. By contrast, the mechanisms involved in programmed +1 (or -2) slippage are more varied and often poorly characterized. Recently, a novel gene, PA-X, was discovered in influenza A virus and found to be expressed via a shift to the +1 reading frame. Here, we identify, by mass spectrometric analysis, both the site (UCC_UUU_CGU) and direction (+1) of the frameshifting that is involved in PA-X expression. Related sites are identified in other virus genes that have previously been proposed to be expressed via +1 frameshifting. As these viruses infect insects (chronic bee paralysis virus), plants (fijiviruses and amalgamaviruses) and vertebrates (influenza A virus), such motifs may form a new class of +1 frameshift-inducing sequences that are active in diverse eukaryotes.

Keywords: PA-X; genetic recoding; influenza virus; mass spectrometry; ribosomal frameshifting; translation.

Figure 1.

Mass spectrometric analysis of the PA-X-GFP frameshift fusion protein. (a) Translation map of influenza A virus segment 3 showing full-length PA and the transframe fusion PA-X that comprises the N-terminal domain of PA fused to a C-terminal tail encoded by the +1 reading frame. (b) Map of the construct used to purify the product of frameshifting on the PA-X frameshift cassette. (c) Complete amino acid sequence of PA-X-GFP. Amino acids encoded by the zero-frame are highlighted in mauve; amino acids encoded by the +1 frame are highlighted in pale yellow (X) or cyan (GFP). The eight peptides identified by mass spectrometry are indicated in red (note that the sequence GES…EGR corresponds to three detected peptides GES…LLK, HRF…EGR and FEI…EGR). The peptide spanning the frameshift site is underlined in green. (d) MS/MS fragmentation spectrum of the shift site peptide GLWDSFVSPR. The inset shows the peptide sequence with ‘b-’ and ‘y’-type fragment ions that strongly support the shift site peptide identified in the nano-LC/MS/MS analysis. Several additional fragment ions, corresponding to H₂O losses from b and y series ions and doubly charged fragment ions, are also present in the spectrum to further support the sequence (assignments not labelled in the figure). (e) Nucleotide sequence in the vicinity of the frameshift site UCC_UUU_CGU, with conceptual amino acid translations in all three reading frames. The product of +1 frameshifting is indicated in red. The green-underlined peptide, which spans the shift site, is compatible with +1, but not –2, frameshifting.

Figure 2.

Predicted sites of ‘PAX-like’ +1 frameshifting in (a) fijiviruses, (b) chronic bee paralysis and Lake Sinai viruses, and (c) amalgamaviruses. FDV, Fiji disease virus; MRCV, mal de Rio Cuarto virus; RBSDV, rice black-streaked dwarf virus; SRBSDV, southern rice black-streaked dwarf virus; CBPV, chronic bee paralysis virus; LSV, Lake Sinai virus; BBLV, blueberry latent virus; RhVA, rhododendron virus A; VCVM, Vicia cryptic virus M. In all cases, the predicted shift site occurs near the 5′ end of the overlap region between the zero-frame and +1 frame ORFs. Predicted shift sites are highlighted in blue. Dashes in CBPV indicate alignment gaps. Spaces separate zero-frame codons. Note that, downstream of the shift site, the sequences are predicted to be coding in both the zero and +1 frames, and this generally corresponds to enhanced conservation at the nucleotide level. The amalgamavirus sequences are highly divergent, and the precise alignment between BBLV and RhVA+VCVM is ambiguous in this region. GenBank accession numbers, and sequence coordinates of 5′ terminal nucleotides, are indicated at left.