3' cap-independent translation enhancers of plant viruses

Abstract

In the absence of a 5' cap, plant positive-strand RNA viruses have evolved a number of different elements in their 3' untranslated region (UTR) to attract initiation factors and/or ribosomes to their templates. These 3' cap-independent translational enhancers (3' CITEs) take different forms, such as I-shaped, Y-shaped, T-shaped, or pseudoknotted structures, or radiate multiple helices from a central hub. Common features of most 3' CITEs include the ability to bind a component of the translation initiation factor eIF4F complex and to engage in an RNA-RNA kissing-loop interaction with a hairpin loop located at the 5' end of the RNA. The two T-shaped structures can bind to ribosomes and ribosomal subunits, with one structure also able to engage in a simultaneous long-distance RNA-RNA interaction. Several of these 3' CITEs are interchangeable and there is evidence that natural recombination allows exchange of modular CITE units, which may overcome genetic resistance or extend the virus's host range.

Figure 1

Predicted structure of the STNV TED. Sequences involved in putative long-distance RNA-RNA interaction with 5′ sequences (boxed) are in green. Yellow boxes denote the conserved interacting sequences found in carmovirus TED-like, ISS, and PTE class CITEs. Sequences in blue are conserved between the STNV TED and the PLPV and PCRPV TED-like structures. Numbering is from the 5′ end of the viral genome. Abbreviations: CITE, cap-independent translational enhancer; CbMV, Calibrachoa mottle virus; ISS, I-shaped structure; PLPV, Pelargonium line pattern virus; PCRPV, Pelargonium chlorotic ring pattern virus; PTE, Panicum mosaic virus-like translational enhancer; STNV, Satellite tobacco necrosis virus; TED, translation enhancer domain.

Figure 2

Secondary structures of representative BTEs from the Luteovirus (BYDV), Necrovirus (TNV-D), and Dianthovirus (RCNMV1) genera. Stem (S) or stem loop (SL) numbers are indicated. The 17-nucleotide conserved sequence is in italics. Regions shaded in blue are protected from chemical modification by a functional truncation of eIF4G, indicating the likely eIF4G-binding site (24). Note that this includes most of the 17-nucleotide conserved sequence and other bases around the central hub. Green bases pair with the 5′ UTR for efficient translation. Loops in RCNMV1 BTE complement bases in the 5′ UTR, but this complementarity is not necessary for efficient translation (49). This figure is modified and reprinted with permission from Reference . Abbreviations: BTE, Barley yellow dwarf virus-like element; BYDV, Barley yellow dwarf virus; RCNMV1, Red clover necrotic mosaic virus RNA1; TNV-D, Tobacco necrosis virus-D; UTR, untranslated region.

Figure 3

Structures of three PTEs. (a) Representative PTE structures found in viruses in different genera (Umbravirus, PEMV2; Panicovirus, PMV; Carmovirus, SCV). Sequences involved in putative long-distance RNA-RNA interaction with 5′ sequences (boxed) are in green typeface. Yellow boxes denote the conserved interacting sequences found in carmovirus TED-like, ISS, and PTE class CITEs. The PEMV2 PTE has no interacting sequence but instead uses the interacting sequence of the adjacent kl-TSS 3′ CITE (see Figure 5b,c for the structure of this CITE). Potential pseudoknots are indicated by the double-headed arrow. Numbering is from the 5′ end of the viral genome. (b) Model of eIF4E docked to the PEMV PTE. This model is consistent with structure-probing and footprinting data. Reprinted from Reference with permission. Abbreviations: kl-TSS, kissing-loop T-shaped structure; PEMV2, Pea enation mosaic virus RNA 2; PMV, Panicum mosaic virus; PTE, Panicum mosaic virus-like translational enhancer; SCV, Saguaro cactus virus; TED, translation enhancer domain; ISS, I-shaped structure; CITE, cap-independent translational enhancer.

Figure 4

ISS and YSS secondary structures. Bases in green typeface pair with sequences in the 5′ UTR. (a) Alternative secondary structures of the MNeSV ISS. Structure i was predicted by Mfold and was supported, in part, by chemical probing and mutagenesis (34). Structure ii is the predicted structure of the MNeSV ISS selected in a chimeric replicating virus consisting of Carnation Italian ringspot virus containing an ISS in place of a YSS translational enhancer (35). All five known ISS, except for the MNSV-Mα5 ISS, can form both structures. MNSV-Mα5 can form a similar structure in an inverted orientation (right). Bases in blue typeface are conserved in all but the MNSV-Mα5 ISS, and mutation of these bases in the MNeSV ISS reduces its function (36, 37). Bases in gray typeface are those in the inverted MNSV-Mα5 ISS that defy the consensus. Bases in yellow boxes contain the conserved carmovirus interaction sequence. (b) Secondary structure of the TBSV YSS, supported by chemical probing and mutagenesis (8). Abbreviations: ISS, I-shaped structure; MNeSV, Maize necrotic streak virus; MNSV, Melon necrotic spot virus; TBSV, Tomato bushy stunt virus; UTR, untranslated region; YSS, Y-shaped structure.

Figure 5

Two-dimensional and three-dimensional structures of the TCV TSS and PEMV kl-TSS. (a) The TCV TSS, composed of three hairpins and two pseudoknots. No long-distance interaction is discernible for the TCV TSS or nearby sequences (55). (b, c) Two possible configurations of the PEMV kl-TSS. The loop of kl-TSS hairpin 3H1 engages in an RNA-RNA interaction with a 5′ coding sequence hairpin (see Figure 3a), which is compatible with simultaneous ribosome binding (F. Gao & A.E. Simon, unpublished data). (b) Hairpin 3H1 mimics a short anticodon stem. (c) Hairpin 3H2 occupies the anticodon stem position. The TCV TSS binds to 60S and 80S ribosomes (54), whereas the PEMV kl-TSS binds to 40S, 60S, and 80S ribosomes (10). The TCV TSS and PEMV kl-TSS do not compete with each other for binding to plant 80S ribosomes (F. Gao & A.E. Simon, unpublished data). Three-dimensional structures courtesy of B. Shapiro & W. Kasprzak (National Cancer Institute). Abbreviations: kl, kissing loop; PEMV, Pea enation mosaic virus; TCV, Turnip crinkle virus; TSS, T-shaped structure.

Figure 6

Phylogenetic tree of selected Tombusviridae and Tombusviridae-related viruses based on RdRp sequences. Virus acronyms are color-coded to match the type of 3′ CITE (two-dimensional structure at left) they contain. Lighter-shaded loops in the secondary structure diagrams indicate sequences known or predicted to base-pair to the 5′ end of the viral RNA. CITEs have not been identified for viruses in black typeface. RdRp sequences were aligned using the Muscle multiple sequence alignment algorithm and optimized using JalView version 2.6.1. Trees were generated from the alignment using MEGA5 via the neighbor-joining method and 1,000 replicates for the bootstrap. The authors thank Nikki Krueger for constructing the tree. Abbreviations: RdRp, RNA-dependent RNA polymerase; CITE, cap-independent translational enhancer; TED, translation enhancer domain; PTE, Panicum mosaic virus-like translational enhancer; BTE, Barley yellow dwarf virus-like element; S, stem; SL, stem loop.