Hepatitis C virus internal ribosome entry site (IRES) stem loop IIId contains a phylogenetically conserved GGG triplet essential for translation and IRES folding

Abstract

The hepatitis C virus (HCV) internal ribosome entry site (IRES) is a highly structured RNA element that directs cap-independent translation of the viral polyprotein. Morpholino antisense oligonucleotides directed towards stem loop IIId drastically reduced HCV IRES activity. Mutagenesis studies of this region showed that the GGG triplet (nucleotides 266 through 268) of the hexanucleotide apical loop of stem loop IIId is essential for IRES activity both in vitro and in vivo. Sequence comparison showed that apical loop nucleotides (UUGGGU) were absolutely conserved across HCV genotypes and the GGG triplet was strongly conserved among related Flavivirus and Pestivirus nontranslated regions. Chimeric IRES elements with IIId derived from GB virus B (GBV-B) in the context of the HCV IRES possess translational activity. Mutations within the IIId stem loop that abolish IRES activity also affect the RNA structure in RNase T(1)-probing studies, demonstrating the importance of correct RNA folding to IRES function.

FIG. 1

(A) Schematic representation of plasmid used for analysis containing the HCV IRES. T7 is the location of the T7 RNA polymerase promoter sequences for transcription of bicistronic RNA. Reporter RLUC is under translational control of the Xenopus β-globin 5′NTR while ΔCore (hashed box)/FLUC is under translational control of the HCV IRES as described in Materials and Methods. (B) Morpholino antisense oligonucleotides used for antisense inhibition studies. Antisense nucleotide sequences are listed in underlined text within each set, while mismatch controls are listed below. Dots represent identical sequences, while lowercase letters denote base pair substitutions. m(4) denotes a 4-bp mismatch, and RDM indicates random sequences. (C) Secondary model of the HCV IRES (15), including all sequences (nucleotides 1 through 408) used to construct the assay plasmid shown in panel A. Locations of the antisense targets are shown with oval boxes. IIId loop nucleotides 253 through 279 and numbers corresponding to antisense target sequences are shown in bold type.

FIG. 2

Morpholino antisense inhibition of the HCV IRES. (A) Equal concentrations (200 nM) of various antisense oligonucleotides (see Fig. 1B) were included in coupled TNT transcription/translation reactions programmed with pT7βR(1b/408)P. Percents inhibition were determined by direct comparison of antisense oligonucleotide-containing samples to control reaction samples without antisense addition. Open bars represent FLUC (IRES) inhibition; shaded bars represent RLUC inhibition. (B) HCVm260-279 antisense oligonucleotides were added to TNT transcription-translation reactions at assay concentrations ranging from 25 to 200 nM. Triangles represent pT7βR(1b/408)P; squares represent pT7βR(EMCV)P. Open figures illustrate FLUC (IRES) inhibition patterns;shaded figures represent RLUC inhibition of each respective assay plasmid.

FIG. 3

Mutational analysis of the HCV IRES IIId apical loop. (A) Specific IIId apical loop mutations and secondary structures. Nucleotide substitutions are denoted by lowercase letters, with wild-type (WT) IIId sequences and structures listed at the top for comparison. Abbreviations: LD, loop deleted; LC, loop conserved; SD, sequence deleted; SM, semimaintained; SC, sequence complementary; SV, sequence varied 3′ to 5′. (B) Wild-type bicistronic plasmid DNA or plasmid DNA samples containing mutations were used to program in vitro-coupled transcription-translation RRL reactions or were transfected into BT7-H cells. Relative IRES efficiencies were determined for each mutation by direct comparison of FLUC/RLUC ratios to wild-type values (arbitrarily 100%). Open bars represent in vitro RRL mean IRES efficiencies; shaded bars represent the same from BT7-H cell-based transfections. Duplicate samples were analyzed in three separate experiments; error bars represent standard errors of the mean.

FIG. 4

Mapping of IIId nucleotides essential for IRES translation. (A and C) Specific IIId apical loop mutations and secondary structures. Nucleotide substitutions are denoted by lowercase letters, with wild-type (WT) IIId sequences and structures listed at the top for comparison. (B and D) Relative IRES translational efficiencies of point or multiple G mutations, respectively, were determined as described in the legend to Fig. 3. Open bars, in vitro RRL mean IRES efficiencies; shaded bars, the same from BT7-H cell-based transfections.

FIG. 5

IIId stem-loop sequence alignment of HCV genotypes 1 through 6, related Flavivirus GBV-B, and pestiviruses BVDV and CSFV. Nucleotide numbers in viral genomes are listed along with genotype and accession number (GenBank). HCV IRES genotype 1b used in this study is underlined at the top, with the conserved GGG triplet identified by asterisks. Listed below genotype 1b are the sequence alignments. Identical sequences are represented by dots, and nucleotide variations are indicated by lowercase letters, while dashes denote gaps in alignments.

FIG. 6

Chimeric substitutions and effects on relative IRES translation. (A) Specific IIId apical loop mutations and secondary structures. Nucleotide substitutions are denoted by lowercase letters, with wild-type (WT) IIId sequences and structures listed at the top for comparison. (B) Relative IRES translational efficiencies of chimeric IRESs were determined as described in the legend to Fig. 3. Open bars represent in vitro RRL mean IRES efficiencies; shaded bars represent the same from BT7-H cell-based transfections.

FIG. 7

RNase T₁ probing of wild-type HCV IRES and GBV-B/HCV chimeric RNA in the absence (−) and presence (+) of 2.5 mM MgCl₂. Addition of MgCl₂ resulted in cleavage patterns similar to wild-type HCV for GBV-B1 and GBV-B3 (lanes 5, 9, and 17) but different for GBV-B2 (lane 13). The location of the IIId apical loop is indicated by an arrow, and the asterisks indicate chimeric loops. A bracket indicates the region of GBV-B2 that was most different from others. This region corresponded to loop IIId and upstream regions extending to G243. Lanes 1, 2, and 3 contain an RNase U₂ sequencing ladder (denaturing conditions), an RNase T₁ sequencing ladder (denaturing conditions), and a hydrolysis ladder, respectively. These samples were not all run on the same gel; therefore, the run times differ slightly.

FIG. 8

Summary figure compiling IIId apical mutations and chimeras with their relative IRES translational efficiencies compared to wild-type HCV IRES. Annotations are the same as those listed in the legend to Fig. 5, except that apical loop sequences are listed in bold type. Percent IRES activity is shown on the right with results listed as in vitro- and cell-based results. Single values denote the same result in both in vitro- and cell-based assays.