Hepatitis C Virus Translation Regulation

Figures

• 

  Figure 1. cis-Elements in the hepatitis C virus (HCV) RNA genome that are involved in translation regulation. The HCV plus strand RNA genome. The internal ribosome entry site (IRES) in the HCV 5′-untranslated region (5′UTR), the entire 3′UTR and the cis-acting replication element (CRE) in the NS5B coding region are involved in translation regulation [22,23,24]. Those regions of the 5′UTR, 3′UTR, and CRE that bind to the ribosomal 40S subunit are underlayed in light yellow. Stem-loops (SLs) in the 5′UTR are numbered by roman numerals. The region of the IRES is surrounded by a dotted line. The IRES includes SLs II–IV of the 5′UTR but spans into the core protein coding region. The canonical AUG start codon in SL IV of the 5′UTR (black circle) gives rise to translation of the polyprotein which is cleaved to yield structural proteins and non-structural (NS-) proteins, including the viral replicase NS5B. The 3′UTR contains the variable region, a poly(U/C) tract (U/C), and the so-called 3′X region including SLs 1, 2, and 3. The stem-loop 5BSL3.2 in the 3′-region of the NS5B coding region is the CRE, flanked by upstream stem-loop 5BSL3.1 and downstream 5BSL3.3. The polyprotein stop codon is…

• 

  Figure 2. Sequence and RNA secondary structure of the HCV IRES. (A) The 5′UTR of HCV and its canonical structure [63,109,110,111,112,113,114]. The sequence shown is from genotype 2a (J6 isolate), with length variations among HCV isolates indicated by short horizontal dashes at nucleotide positions. Nucleotide numbering is according to the MAFFT alignment in the supplement of [63] (http://www.rna.uni-jena.de/supplements/hcv/). For comparison, the first nucleotide of the core coding sequence (AUG, in bold) in the SL IV is nucleotide No. 342 in genotype 1b (Con1) and No. 341 in genotype 2a (J6 and JFH1 isolates). IRES SL domains are indicated by roman numerals. microRNA-122 (miR-122, red) binding is indicated, miR-122 target sequences are boxed. Pseudoknot (PK) base pairing and the interaction between SL II and SL IV is indicated; (B) Secondary structure model of the 5′UTR with alternative folding of SL IIalt. The structure is largely according to [72,77,115], with minor modifications according to our RNAalifold outputs using several genotypes (not shown).

• 

  Figure 3. Conformational dynamics of HCV 5′UTR structure. The conserved canonical 5′UTR structure [63,109,110,111,112,113,114] is shown in the middle (structure A). 5′UTR sequences which have a strong intrinsic tendency to fold into a distinct secondary and tertiary structure are depicted in bold. The constitutive double pseudoknot (PK) interactions involving SL IIIe (PK2) and IIIf (PK1) and the possible interaction between SL II and SL IV are indicated. The 5′UTR structure B [72,77,115] has the region between SLs I and III refolded and forms an alternative version of SL II (SL IIalt). This SL IIalt can form in the absence of miR-122, then hiding both miR-122 binding sites. Binding of miR-122 promotes refolding of structure B to structure A [72,115,122]. The 5′UTR structure C can form by fold-back of nucleotides 428–442 in the base of SL VI in the core protein coding region to hybridize with sequences downstream of SL I. In this state, the translation activity of the IRES is decreased [130,131,132,133,134]. This hybridization also interferes with binding of miR-122 to the 5′UTR. In turn, binding of miR-122 induces refolding of structure C to structure A [135]. Structure D is a yet…

• 

  Figure 4. Binding of the HCV IRES to the ribosome and to eIF3. (A) The small ribosomal 40S subunit (yellow) with bound eIF3 (red). eIF3 makes multiple contacts to the 40S subunit using several subunits, including eIF3a and eIF3c; (B) The IRES of classical swine fever virus (CSFV) (pink) without IRES domain II, binding to the 40S subunit (yellow) and to eIF3 (red). The IRES has displaced eIF3 completely from its binding to the 40S subunit (compare panel A) but keeps it connected to the 40S subunit only indirectly by contacts between IRES SL IIIabc and eIF3. Figures A and B were reprinted from [150] (Figure 2F) and slightly modified. Reprinted with permission from Elsevier (licence no. 4761840166773); (C) The HCV IRES (pink) binding to the 40S subunit (yellow) in the complete 80S ribosome (60S subunit in blue). The IRES SL IIIdef/PK is in close contact with the 40S subunit, the SLs II and IV are positioned in the mRNA entry channel, and the SL IIIb is pointing to the solvent side for binding eIF3. The IRES is not touching the 60S subunit [121,138]. Figure C was modified from the left panel of Figure 1C in [121]. Reproduced with permission from EMBO; (D) Schematic illustration of the HCV IRES…

• 

  Figure 5. The steps involved in translation initiation at the HCV IRES. (A) The canonical way of translation initiation with the HCV IRES under conditions of sufficient eIF2 activity. In vivo, the IRES most likely binds to 40S-eIF3 post-termination complexes (top right), while in vitro-studies also suggest that binding of 40S and eIF3 can occur subsequently (top left). Then, the ternary 40S-IRES-eIF3 complex acquires eIF2 charged with tRNAiMet and GTP, as well as eIF1A and eIF5. After locating the HCV AUG start codon, eIF5 catalyzes release of decharged eIF2-GDP from the ribosome. eIF5B then causes subunit joining, and eIF5B-GDP, eIF1A, and eIF3 leave the complex which is, then, ready for the first translation elongation step. “Hitchhiking” of the IRES on translating 80S ribosomes is not shown (see main text); (B and C) Alternative translation initiation pathways for the HCV IRES under stress conditions leading to eIF2α phosphorylation, i.e., under limited eIF2 availability. (B) Binary IRES-40S complexes, which can or cannot also bind eIF3, bind either eIF2A (left), eIF5B (right) or both eIF2A and eIF5B in combination (middle). In the cases when eIF2A is present, it delivers the tRNAiMet.…

• 

  Figure 6. IRES trans-acting factors (ITAFs) that modulate HCV IRES activity. Important functional domains are shown in dark grey. In most cases, these domains are RNA-binding domains, similar to the RNA-recognition motif domain (RRM) [194] or the K-homology (KH) domains of hnRNP proteins [195].

• 

  Figure 7. Interactions between HCV 5′- and 3′UTR. The HCV RNA is shown as a black line, with most of the coding region shortened as a dotted line. RNA–RNA long-range interactions (LRIs) are shown with green lines. Interactions mediated by proteins, by the 40S subunit, and by eIF3 are shown with blue lines. The proteins are shown as transparent bodies with blue lines. Other indications are as in previous figures. For details, please see main text.

• 

  Figure 8. Translation of the HCV alternative reading frame (ARF) or core+1 frame. (A) Overview over starts and stops that result in expression of a variety of different ARF of core+1 protein, among HCV isolates. The HCV 5′UTR and the downstream core region is shown essentially as in Figure 1. The start AUG for the canonical core protein is shown in SL IV as black filled circle. The start codons of core+1 protein are shown as grey filled circles at the core+1 frame for positions 26, 58, 85, and 87. The canonical core protein and the various core+1 ARF products are shown as boxes below. The major products produced in genotype 2a are shown in grey, other products produced by initiation at codon 58 or by termination at other more downstream stops in other genotypes are underlayed in white. Start codon usage is depicted; (B) Amino acid sequences of representative HCV genotypes and subtypes (selected from [63]); genotypes and NCBI nucleotide database accession numbers are given on the left, as well as abbreviations of some well-known isolates. The AA sequence starts with codon 26 of the core+1 frame, when the nucleotide No. 5 of the canonical core frame, AUGAG, is nucleotide No. 1 of the core+1…