The RNA genome of the hepatitis C virus (HCV) contains an internal ribosome entry site (IRES), which binds to the hostcell 40S ribosomal subunit and initiates protein translation in the absence of most initiation factors. [1,2] Recruitment of the ribosomal subunit to the HCV RNA is driven by the high affinity of the IRES-40S interaction.[3] The IRES sequence adopts a highly ordered secondary structure (Figure 1 a). [4] The three-dimensional architecture of the IRES is dominated by independently folding RNA domains.[5] Structures of IRES-40S complexes have been studied by cryo-electron microscopy (cryo-EM) revealing the overall shape of the RNA. [6,7] Higher-resolution structures of individual subdomains, including II [8] and IIIa-e, [9][10][11][12] have been determined by crystallography and NMR spectroscopy. Cryo-EM studies revealed that the domain II, which plays an important structural role in HCV translation, [13] adopts an L-shaped conformation that directs the apical hairpin loop IIb to overlap with the ribosomal E site in the proximity of the P site.[6] Binding of domain II induces a conformational change in the 40S head [7] and closes the messenger RNA (mRNA) binding cleft.[6] NMR studies suggested that subdomain IIa might be a flexible hinge whose bent state is stabilized by binding of divalent metal ions. [8] To investigate the molecular architecture of the IRES domain II kink and its metal-ion-dependent stabilization, we have used X-ray crystallography and structure-guided incorporation of fluorescent labels. For X-ray crystal structure determination (see Figure S4 and Table S1 in the Supporting Information), an oligonucleotide (IIa-1) was used that contained residues 49-69 and 100-115 of the HCV IRES (Figure 1). The overall structure of the RNA revealed a bent architecture for the IIa subdomain. The two stems that are flanking the internal bulge are arranged at a right angle (Figure 1 b). [14] The upper stem (residues 58-69/100-110) forms a continuous helix that contains both standard and noncanonical base pairs that adopt cis-Watson-Crick geometries. U106 is looped out from the stem to allow continuous stacking of the flanking base pairs. The arrangement of residues of the internal bulge introduces a right-angled bend in the RNA Figure 1. a) Secondary structure of the HCV 5'-nontranslated region that contains the IRES element and the subdomain IIa-1 construct used for crystallization and fluorescence labeling. The position of the IIa-1 construct within the IRES is indicated by a box. The nucleotides shown in red were changed in the construct from the HCV (genotype 1b) sequence to improve RNA stability. Non-Watson-Crick base pairs are indicated by "o" symbols. The numbering scheme for the construct has been adopted from the full HCV RNA. b) The three-dimensional structure of the IIa-1 RNA. 2007, 119, 230 -233 between the base pairs G52-C111 and C58-G110 that is stabilized by a combination of base stacking, hydrogen bonding, and metal-ion participation (see Figure 2 and Figure S5 in the Supporting ...