PMC

NS4B binds specifically to the 3′ terminus of the HCV negative-strand RNA. (a) Four HCV probes were designed. 5′ UTR pos and 3′ UTR pos corresponded to the 5′ UTR and 3′ UTR sequences of the positive viral strand, respectively, and 5′ negative terminus and 3′ negative terminus corresponded to the 5′ and 3′ terminal regions of the negative strand, respectively. The position of these sequences with respect to the HCV open reading frame (ORF) is shown. (b) Fractional binding of NS4B to equimolar concentrations of the four HCV probes and to a non-HCV RNA (delta virus RNA) probe. The 3′ terminus of the negative genome strand is favored by > 5×. Each data point represents the mean of 10–20 replicates, and the bars represent the standard error.

Identification of RNA binding domains within NS4B. (a) Arginine-rich sequences known to confer RNA binding found in viral and bacteriophage proteins^,,. Elements in the terminal loop of NS4B that conform with arginine rich-like motifs are shown at the bottom. Proteins listed are HIV Rev and Tat; λ, Φ21 and P22 bacteriophage antiterminator N proteins; triple gene block protein 1 (TGBp1) of the bamboo mosaic virus (BaMV) and poliovirus 2C. (b) Positively charged amino acids (highlighted) within the primary sequence of NS4B (genotype 1a). (c) Schematic diagram indicating predicted transmembrane and intracellular domains of NS4B. Conserved positively charged amino acids in the C-terminal segment of NS4B are shown in red. (d) Arginine residues mediate RNA binding by NS4B. Binding of wild-type NS4B-GFP, RRa and RRb NS4B-GFP mutants and NS5A(AH)-GFP to the 3′ terminus of the negative viral RNA strand (1.5 nM) was determined by microfluidic affinity analysis. Bars represent s.d.

Small-molecule screen reveals that clemizole hydrochloride inhibits RNA binding by NS4B and HCV RNA replication in cell culture. (a) The first screen represented a low-stringency measurement of inhibition of 1,280 compounds where the latter were categorized as having high (green), ambiguous (blue) or no (red) inhibition. Based on the initial screen, 214 compounds were then measured again with higher stringency and with a greater number of replicates and the best 18 inhibitors were tested for their ability to inhibit HCV replication via a cellular assay. (b) In vitro inhibition of NS4B-RNA binding by the top 18 small molecules. (c) HCV luciferase reporter–linked cellular assay showing that clemizole inhibits HCV replication (left axis, blue ◆) with no measurable toxicity to the cell as measured by Alamar Blue (right axis, red □). (d) In vitro NS4B-RNA binding: inhibition curve of clemizole. Each data point represents the mean of 10–20 replicates and the bars represent the standard error.

Protein-RNA interactions measured on microfluidic platform. (a) Target RNA sequences used to study binding of HuD to RNA and comparison of K_d values measured using microfluidic affinity analysis to values previously measured by gel shift assay^,. ND, not determined. (b) Binding curve of HuD to increasing concentration of AU3 (●) and AU3 mutant (Δ) RNA, as determined in the microfluidic affinity assay. Normalized mean values for 10–20 replicates measured in two independent experiments are shown for each graph. Error bars represent s.d. (c) Fluorescent images from microfluidic chip. Left: NS4B-GFP and NS5A(AH)-GFP were anchored to the microfluidic device surface via its interaction with anti-GFP Middle: an RNA probe corresponding to the 3′ terminal region of the negative viral strand was labeled with Cy5 and incubated with the proteins on the device. Cy5 signal representing bound RNA is shown following a brief wash. Right: an overlay of the Cy5 signal and GFP signal representing bound RNA to protein ratio. (d) In vitro binding curve of NS4B to serial dilutions of the RNA probe. Each data point represents the mean of 10–20 replicates, and the bars represent the standard error.

Clemizol-resistant mutant. Replicon cells were passaged in the presence of clemizole and individual colonies were isolated, propagated and the HCV genomes harbored within were subjected to sequence analysis, (a) Schematic diagram indicating predicted transmembrane and intracellular segments of NS4B. Conserved positively charged amino acids are shown in red. The clemizole-resistant mutation, W55R, is shown in green, (b) HCV replication in Huh7.5 cells electroporated with 50 µg of whole-cell RNA extracted from cells harboring either wild type or the W55R mutant clone, followed by growth in the absence (white bars) or presence (gray bars) of 10 µM clemizole. Results represent relative numbers of colonies obtained compared to each corresponding untreated control, (c) HCV replication assays initiated by electroporation of in vitro transcribed luciferase reporter-linked wild-type or W55R mutant HCV RNA genomes performed in the absence (white bars) or presence (gray bars) of 10 µM clemizole. Results represent replication level of each genome relative to its untreated level, (d) HCV RNA binding of wild-type NS4B and the W55R NS4B mutant as measured in vitro by microfluidics in the presence of 10 nM clemizole (gray bars) and its absence (white bars), (e) In vitro binding curves of W55R NS4B mutant (solid line, O) and wild-type NS4B (broken line, ▲) to serial dilutions of the RNA probe. Each data point represents the mean of 10–20 replicates, and the bars represent the standard error.